{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 28.06361085126286,
  "eval_steps": 500,
  "global_step": 30000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "action_query_batch_std": 0.44519842267036436,
      "action_query_norm": 45.25465393066406,
      "action_relation_loss": 0.06319189928472042,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.009354536950420954,
      "flow_matching_loss": 0.919458419084549,
      "grad_norm": 7.214541435241699,
      "h_stock_valid_norm": 80.02765197753907,
      "invalid_tail_samples": 24.4,
      "learning_rate": 6.000000000000001e-07,
      "loss": 0.9827,
      "positive_teacher_pairs": 26709.2,
      "query_h_norm_ratio": 0.5654894173145294,
      "relation_loss": 0.06319189928472042,
      "selected_pairs": 26709.2,
      "step": 10,
      "student_distance_mean": 28.515333557128905,
      "teacher_distance_mean": 0.8458072602748871,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.06319189928472042,
      "weighted_relation_loss": 0.06319189928472042
    },
    {
      "action_query_batch_std": 0.4452923595905304,
      "action_query_norm": 45.2546630859375,
      "action_relation_loss": 0.06521583907306194,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.018709073900841908,
      "flow_matching_loss": 0.9060794293880463,
      "grad_norm": 6.968599319458008,
      "h_stock_valid_norm": 79.86261520385742,
      "invalid_tail_samples": 22.4,
      "learning_rate": 1.2666666666666667e-06,
      "loss": 0.9713,
      "positive_teacher_pairs": 27177.8,
      "query_h_norm_ratio": 0.5666582882404327,
      "relation_loss": 0.06521583907306194,
      "selected_pairs": 27177.8,
      "step": 20,
      "student_distance_mean": 28.5338716506958,
      "teacher_distance_mean": 0.8513376951217652,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.06521583907306194,
      "weighted_relation_loss": 0.06521583907306194
    },
    {
      "action_query_batch_std": 0.44308011829853056,
      "action_query_norm": 45.25467643737793,
      "action_relation_loss": 0.06156411655247211,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.02806361085126286,
      "flow_matching_loss": 0.7607109785079956,
      "grad_norm": 4.149909019470215,
      "h_stock_valid_norm": 79.76971588134765,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.9333333333333336e-06,
      "loss": 0.8223,
      "positive_teacher_pairs": 26755.1,
      "query_h_norm_ratio": 0.5673177897930145,
      "relation_loss": 0.06156411655247211,
      "selected_pairs": 26755.1,
      "step": 30,
      "student_distance_mean": 28.366924476623534,
      "teacher_distance_mean": 0.8455843806266785,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.06156411655247211,
      "weighted_relation_loss": 0.06156411655247211
    },
    {
      "action_query_batch_std": 0.4399962991476059,
      "action_query_norm": 45.25469551086426,
      "action_relation_loss": 0.059114628657698634,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.037418147801683815,
      "flow_matching_loss": 0.5642935127019882,
      "grad_norm": 1.3948876857757568,
      "h_stock_valid_norm": 79.66342010498047,
      "invalid_tail_samples": 23.1,
      "learning_rate": 2.6e-06,
      "loss": 0.6234,
      "positive_teacher_pairs": 27012.3,
      "query_h_norm_ratio": 0.5680768370628357,
      "relation_loss": 0.059114628657698634,
      "selected_pairs": 27012.3,
      "step": 40,
      "student_distance_mean": 28.186686325073243,
      "teacher_distance_mean": 0.8578713417053223,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.059114628657698634,
      "weighted_relation_loss": 0.059114628657698634
    },
    {
      "action_query_batch_std": 0.4309561252593994,
      "action_query_norm": 45.25471076965332,
      "action_relation_loss": 0.057878477871418,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.04677268475210477,
      "flow_matching_loss": 0.431062114238739,
      "grad_norm": 1.2313226461410522,
      "h_stock_valid_norm": 79.75285568237305,
      "invalid_tail_samples": 25.8,
      "learning_rate": 3.2666666666666666e-06,
      "loss": 0.4889,
      "positive_teacher_pairs": 26396.6,
      "query_h_norm_ratio": 0.5674397826194764,
      "relation_loss": 0.057878477871418,
      "selected_pairs": 26396.6,
      "step": 50,
      "student_distance_mean": 27.575566673278807,
      "teacher_distance_mean": 0.8490286767482758,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.057878477871418,
      "weighted_relation_loss": 0.057878477871418
    },
    {
      "action_query_batch_std": 0.41570029556751253,
      "action_query_norm": 45.25473403930664,
      "action_relation_loss": 0.05633627809584141,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.05612722170252572,
      "flow_matching_loss": 0.3551485061645508,
      "grad_norm": 1.0412205457687378,
      "h_stock_valid_norm": 79.84184188842774,
      "invalid_tail_samples": 21.6,
      "learning_rate": 3.9333333333333335e-06,
      "loss": 0.4115,
      "positive_teacher_pairs": 27364.8,
      "query_h_norm_ratio": 0.5668071866035461,
      "relation_loss": 0.05633627809584141,
      "selected_pairs": 27364.8,
      "step": 60,
      "student_distance_mean": 26.568267250061034,
      "teacher_distance_mean": 0.8570368528366089,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.05633627809584141,
      "weighted_relation_loss": 0.05633627809584141
    },
    {
      "action_query_batch_std": 0.406466481089592,
      "action_query_norm": 45.25474243164062,
      "action_relation_loss": 0.053385843709111216,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.06548175865294668,
      "flow_matching_loss": 0.3052156984806061,
      "grad_norm": 0.9774118661880493,
      "h_stock_valid_norm": 80.44348907470703,
      "invalid_tail_samples": 25.5,
      "learning_rate": 4.6e-06,
      "loss": 0.3586,
      "positive_teacher_pairs": 26454.3,
      "query_h_norm_ratio": 0.5625751495361329,
      "relation_loss": 0.053385843709111216,
      "selected_pairs": 26454.3,
      "step": 70,
      "student_distance_mean": 25.913517951965332,
      "teacher_distance_mean": 0.8526943862438202,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.053385843709111216,
      "weighted_relation_loss": 0.053385843709111216
    },
    {
      "action_query_batch_std": 0.39787589609622953,
      "action_query_norm": 45.25475730895996,
      "action_relation_loss": 0.05208288617432118,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.07483629560336763,
      "flow_matching_loss": 0.27009828984737394,
      "grad_norm": 0.704868733882904,
      "h_stock_valid_norm": 81.80571060180664,
      "invalid_tail_samples": 24.8,
      "learning_rate": 5.266666666666667e-06,
      "loss": 0.3222,
      "positive_teacher_pairs": 26614.2,
      "query_h_norm_ratio": 0.5532117784023285,
      "relation_loss": 0.05208288617432118,
      "selected_pairs": 26614.2,
      "step": 80,
      "student_distance_mean": 25.345639801025392,
      "teacher_distance_mean": 0.8562870621681213,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.05208288617432118,
      "weighted_relation_loss": 0.05208288617432118
    },
    {
      "action_query_batch_std": 0.39124230444431307,
      "action_query_norm": 45.25475921630859,
      "action_relation_loss": 0.04759731367230415,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.08419083255378859,
      "flow_matching_loss": 0.2521683186292648,
      "grad_norm": 0.6291298270225525,
      "h_stock_valid_norm": 83.08995132446289,
      "invalid_tail_samples": 23.6,
      "learning_rate": 5.933333333333334e-06,
      "loss": 0.2998,
      "positive_teacher_pairs": 26897.3,
      "query_h_norm_ratio": 0.54466552734375,
      "relation_loss": 0.04759731367230415,
      "selected_pairs": 26897.3,
      "step": 90,
      "student_distance_mean": 24.91824722290039,
      "teacher_distance_mean": 0.8514817237854004,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.04759731367230415,
      "weighted_relation_loss": 0.04759731367230415
    },
    {
      "action_query_batch_std": 0.3954827517271042,
      "action_query_norm": 45.2547420501709,
      "action_relation_loss": 0.04510456323623657,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.09354536950420954,
      "flow_matching_loss": 0.22030310779809953,
      "grad_norm": 0.46649086475372314,
      "h_stock_valid_norm": 84.18203430175781,
      "invalid_tail_samples": 23.8,
      "learning_rate": 6.6e-06,
      "loss": 0.2654,
      "positive_teacher_pairs": 26853.0,
      "query_h_norm_ratio": 0.5375893414020538,
      "relation_loss": 0.04510456323623657,
      "selected_pairs": 26853.0,
      "step": 100,
      "student_distance_mean": 25.174398040771486,
      "teacher_distance_mean": 0.8571404814720154,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.04510456323623657,
      "weighted_relation_loss": 0.04510456323623657
    },
    {
      "action_query_batch_std": 0.40022508800029755,
      "action_query_norm": 45.25470809936523,
      "action_relation_loss": 0.04333480000495911,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.1028999064546305,
      "flow_matching_loss": 0.2042098358273506,
      "grad_norm": 0.45488178730010986,
      "h_stock_valid_norm": 85.22114944458008,
      "invalid_tail_samples": 26.6,
      "learning_rate": 7.266666666666668e-06,
      "loss": 0.2475,
      "positive_teacher_pairs": 26207.9,
      "query_h_norm_ratio": 0.5310316920280457,
      "relation_loss": 0.04333480000495911,
      "selected_pairs": 26207.9,
      "step": 110,
      "student_distance_mean": 25.5108736038208,
      "teacher_distance_mean": 0.8551790535449981,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.04333480000495911,
      "weighted_relation_loss": 0.04333480000495911
    },
    {
      "action_query_batch_std": 0.404365086555481,
      "action_query_norm": 45.25468559265137,
      "action_relation_loss": 0.040216144919395444,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.11225444340505145,
      "flow_matching_loss": 0.1934616357088089,
      "grad_norm": 0.6344209909439087,
      "h_stock_valid_norm": 86.17413024902343,
      "invalid_tail_samples": 24.7,
      "learning_rate": 7.933333333333334e-06,
      "loss": 0.2337,
      "positive_teacher_pairs": 26649.7,
      "query_h_norm_ratio": 0.525160139799118,
      "relation_loss": 0.040216144919395444,
      "selected_pairs": 26649.7,
      "step": 120,
      "student_distance_mean": 25.759596061706542,
      "teacher_distance_mean": 0.8527304410934449,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.040216144919395444,
      "weighted_relation_loss": 0.040216144919395444
    },
    {
      "action_query_batch_std": 0.4051499247550964,
      "action_query_norm": 45.25470314025879,
      "action_relation_loss": 0.03966134190559387,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.1216089803554724,
      "flow_matching_loss": 0.18611549586057663,
      "grad_norm": 0.7581837773323059,
      "h_stock_valid_norm": 86.99815292358399,
      "invalid_tail_samples": 21.7,
      "learning_rate": 8.599999999999999e-06,
      "loss": 0.2258,
      "positive_teacher_pairs": 27342.7,
      "query_h_norm_ratio": 0.5201857447624206,
      "relation_loss": 0.03966134190559387,
      "selected_pairs": 27342.7,
      "step": 130,
      "student_distance_mean": 25.82041015625,
      "teacher_distance_mean": 0.8566129565238952,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.03966134190559387,
      "weighted_relation_loss": 0.03966134190559387
    },
    {
      "action_query_batch_std": 0.3961487889289856,
      "action_query_norm": 45.25480079650879,
      "action_relation_loss": 0.03722836971282959,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.13096351730589337,
      "flow_matching_loss": 0.17852823585271835,
      "grad_norm": 0.6783291697502136,
      "h_stock_valid_norm": 88.12196731567383,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.266666666666667e-06,
      "loss": 0.2158,
      "positive_teacher_pairs": 27174.7,
      "query_h_norm_ratio": 0.5135553419589997,
      "relation_loss": 0.03722836971282959,
      "selected_pairs": 27174.7,
      "step": 140,
      "student_distance_mean": 25.237031745910645,
      "teacher_distance_mean": 0.8592345356941223,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.03722836971282959,
      "weighted_relation_loss": 0.03722836971282959
    },
    {
      "action_query_batch_std": 0.38906853199005126,
      "action_query_norm": 45.254964447021486,
      "action_relation_loss": 0.03415284529328346,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.1403180542563143,
      "flow_matching_loss": 0.16998847126960753,
      "grad_norm": 0.5640113949775696,
      "h_stock_valid_norm": 89.1622184753418,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.933333333333334e-06,
      "loss": 0.2041,
      "positive_teacher_pairs": 26826.3,
      "query_h_norm_ratio": 0.5075610697269439,
      "relation_loss": 0.03415284529328346,
      "selected_pairs": 26826.3,
      "step": 150,
      "student_distance_mean": 24.773453140258788,
      "teacher_distance_mean": 0.8583857834339141,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.03415284529328346,
      "weighted_relation_loss": 0.03415284529328346
    },
    {
      "action_query_batch_std": 0.38819784820079806,
      "action_query_norm": 45.25516700744629,
      "action_relation_loss": 0.03402448259294033,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.14967259120673526,
      "flow_matching_loss": 0.15546735674142836,
      "grad_norm": 0.405224472284317,
      "h_stock_valid_norm": 90.23874435424804,
      "invalid_tail_samples": 24.8,
      "learning_rate": 1.06e-05,
      "loss": 0.1895,
      "positive_teacher_pairs": 26622.1,
      "query_h_norm_ratio": 0.5015075296163559,
      "relation_loss": 0.03402448259294033,
      "selected_pairs": 26622.1,
      "step": 160,
      "student_distance_mean": 24.6736047744751,
      "teacher_distance_mean": 0.8498655676841735,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.03402448259294033,
      "weighted_relation_loss": 0.03402448259294033
    },
    {
      "action_query_batch_std": 0.386635485291481,
      "action_query_norm": 45.255333709716794,
      "action_relation_loss": 0.03265544716268778,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.15902712815715622,
      "flow_matching_loss": 0.1487676188349724,
      "grad_norm": 0.45525887608528137,
      "h_stock_valid_norm": 91.3005256652832,
      "invalid_tail_samples": 22.0,
      "learning_rate": 1.1266666666666667e-05,
      "loss": 0.1814,
      "positive_teacher_pairs": 27272.6,
      "query_h_norm_ratio": 0.4956775724887848,
      "relation_loss": 0.03265544716268778,
      "selected_pairs": 27272.6,
      "step": 170,
      "student_distance_mean": 24.630986213684082,
      "teacher_distance_mean": 0.8486714065074921,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.03265544716268778,
      "weighted_relation_loss": 0.03265544716268778
    },
    {
      "action_query_batch_std": 0.3882857024669647,
      "action_query_norm": 45.255475997924805,
      "action_relation_loss": 0.03188955653458834,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.16838166510757718,
      "flow_matching_loss": 0.1413273334503174,
      "grad_norm": 0.5308682322502136,
      "h_stock_valid_norm": 91.82629623413087,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.1933333333333333e-05,
      "loss": 0.1732,
      "positive_teacher_pairs": 27079.9,
      "query_h_norm_ratio": 0.49283901453018186,
      "relation_loss": 0.03188955653458834,
      "selected_pairs": 27079.9,
      "step": 180,
      "student_distance_mean": 24.621802139282227,
      "teacher_distance_mean": 0.8509422183036804,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.03188955653458834,
      "weighted_relation_loss": 0.03188955653458834
    },
    {
      "action_query_batch_std": 0.3891027867794037,
      "action_query_norm": 45.25562973022461,
      "action_relation_loss": 0.0320635724812746,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.17773620205799812,
      "flow_matching_loss": 0.14214842021465302,
      "grad_norm": 0.5838639140129089,
      "h_stock_valid_norm": 92.18766784667969,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.2600000000000001e-05,
      "loss": 0.1742,
      "positive_teacher_pairs": 26875.0,
      "query_h_norm_ratio": 0.49091089367866514,
      "relation_loss": 0.0320635724812746,
      "selected_pairs": 26875.0,
      "step": 190,
      "student_distance_mean": 24.7142765045166,
      "teacher_distance_mean": 0.8458907246589661,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0320635724812746,
      "weighted_relation_loss": 0.0320635724812746
    },
    {
      "action_query_batch_std": 0.39958467781543733,
      "action_query_norm": 45.255741119384766,
      "action_relation_loss": 0.031121252477169035,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.18709073900841908,
      "flow_matching_loss": 0.1388360694050789,
      "grad_norm": 0.3691250681877136,
      "h_stock_valid_norm": 92.53959274291992,
      "invalid_tail_samples": 23.5,
      "learning_rate": 1.3266666666666666e-05,
      "loss": 0.17,
      "positive_teacher_pairs": 26918.3,
      "query_h_norm_ratio": 0.4890464872121811,
      "relation_loss": 0.031121252477169035,
      "selected_pairs": 26918.3,
      "step": 200,
      "student_distance_mean": 25.347447776794432,
      "teacher_distance_mean": 0.8618796646595002,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.031121252477169035,
      "weighted_relation_loss": 0.031121252477169035
    },
    {
      "action_query_batch_std": 0.4119757115840912,
      "action_query_norm": 45.25580825805664,
      "action_relation_loss": 0.030871157348155976,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.19644527595884004,
      "flow_matching_loss": 0.13044705614447594,
      "grad_norm": 0.6913148164749146,
      "h_stock_valid_norm": 92.7876693725586,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.3933333333333334e-05,
      "loss": 0.1613,
      "positive_teacher_pairs": 27081.4,
      "query_h_norm_ratio": 0.48773671090602877,
      "relation_loss": 0.030871157348155976,
      "selected_pairs": 27081.4,
      "step": 210,
      "student_distance_mean": 26.124947357177735,
      "teacher_distance_mean": 0.8458437085151672,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.030871157348155976,
      "weighted_relation_loss": 0.030871157348155976
    },
    {
      "action_query_batch_std": 0.42486476600170137,
      "action_query_norm": 45.2557876586914,
      "action_relation_loss": 0.029192883893847464,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.205799812909261,
      "flow_matching_loss": 0.1326353594660759,
      "grad_norm": 0.6899434328079224,
      "h_stock_valid_norm": 92.81553344726562,
      "invalid_tail_samples": 24.5,
      "learning_rate": 1.4599999999999999e-05,
      "loss": 0.1618,
      "positive_teacher_pairs": 26688.9,
      "query_h_norm_ratio": 0.48758949637413024,
      "relation_loss": 0.029192883893847464,
      "selected_pairs": 26688.9,
      "step": 220,
      "student_distance_mean": 26.980838966369628,
      "teacher_distance_mean": 0.848963075876236,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.029192883893847464,
      "weighted_relation_loss": 0.029192883893847464
    },
    {
      "action_query_batch_std": 0.42734754979610445,
      "action_query_norm": 45.255876541137695,
      "action_relation_loss": 0.029610772803425788,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.21515434985968196,
      "flow_matching_loss": 0.1280670315027237,
      "grad_norm": 0.5986344218254089,
      "h_stock_valid_norm": 93.15821990966796,
      "invalid_tail_samples": 23.4,
      "learning_rate": 1.5266666666666667e-05,
      "loss": 0.1577,
      "positive_teacher_pairs": 26947.2,
      "query_h_norm_ratio": 0.48579692244529726,
      "relation_loss": 0.029610772803425788,
      "selected_pairs": 26947.2,
      "step": 230,
      "student_distance_mean": 27.19253044128418,
      "teacher_distance_mean": 0.8595677196979523,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.029610772803425788,
      "weighted_relation_loss": 0.029610772803425788
    },
    {
      "action_query_batch_std": 0.4206335932016373,
      "action_query_norm": 45.25608787536621,
      "action_relation_loss": 0.02738343421369791,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.2245088868101029,
      "flow_matching_loss": 0.12580396085977555,
      "grad_norm": 0.49872103333473206,
      "h_stock_valid_norm": 93.58844528198242,
      "invalid_tail_samples": 24.6,
      "learning_rate": 1.5933333333333332e-05,
      "loss": 0.1532,
      "positive_teacher_pairs": 26659.7,
      "query_h_norm_ratio": 0.4835669606924057,
      "relation_loss": 0.02738343421369791,
      "selected_pairs": 26659.7,
      "step": 240,
      "student_distance_mean": 26.70017776489258,
      "teacher_distance_mean": 0.8667687594890594,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.02738343421369791,
      "weighted_relation_loss": 0.02738343421369791
    },
    {
      "action_query_batch_std": 0.42957811057567596,
      "action_query_norm": 45.25611419677735,
      "action_relation_loss": 0.027063868008553982,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.23386342376052385,
      "flow_matching_loss": 0.12488146424293518,
      "grad_norm": 0.4596802294254303,
      "h_stock_valid_norm": 93.7844955444336,
      "invalid_tail_samples": 23.5,
      "learning_rate": 1.66e-05,
      "loss": 0.1519,
      "positive_teacher_pairs": 26918.7,
      "query_h_norm_ratio": 0.4825552970170975,
      "relation_loss": 0.027063868008553982,
      "selected_pairs": 26918.7,
      "step": 250,
      "student_distance_mean": 27.216525840759278,
      "teacher_distance_mean": 0.8567311465740204,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.027063868008553982,
      "weighted_relation_loss": 0.027063868008553982
    },
    {
      "action_query_batch_std": 0.4380154013633728,
      "action_query_norm": 45.256008529663085,
      "action_relation_loss": 0.027322525903582574,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.2432179607109448,
      "flow_matching_loss": 0.12143545597791672,
      "grad_norm": 0.5063205361366272,
      "h_stock_valid_norm": 93.72392959594727,
      "invalid_tail_samples": 26.7,
      "learning_rate": 1.726666666666667e-05,
      "loss": 0.1488,
      "positive_teacher_pairs": 26196.7,
      "query_h_norm_ratio": 0.48286728858947753,
      "relation_loss": 0.027322525903582574,
      "selected_pairs": 26196.7,
      "step": 260,
      "student_distance_mean": 27.951446533203125,
      "teacher_distance_mean": 0.8574040949344635,
      "valid_action_samples": 229.3,
      "weighted_auxiliary_loss": 0.027322525903582574,
      "weighted_relation_loss": 0.027322525903582574
    },
    {
      "action_query_batch_std": 0.4416928321123123,
      "action_query_norm": 45.256103897094725,
      "action_relation_loss": 0.025983518362045287,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.25257249766136575,
      "flow_matching_loss": 0.11409894302487374,
      "grad_norm": 0.4868370592594147,
      "h_stock_valid_norm": 94.09074478149414,
      "invalid_tail_samples": 21.1,
      "learning_rate": 1.7933333333333337e-05,
      "loss": 0.1401,
      "positive_teacher_pairs": 27476.1,
      "query_h_norm_ratio": 0.48098589181900026,
      "relation_loss": 0.025983518362045287,
      "selected_pairs": 27476.1,
      "step": 270,
      "student_distance_mean": 28.070331001281737,
      "teacher_distance_mean": 0.8486423492431641,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.025983518362045287,
      "weighted_relation_loss": 0.025983518362045287
    },
    {
      "action_query_batch_std": 0.44729403853416444,
      "action_query_norm": 45.25631675720215,
      "action_relation_loss": 0.02679998893290758,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.26192703461178674,
      "flow_matching_loss": 0.1098124623298645,
      "grad_norm": 0.5493093132972717,
      "h_stock_valid_norm": 94.44245147705078,
      "invalid_tail_samples": 22.0,
      "learning_rate": 1.86e-05,
      "loss": 0.1366,
      "positive_teacher_pairs": 27263.9,
      "query_h_norm_ratio": 0.47919716238975524,
      "relation_loss": 0.02679998893290758,
      "selected_pairs": 27263.9,
      "step": 280,
      "student_distance_mean": 28.3196533203125,
      "teacher_distance_mean": 0.8574328243732452,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.02679998893290758,
      "weighted_relation_loss": 0.02679998893290758
    },
    {
      "action_query_batch_std": 0.45789146423339844,
      "action_query_norm": 45.256409072875975,
      "action_relation_loss": 0.027130278572440148,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.27128157156220767,
      "flow_matching_loss": 0.11467377990484237,
      "grad_norm": 0.6915762424468994,
      "h_stock_valid_norm": 94.66059112548828,
      "invalid_tail_samples": 27.7,
      "learning_rate": 1.926666666666667e-05,
      "loss": 0.1418,
      "positive_teacher_pairs": 25961.5,
      "query_h_norm_ratio": 0.4780924618244171,
      "relation_loss": 0.027130278572440148,
      "selected_pairs": 25961.5,
      "step": 290,
      "student_distance_mean": 28.938625717163085,
      "teacher_distance_mean": 0.8567277014255523,
      "valid_action_samples": 228.3,
      "weighted_auxiliary_loss": 0.027130278572440148,
      "weighted_relation_loss": 0.027130278572440148
    },
    {
      "action_query_batch_std": 0.4722477287054062,
      "action_query_norm": 45.25641632080078,
      "action_relation_loss": 0.02622472308576107,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.2806361085126286,
      "flow_matching_loss": 0.11312596574425697,
      "grad_norm": 0.5369192957878113,
      "h_stock_valid_norm": 95.01139984130859,
      "invalid_tail_samples": 24.8,
      "learning_rate": 1.9933333333333334e-05,
      "loss": 0.1394,
      "positive_teacher_pairs": 26620.2,
      "query_h_norm_ratio": 0.4763284206390381,
      "relation_loss": 0.02622472308576107,
      "selected_pairs": 26620.2,
      "step": 300,
      "student_distance_mean": 29.87894344329834,
      "teacher_distance_mean": 0.8537890136241912,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.02622472308576107,
      "weighted_relation_loss": 0.02622472308576107
    },
    {
      "action_query_batch_std": 0.4697927743196487,
      "action_query_norm": 45.256430053710936,
      "action_relation_loss": 0.024994135648012162,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.2899906454630496,
      "flow_matching_loss": 0.11109494864940643,
      "grad_norm": 0.5184224247932434,
      "h_stock_valid_norm": 95.21833801269531,
      "invalid_tail_samples": 25.5,
      "learning_rate": 2.06e-05,
      "loss": 0.1361,
      "positive_teacher_pairs": 26466.7,
      "query_h_norm_ratio": 0.4752917230129242,
      "relation_loss": 0.024994135648012162,
      "selected_pairs": 26466.7,
      "step": 310,
      "student_distance_mean": 29.985829734802245,
      "teacher_distance_mean": 0.852120041847229,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.024994135648012162,
      "weighted_relation_loss": 0.024994135648012162
    },
    {
      "action_query_batch_std": 0.4565098643302917,
      "action_query_norm": 45.25663681030274,
      "action_relation_loss": 0.02610196303576231,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.2993451824134705,
      "flow_matching_loss": 0.11007538810372353,
      "grad_norm": 0.4102122187614441,
      "h_stock_valid_norm": 95.20193099975586,
      "invalid_tail_samples": 22.7,
      "learning_rate": 2.1266666666666667e-05,
      "loss": 0.1362,
      "positive_teacher_pairs": 27105.2,
      "query_h_norm_ratio": 0.47537763714790343,
      "relation_loss": 0.02610196303576231,
      "selected_pairs": 27105.2,
      "step": 320,
      "student_distance_mean": 28.94193916320801,
      "teacher_distance_mean": 0.8549017488956452,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.02610196303576231,
      "weighted_relation_loss": 0.02610196303576231
    },
    {
      "action_query_batch_std": 0.4531555712223053,
      "action_query_norm": 45.256788635253905,
      "action_relation_loss": 0.024861196614801885,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.3086997193638915,
      "flow_matching_loss": 0.10813549160957336,
      "grad_norm": 0.5850563049316406,
      "h_stock_valid_norm": 95.3350112915039,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.1933333333333332e-05,
      "loss": 0.133,
      "positive_teacher_pairs": 26965.0,
      "query_h_norm_ratio": 0.4747141033411026,
      "relation_loss": 0.024861196614801885,
      "selected_pairs": 26965.0,
      "step": 330,
      "student_distance_mean": 28.52683792114258,
      "teacher_distance_mean": 0.8610243082046509,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.024861196614801885,
      "weighted_relation_loss": 0.024861196614801885
    },
    {
      "action_query_batch_std": 0.45859291553497317,
      "action_query_norm": 45.25651168823242,
      "action_relation_loss": 0.024374888651072978,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.31805425631431244,
      "flow_matching_loss": 0.10669438987970352,
      "grad_norm": 0.5177229642868042,
      "h_stock_valid_norm": 95.36331329345703,
      "invalid_tail_samples": 23.5,
      "learning_rate": 2.26e-05,
      "loss": 0.1311,
      "positive_teacher_pairs": 26916.4,
      "query_h_norm_ratio": 0.47457083165645597,
      "relation_loss": 0.024374888651072978,
      "selected_pairs": 26916.4,
      "step": 340,
      "student_distance_mean": 29.117178535461427,
      "teacher_distance_mean": 0.8436944901943206,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.024374888651072978,
      "weighted_relation_loss": 0.024374888651072978
    },
    {
      "action_query_batch_std": 0.4665312021970749,
      "action_query_norm": 45.25659065246582,
      "action_relation_loss": 0.023671958968043326,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.3274087932647334,
      "flow_matching_loss": 0.10677040964365006,
      "grad_norm": 0.4391489624977112,
      "h_stock_valid_norm": 95.51748046875,
      "invalid_tail_samples": 24.1,
      "learning_rate": 2.326666666666667e-05,
      "loss": 0.1304,
      "positive_teacher_pairs": 26783.9,
      "query_h_norm_ratio": 0.47380578219890596,
      "relation_loss": 0.023671958968043326,
      "selected_pairs": 26783.9,
      "step": 350,
      "student_distance_mean": 29.505405616760253,
      "teacher_distance_mean": 0.861287534236908,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.023671958968043326,
      "weighted_relation_loss": 0.023671958968043326
    },
    {
      "action_query_batch_std": 0.4889243394136429,
      "action_query_norm": 45.25648307800293,
      "action_relation_loss": 0.024939489364624024,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.33676333021515437,
      "flow_matching_loss": 0.10832568556070328,
      "grad_norm": 0.5437573790550232,
      "h_stock_valid_norm": 96.0240707397461,
      "invalid_tail_samples": 25.0,
      "learning_rate": 2.3933333333333337e-05,
      "loss": 0.1333,
      "positive_teacher_pairs": 26576.7,
      "query_h_norm_ratio": 0.4713049173355103,
      "relation_loss": 0.024939489364624024,
      "selected_pairs": 26576.7,
      "step": 360,
      "student_distance_mean": 30.935424041748046,
      "teacher_distance_mean": 0.8614113867282868,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.024939489364624024,
      "weighted_relation_loss": 0.024939489364624024
    },
    {
      "action_query_batch_std": 0.5121837884187699,
      "action_query_norm": 45.256438446044925,
      "action_relation_loss": 0.023105508647859097,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.3461178671655753,
      "flow_matching_loss": 0.10299625545740128,
      "grad_norm": 0.4543520510196686,
      "h_stock_valid_norm": 96.36481170654297,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.46e-05,
      "loss": 0.1261,
      "positive_teacher_pairs": 26965.0,
      "query_h_norm_ratio": 0.4696377545595169,
      "relation_loss": 0.023105508647859097,
      "selected_pairs": 26965.0,
      "step": 370,
      "student_distance_mean": 32.64401092529297,
      "teacher_distance_mean": 0.8600010454654694,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.023105508647859097,
      "weighted_relation_loss": 0.023105508647859097
    },
    {
      "action_query_batch_std": 0.49918396174907687,
      "action_query_norm": 45.25651626586914,
      "action_relation_loss": 0.02374581415206194,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.35547240411599623,
      "flow_matching_loss": 0.1009985513985157,
      "grad_norm": 0.34275150299072266,
      "h_stock_valid_norm": 96.42473449707032,
      "invalid_tail_samples": 24.6,
      "learning_rate": 2.5266666666666666e-05,
      "loss": 0.1247,
      "positive_teacher_pairs": 26673.6,
      "query_h_norm_ratio": 0.4693478912115097,
      "relation_loss": 0.02374581415206194,
      "selected_pairs": 26673.6,
      "step": 380,
      "student_distance_mean": 31.816055488586425,
      "teacher_distance_mean": 0.8523722171783448,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.02374581415206194,
      "weighted_relation_loss": 0.02374581415206194
    },
    {
      "action_query_batch_std": 0.4621544271707535,
      "action_query_norm": 45.25733070373535,
      "action_relation_loss": 0.023635001108050346,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.3648269410664172,
      "flow_matching_loss": 0.10474788025021553,
      "grad_norm": 0.4668126702308655,
      "h_stock_valid_norm": 96.63120803833007,
      "invalid_tail_samples": 24.5,
      "learning_rate": 2.5933333333333338e-05,
      "loss": 0.1284,
      "positive_teacher_pairs": 26691.8,
      "query_h_norm_ratio": 0.46835328042507174,
      "relation_loss": 0.023635001108050346,
      "selected_pairs": 26691.8,
      "step": 390,
      "student_distance_mean": 29.10133304595947,
      "teacher_distance_mean": 0.8537618100643158,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.023635001108050346,
      "weighted_relation_loss": 0.023635001108050346
    },
    {
      "action_query_batch_std": 0.46371265649795534,
      "action_query_norm": 45.256858825683594,
      "action_relation_loss": 0.023252296075224877,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.37418147801683815,
      "flow_matching_loss": 0.10461004227399825,
      "grad_norm": 0.43041568994522095,
      "h_stock_valid_norm": 96.75028839111329,
      "invalid_tail_samples": 23.9,
      "learning_rate": 2.6600000000000003e-05,
      "loss": 0.1279,
      "positive_teacher_pairs": 26829.5,
      "query_h_norm_ratio": 0.467772376537323,
      "relation_loss": 0.023252296075224877,
      "selected_pairs": 26829.5,
      "step": 400,
      "student_distance_mean": 29.350113487243654,
      "teacher_distance_mean": 0.8585773944854737,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.023252296075224877,
      "weighted_relation_loss": 0.023252296075224877
    },
    {
      "action_query_batch_std": 0.47937061786651614,
      "action_query_norm": 45.256323623657224,
      "action_relation_loss": 0.023377373814582825,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.38353601496725914,
      "flow_matching_loss": 0.10042736530303956,
      "grad_norm": 0.4825267195701599,
      "h_stock_valid_norm": 96.63373031616212,
      "invalid_tail_samples": 22.1,
      "learning_rate": 2.7266666666666668e-05,
      "loss": 0.1238,
      "positive_teacher_pairs": 27249.4,
      "query_h_norm_ratio": 0.4683292776346207,
      "relation_loss": 0.023377373814582825,
      "selected_pairs": 27249.4,
      "step": 410,
      "student_distance_mean": 30.48000297546387,
      "teacher_distance_mean": 0.8467832922935485,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.023377373814582825,
      "weighted_relation_loss": 0.023377373814582825
    },
    {
      "action_query_batch_std": 0.49343817830085757,
      "action_query_norm": 45.256826400756836,
      "action_relation_loss": 0.0220407884567976,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.3928905519176801,
      "flow_matching_loss": 0.1005633868277073,
      "grad_norm": 0.4436739385128021,
      "h_stock_valid_norm": 96.72578735351563,
      "invalid_tail_samples": 24.0,
      "learning_rate": 2.7933333333333332e-05,
      "loss": 0.1226,
      "positive_teacher_pairs": 26803.8,
      "query_h_norm_ratio": 0.4678899884223938,
      "relation_loss": 0.0220407884567976,
      "selected_pairs": 26803.8,
      "step": 420,
      "student_distance_mean": 31.341345596313477,
      "teacher_distance_mean": 0.8591770887374878,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0220407884567976,
      "weighted_relation_loss": 0.0220407884567976
    },
    {
      "action_query_batch_std": 0.48096750378608705,
      "action_query_norm": 45.25695991516113,
      "action_relation_loss": 0.022524821758270263,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.402245088868101,
      "flow_matching_loss": 0.10473527312278748,
      "grad_norm": 0.30535584688186646,
      "h_stock_valid_norm": 96.53056945800782,
      "invalid_tail_samples": 23.2,
      "learning_rate": 2.86e-05,
      "loss": 0.1273,
      "positive_teacher_pairs": 26985.8,
      "query_h_norm_ratio": 0.4688363939523697,
      "relation_loss": 0.022524821758270263,
      "selected_pairs": 26985.8,
      "step": 430,
      "student_distance_mean": 30.591602325439453,
      "teacher_distance_mean": 0.8617853462696076,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.022524821758270263,
      "weighted_relation_loss": 0.022524821758270263
    },
    {
      "action_query_batch_std": 0.4746852993965149,
      "action_query_norm": 45.25670967102051,
      "action_relation_loss": 0.022257195971906184,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.411599625818522,
      "flow_matching_loss": 0.10055308789014816,
      "grad_norm": 0.35708656907081604,
      "h_stock_valid_norm": 96.64959716796875,
      "invalid_tail_samples": 26.0,
      "learning_rate": 2.926666666666667e-05,
      "loss": 0.1228,
      "positive_teacher_pairs": 26354.5,
      "query_h_norm_ratio": 0.4682571619749069,
      "relation_loss": 0.022257195971906184,
      "selected_pairs": 26354.5,
      "step": 440,
      "student_distance_mean": 30.131527137756347,
      "teacher_distance_mean": 0.8463669955730438,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.022257195971906184,
      "weighted_relation_loss": 0.022257195971906184
    },
    {
      "action_query_batch_std": 0.48006944060325624,
      "action_query_norm": 45.25662536621094,
      "action_relation_loss": 0.021282754093408584,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.42095416276894293,
      "flow_matching_loss": 0.09970473125576973,
      "grad_norm": 0.4277660846710205,
      "h_stock_valid_norm": 96.94928665161133,
      "invalid_tail_samples": 23.1,
      "learning_rate": 2.9933333333333337e-05,
      "loss": 0.121,
      "positive_teacher_pairs": 27014.6,
      "query_h_norm_ratio": 0.4668083071708679,
      "relation_loss": 0.021282754093408584,
      "selected_pairs": 27014.6,
      "step": 450,
      "student_distance_mean": 30.47330665588379,
      "teacher_distance_mean": 0.8585263848304748,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.021282754093408584,
      "weighted_relation_loss": 0.021282754093408584
    },
    {
      "action_query_batch_std": 0.4792586177587509,
      "action_query_norm": 45.25680656433106,
      "action_relation_loss": 0.02327370084822178,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.4303086997193639,
      "flow_matching_loss": 0.09763320162892342,
      "grad_norm": 0.47225242853164673,
      "h_stock_valid_norm": 96.57506866455078,
      "invalid_tail_samples": 24.8,
      "learning_rate": 3.06e-05,
      "loss": 0.1209,
      "positive_teacher_pairs": 26617.4,
      "query_h_norm_ratio": 0.46861917674541476,
      "relation_loss": 0.02327370084822178,
      "selected_pairs": 26617.4,
      "step": 460,
      "student_distance_mean": 30.348296928405762,
      "teacher_distance_mean": 0.8538972079753876,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.02327370084822178,
      "weighted_relation_loss": 0.02327370084822178
    },
    {
      "action_query_batch_std": 0.4940468430519104,
      "action_query_norm": 45.256442642211915,
      "action_relation_loss": 0.022259348630905153,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.43966323666978485,
      "flow_matching_loss": 0.09420531019568443,
      "grad_norm": 0.28766506910324097,
      "h_stock_valid_norm": 96.31457214355468,
      "invalid_tail_samples": 27.6,
      "learning_rate": 3.126666666666666e-05,
      "loss": 0.1165,
      "positive_teacher_pairs": 25988.2,
      "query_h_norm_ratio": 0.46988321244716647,
      "relation_loss": 0.022259348630905153,
      "selected_pairs": 25988.2,
      "step": 470,
      "student_distance_mean": 31.242060470581055,
      "teacher_distance_mean": 0.8478752493858337,
      "valid_action_samples": 228.4,
      "weighted_auxiliary_loss": 0.022259348630905153,
      "weighted_relation_loss": 0.022259348630905153
    },
    {
      "action_query_batch_std": 0.5286127209663392,
      "action_query_norm": 45.25654182434082,
      "action_relation_loss": 0.022496894374489783,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.4490177736202058,
      "flow_matching_loss": 0.09681384041905403,
      "grad_norm": 0.36745336651802063,
      "h_stock_valid_norm": 96.13996124267578,
      "invalid_tail_samples": 25.7,
      "learning_rate": 3.1933333333333335e-05,
      "loss": 0.1193,
      "positive_teacher_pairs": 26416.2,
      "query_h_norm_ratio": 0.47073734402656553,
      "relation_loss": 0.022496894374489783,
      "selected_pairs": 26416.2,
      "step": 480,
      "student_distance_mean": 33.40840187072754,
      "teacher_distance_mean": 0.858701741695404,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.022496894374489783,
      "weighted_relation_loss": 0.022496894374489783
    },
    {
      "action_query_batch_std": 0.5118881613016129,
      "action_query_norm": 45.2566535949707,
      "action_relation_loss": 0.021813048608601095,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.4583723105706268,
      "flow_matching_loss": 0.09601019844412803,
      "grad_norm": 0.36277130246162415,
      "h_stock_valid_norm": 95.84358444213868,
      "invalid_tail_samples": 25.0,
      "learning_rate": 3.26e-05,
      "loss": 0.1178,
      "positive_teacher_pairs": 26572.2,
      "query_h_norm_ratio": 0.4721961826086044,
      "relation_loss": 0.021813048608601095,
      "selected_pairs": 26572.2,
      "step": 490,
      "student_distance_mean": 32.38370590209961,
      "teacher_distance_mean": 0.8495366454124451,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.021813048608601095,
      "weighted_relation_loss": 0.021813048608601095
    },
    {
      "action_query_batch_std": 0.48167942464351654,
      "action_query_norm": 45.25676002502441,
      "action_relation_loss": 0.0212163295596838,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.4677268475210477,
      "flow_matching_loss": 0.09926914349198342,
      "grad_norm": 0.30514898896217346,
      "h_stock_valid_norm": 95.53792114257813,
      "invalid_tail_samples": 25.1,
      "learning_rate": 3.326666666666667e-05,
      "loss": 0.1205,
      "positive_teacher_pairs": 26546.6,
      "query_h_norm_ratio": 0.47370740175247195,
      "relation_loss": 0.0212163295596838,
      "selected_pairs": 26546.6,
      "step": 500,
      "student_distance_mean": 30.547501945495604,
      "teacher_distance_mean": 0.8632226824760437,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0212163295596838,
      "weighted_relation_loss": 0.0212163295596838
    },
    {
      "action_query_batch_std": 0.49905869364738464,
      "action_query_norm": 45.25616645812988,
      "action_relation_loss": 0.021084764786064625,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.47708138447146864,
      "flow_matching_loss": 0.09469894990324974,
      "grad_norm": 0.4034586250782013,
      "h_stock_valid_norm": 95.89756851196289,
      "invalid_tail_samples": 22.9,
      "learning_rate": 3.3933333333333336e-05,
      "loss": 0.1158,
      "positive_teacher_pairs": 27058.2,
      "query_h_norm_ratio": 0.47192347049713135,
      "relation_loss": 0.021084764786064625,
      "selected_pairs": 27058.2,
      "step": 510,
      "student_distance_mean": 31.834549713134766,
      "teacher_distance_mean": 0.8572067856788635,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.021084764786064625,
      "weighted_relation_loss": 0.021084764786064625
    },
    {
      "action_query_batch_std": 0.4881055921316147,
      "action_query_norm": 45.2558235168457,
      "action_relation_loss": 0.021940629929304123,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.4864359214218896,
      "flow_matching_loss": 0.0939585581421852,
      "grad_norm": 0.5429545044898987,
      "h_stock_valid_norm": 95.86923370361328,
      "invalid_tail_samples": 21.1,
      "learning_rate": 3.46e-05,
      "loss": 0.1159,
      "positive_teacher_pairs": 27478.1,
      "query_h_norm_ratio": 0.472059366106987,
      "relation_loss": 0.021940629929304123,
      "selected_pairs": 27478.1,
      "step": 520,
      "student_distance_mean": 30.802318000793456,
      "teacher_distance_mean": 0.8518754422664643,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.021940629929304123,
      "weighted_relation_loss": 0.021940629929304123
    },
    {
      "action_query_batch_std": 0.48475478291511537,
      "action_query_norm": 45.25548782348633,
      "action_relation_loss": 0.021802045591175555,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.49579045837231056,
      "flow_matching_loss": 0.09513184577226638,
      "grad_norm": 0.6371195316314697,
      "h_stock_valid_norm": 96.21396026611328,
      "invalid_tail_samples": 26.2,
      "learning_rate": 3.526666666666667e-05,
      "loss": 0.1169,
      "positive_teacher_pairs": 26298.5,
      "query_h_norm_ratio": 0.4703699737787247,
      "relation_loss": 0.021802045591175555,
      "selected_pairs": 26298.5,
      "step": 530,
      "student_distance_mean": 30.819536209106445,
      "teacher_distance_mean": 0.8477982223033905,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.021802045591175555,
      "weighted_relation_loss": 0.021802045591175555
    },
    {
      "action_query_batch_std": 0.5049764811992645,
      "action_query_norm": 45.256241989135745,
      "action_relation_loss": 0.021209322661161423,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.5051449953227315,
      "flow_matching_loss": 0.09623007550835609,
      "grad_norm": 0.38474512100219727,
      "h_stock_valid_norm": 97.36184539794922,
      "invalid_tail_samples": 22.9,
      "learning_rate": 3.593333333333334e-05,
      "loss": 0.1174,
      "positive_teacher_pairs": 27060.5,
      "query_h_norm_ratio": 0.464829421043396,
      "relation_loss": 0.021209322661161423,
      "selected_pairs": 27060.5,
      "step": 540,
      "student_distance_mean": 31.8910701751709,
      "teacher_distance_mean": 0.8710597097873688,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.021209322661161423,
      "weighted_relation_loss": 0.021209322661161423
    },
    {
      "action_query_batch_std": 0.5155073523521423,
      "action_query_norm": 45.256065368652344,
      "action_relation_loss": 0.02092938609421253,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.5144995322731525,
      "flow_matching_loss": 0.09696928486227989,
      "grad_norm": 0.37790387868881226,
      "h_stock_valid_norm": 97.46875839233398,
      "invalid_tail_samples": 26.5,
      "learning_rate": 3.66e-05,
      "loss": 0.1179,
      "positive_teacher_pairs": 26225.1,
      "query_h_norm_ratio": 0.4643148332834244,
      "relation_loss": 0.02092938609421253,
      "selected_pairs": 26225.1,
      "step": 550,
      "student_distance_mean": 32.73186473846435,
      "teacher_distance_mean": 0.8591363251209259,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.02092938609421253,
      "weighted_relation_loss": 0.02092938609421253
    },
    {
      "action_query_batch_std": 0.48947266340255735,
      "action_query_norm": 45.25603828430176,
      "action_relation_loss": 0.02126255426555872,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.5238540692235735,
      "flow_matching_loss": 0.094685410708189,
      "grad_norm": 0.43230271339416504,
      "h_stock_valid_norm": 97.18167495727539,
      "invalid_tail_samples": 24.7,
      "learning_rate": 3.726666666666667e-05,
      "loss": 0.1159,
      "positive_teacher_pairs": 26645.2,
      "query_h_norm_ratio": 0.46568822860717773,
      "relation_loss": 0.02126255426555872,
      "selected_pairs": 26645.2,
      "step": 560,
      "student_distance_mean": 31.105400466918944,
      "teacher_distance_mean": 0.8429630935192108,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.02126255426555872,
      "weighted_relation_loss": 0.02126255426555872
    },
    {
      "action_query_batch_std": 0.4853668540716171,
      "action_query_norm": 45.25564804077148,
      "action_relation_loss": 0.02088518477976322,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.5332086061739943,
      "flow_matching_loss": 0.09107317179441451,
      "grad_norm": 0.40171584486961365,
      "h_stock_valid_norm": 96.6721565246582,
      "invalid_tail_samples": 23.8,
      "learning_rate": 3.793333333333334e-05,
      "loss": 0.112,
      "positive_teacher_pairs": 26852.9,
      "query_h_norm_ratio": 0.46813870668411256,
      "relation_loss": 0.02088518477976322,
      "selected_pairs": 26852.9,
      "step": 570,
      "student_distance_mean": 30.806797599792482,
      "teacher_distance_mean": 0.8552877366542816,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.02088518477976322,
      "weighted_relation_loss": 0.02088518477976322
    },
    {
      "action_query_batch_std": 0.4980041414499283,
      "action_query_norm": 45.25537338256836,
      "action_relation_loss": 0.02060304619371891,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.5425631431244153,
      "flow_matching_loss": 0.09533087611198425,
      "grad_norm": 0.3441754877567291,
      "h_stock_valid_norm": 96.29734878540039,
      "invalid_tail_samples": 22.3,
      "learning_rate": 3.86e-05,
      "loss": 0.1159,
      "positive_teacher_pairs": 27201.6,
      "query_h_norm_ratio": 0.46995659470558165,
      "relation_loss": 0.02060304619371891,
      "selected_pairs": 27201.6,
      "step": 580,
      "student_distance_mean": 31.59707851409912,
      "teacher_distance_mean": 0.8586687326431275,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.02060304619371891,
      "weighted_relation_loss": 0.02060304619371891
    },
    {
      "action_query_batch_std": 0.5132920205593109,
      "action_query_norm": 45.25534286499023,
      "action_relation_loss": 0.021050027385354043,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.5519176800748363,
      "flow_matching_loss": 0.09511511996388436,
      "grad_norm": 0.286603718996048,
      "h_stock_valid_norm": 96.40039443969727,
      "invalid_tail_samples": 23.8,
      "learning_rate": 3.926666666666667e-05,
      "loss": 0.1162,
      "positive_teacher_pairs": 26848.9,
      "query_h_norm_ratio": 0.4694548398256302,
      "relation_loss": 0.021050027385354043,
      "selected_pairs": 26848.9,
      "step": 590,
      "student_distance_mean": 32.507528114318845,
      "teacher_distance_mean": 0.8379797995090484,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.021050027385354043,
      "weighted_relation_loss": 0.021050027385354043
    },
    {
      "action_query_batch_std": 0.5321188569068909,
      "action_query_norm": 45.255349731445314,
      "action_relation_loss": 0.01927350088953972,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.5612722170252572,
      "flow_matching_loss": 0.09081926345825195,
      "grad_norm": 0.3088436722755432,
      "h_stock_valid_norm": 96.36191635131836,
      "invalid_tail_samples": 25.9,
      "learning_rate": 3.993333333333333e-05,
      "loss": 0.1101,
      "positive_teacher_pairs": 26365.9,
      "query_h_norm_ratio": 0.4696418046951294,
      "relation_loss": 0.01927350088953972,
      "selected_pairs": 26365.9,
      "step": 600,
      "student_distance_mean": 33.75249481201172,
      "teacher_distance_mean": 0.8593577206134796,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.01927350088953972,
      "weighted_relation_loss": 0.01927350088953972
    },
    {
      "action_query_batch_std": 0.5033578127622604,
      "action_query_norm": 45.25498962402344,
      "action_relation_loss": 0.020666810870170593,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.5706267539756782,
      "flow_matching_loss": 0.09117596000432968,
      "grad_norm": 0.3631654679775238,
      "h_stock_valid_norm": 96.49747695922852,
      "invalid_tail_samples": 25.7,
      "learning_rate": 4.0600000000000004e-05,
      "loss": 0.1118,
      "positive_teacher_pairs": 26413.6,
      "query_h_norm_ratio": 0.46897701621055604,
      "relation_loss": 0.020666810870170593,
      "selected_pairs": 26413.6,
      "step": 610,
      "student_distance_mean": 32.01097869873047,
      "teacher_distance_mean": 0.8554817795753479,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.020666810870170593,
      "weighted_relation_loss": 0.020666810870170593
    },
    {
      "action_query_batch_std": 0.5182930022478104,
      "action_query_norm": 45.254581451416016,
      "action_relation_loss": 0.020346792042255403,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.5799812909260992,
      "flow_matching_loss": 0.08780078813433648,
      "grad_norm": 0.4012930393218994,
      "h_stock_valid_norm": 96.72244491577149,
      "invalid_tail_samples": 22.0,
      "learning_rate": 4.126666666666667e-05,
      "loss": 0.1081,
      "positive_teacher_pairs": 27270.6,
      "query_h_norm_ratio": 0.46788251399993896,
      "relation_loss": 0.020346792042255403,
      "selected_pairs": 27270.6,
      "step": 620,
      "student_distance_mean": 32.8489408493042,
      "teacher_distance_mean": 0.8501481115818024,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.020346792042255403,
      "weighted_relation_loss": 0.020346792042255403
    },
    {
      "action_query_batch_std": 0.5139294534921646,
      "action_query_norm": 45.25493125915527,
      "action_relation_loss": 0.01992031019181013,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.5893358278765201,
      "flow_matching_loss": 0.09370923489332199,
      "grad_norm": 0.3455600142478943,
      "h_stock_valid_norm": 97.26699829101562,
      "invalid_tail_samples": 26.2,
      "learning_rate": 4.1933333333333334e-05,
      "loss": 0.1136,
      "positive_teacher_pairs": 26306.5,
      "query_h_norm_ratio": 0.4652719110250473,
      "relation_loss": 0.01992031019181013,
      "selected_pairs": 26306.5,
      "step": 630,
      "student_distance_mean": 32.415658187866214,
      "teacher_distance_mean": 0.8518160104751586,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.01992031019181013,
      "weighted_relation_loss": 0.01992031019181013
    },
    {
      "action_query_batch_std": 0.502452152967453,
      "action_query_norm": 45.25472717285156,
      "action_relation_loss": 0.02089403010904789,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.598690364826941,
      "flow_matching_loss": 0.09378673955798149,
      "grad_norm": 0.4940584897994995,
      "h_stock_valid_norm": 97.50380783081054,
      "invalid_tail_samples": 22.2,
      "learning_rate": 4.26e-05,
      "loss": 0.1147,
      "positive_teacher_pairs": 27222.9,
      "query_h_norm_ratio": 0.4641346275806427,
      "relation_loss": 0.02089403010904789,
      "selected_pairs": 27222.9,
      "step": 640,
      "student_distance_mean": 31.880283546447753,
      "teacher_distance_mean": 0.8516839206218719,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.02089403010904789,
      "weighted_relation_loss": 0.02089403010904789
    },
    {
      "action_query_batch_std": 0.5187026143074036,
      "action_query_norm": 45.25464630126953,
      "action_relation_loss": 0.01858647521585226,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.608044901777362,
      "flow_matching_loss": 0.08914723470807076,
      "grad_norm": 0.2590837776660919,
      "h_stock_valid_norm": 97.4058349609375,
      "invalid_tail_samples": 24.2,
      "learning_rate": 4.3266666666666664e-05,
      "loss": 0.1077,
      "positive_teacher_pairs": 26754.4,
      "query_h_norm_ratio": 0.46460055112838744,
      "relation_loss": 0.01858647521585226,
      "selected_pairs": 26754.4,
      "step": 650,
      "student_distance_mean": 32.913783645629884,
      "teacher_distance_mean": 0.8521072566509247,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.01858647521585226,
      "weighted_relation_loss": 0.01858647521585226
    },
    {
      "action_query_batch_std": 0.5109952032566071,
      "action_query_norm": 45.2540843963623,
      "action_relation_loss": 0.01984447780996561,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.617399438727783,
      "flow_matching_loss": 0.09398146569728852,
      "grad_norm": 0.33037543296813965,
      "h_stock_valid_norm": 97.38190002441407,
      "invalid_tail_samples": 24.6,
      "learning_rate": 4.3933333333333335e-05,
      "loss": 0.1138,
      "positive_teacher_pairs": 26672.1,
      "query_h_norm_ratio": 0.4647092461585999,
      "relation_loss": 0.01984447780996561,
      "selected_pairs": 26672.1,
      "step": 660,
      "student_distance_mean": 32.31416683197021,
      "teacher_distance_mean": 0.8574969232082367,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.01984447780996561,
      "weighted_relation_loss": 0.01984447780996561
    },
    {
      "action_query_batch_std": 0.5403798520565033,
      "action_query_norm": 45.25357780456543,
      "action_relation_loss": 0.020305102691054344,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.6267539756782039,
      "flow_matching_loss": 0.09225097596645356,
      "grad_norm": 0.4365817606449127,
      "h_stock_valid_norm": 96.64236145019531,
      "invalid_tail_samples": 21.8,
      "learning_rate": 4.46e-05,
      "loss": 0.1126,
      "positive_teacher_pairs": 27316.5,
      "query_h_norm_ratio": 0.46826314330101015,
      "relation_loss": 0.020305102691054344,
      "selected_pairs": 27316.5,
      "step": 670,
      "student_distance_mean": 34.25883140563965,
      "teacher_distance_mean": 0.8672927916049957,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.020305102691054344,
      "weighted_relation_loss": 0.020305102691054344
    },
    {
      "action_query_batch_std": 0.5473604679107666,
      "action_query_norm": 45.25357246398926,
      "action_relation_loss": 0.02023650910705328,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.6361085126286249,
      "flow_matching_loss": 0.09372643604874611,
      "grad_norm": 0.6931991577148438,
      "h_stock_valid_norm": 95.71630249023437,
      "invalid_tail_samples": 23.5,
      "learning_rate": 4.526666666666667e-05,
      "loss": 0.114,
      "positive_teacher_pairs": 26925.6,
      "query_h_norm_ratio": 0.4727943897247314,
      "relation_loss": 0.02023650910705328,
      "selected_pairs": 26925.6,
      "step": 680,
      "student_distance_mean": 34.81376075744629,
      "teacher_distance_mean": 0.8570481896400451,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.02023650910705328,
      "weighted_relation_loss": 0.02023650910705328
    },
    {
      "action_query_batch_std": 0.5111295938491821,
      "action_query_norm": 45.253808212280276,
      "action_relation_loss": 0.01949322447180748,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.6454630495790459,
      "flow_matching_loss": 0.09199377298355102,
      "grad_norm": 0.41025716066360474,
      "h_stock_valid_norm": 95.7040802001953,
      "invalid_tail_samples": 21.5,
      "learning_rate": 4.5933333333333336e-05,
      "loss": 0.1115,
      "positive_teacher_pairs": 27382.9,
      "query_h_norm_ratio": 0.4728526145219803,
      "relation_loss": 0.01949322447180748,
      "selected_pairs": 27382.9,
      "step": 690,
      "student_distance_mean": 32.475844383239746,
      "teacher_distance_mean": 0.8531707227230072,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.01949322447180748,
      "weighted_relation_loss": 0.01949322447180748
    },
    {
      "action_query_batch_std": 0.5204510927200318,
      "action_query_norm": 45.25314865112305,
      "action_relation_loss": 0.020291266217827796,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.6548175865294668,
      "flow_matching_loss": 0.09167516455054284,
      "grad_norm": 0.3894639313220978,
      "h_stock_valid_norm": 96.34866867065429,
      "invalid_tail_samples": 25.0,
      "learning_rate": 4.660000000000001e-05,
      "loss": 0.112,
      "positive_teacher_pairs": 26574.2,
      "query_h_norm_ratio": 0.4696871429681778,
      "relation_loss": 0.020291266217827796,
      "selected_pairs": 26574.2,
      "step": 700,
      "student_distance_mean": 33.091475296020505,
      "teacher_distance_mean": 0.8632028281688691,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.020291266217827796,
      "weighted_relation_loss": 0.020291266217827796
    },
    {
      "action_query_batch_std": 0.49502935707569123,
      "action_query_norm": 45.252310562133786,
      "action_relation_loss": 0.01966202221810818,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.6641721234798877,
      "flow_matching_loss": 0.09121390208601951,
      "grad_norm": 0.4862385392189026,
      "h_stock_valid_norm": 97.02265243530273,
      "invalid_tail_samples": 23.9,
      "learning_rate": 4.726666666666667e-05,
      "loss": 0.1109,
      "positive_teacher_pairs": 26826.8,
      "query_h_norm_ratio": 0.4664168477058411,
      "relation_loss": 0.01966202221810818,
      "selected_pairs": 26826.8,
      "step": 710,
      "student_distance_mean": 31.56130485534668,
      "teacher_distance_mean": 0.8586376190185547,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.01966202221810818,
      "weighted_relation_loss": 0.01966202221810818
    },
    {
      "action_query_batch_std": 0.5072650104761124,
      "action_query_norm": 45.25191116333008,
      "action_relation_loss": 0.019692231714725495,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.6735266604303087,
      "flow_matching_loss": 0.09144746214151382,
      "grad_norm": 0.39159679412841797,
      "h_stock_valid_norm": 97.87732315063477,
      "invalid_tail_samples": 21.4,
      "learning_rate": 4.793333333333334e-05,
      "loss": 0.1111,
      "positive_teacher_pairs": 27413.5,
      "query_h_norm_ratio": 0.4623350232839584,
      "relation_loss": 0.019692231714725495,
      "selected_pairs": 27413.5,
      "step": 720,
      "student_distance_mean": 32.365109252929685,
      "teacher_distance_mean": 0.8580571472644806,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.019692231714725495,
      "weighted_relation_loss": 0.019692231714725495
    },
    {
      "action_query_batch_std": 0.525002783536911,
      "action_query_norm": 45.25178108215332,
      "action_relation_loss": 0.01904018959030509,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.6828811973807296,
      "flow_matching_loss": 0.08895643204450607,
      "grad_norm": 0.4402831196784973,
      "h_stock_valid_norm": 98.27630844116212,
      "invalid_tail_samples": 23.3,
      "learning_rate": 4.86e-05,
      "loss": 0.108,
      "positive_teacher_pairs": 26964.0,
      "query_h_norm_ratio": 0.4604576587677002,
      "relation_loss": 0.01904018959030509,
      "selected_pairs": 26964.0,
      "step": 730,
      "student_distance_mean": 33.45682716369629,
      "teacher_distance_mean": 0.8464121282100677,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.01904018959030509,
      "weighted_relation_loss": 0.01904018959030509
    },
    {
      "action_query_batch_std": 0.5296186685562134,
      "action_query_norm": 45.251416778564455,
      "action_relation_loss": 0.019914195127785205,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.6922357343311506,
      "flow_matching_loss": 0.0905018426477909,
      "grad_norm": 0.27078160643577576,
      "h_stock_valid_norm": 98.1809211730957,
      "invalid_tail_samples": 22.1,
      "learning_rate": 4.926666666666667e-05,
      "loss": 0.1104,
      "positive_teacher_pairs": 27245.2,
      "query_h_norm_ratio": 0.46090319752693176,
      "relation_loss": 0.019914195127785205,
      "selected_pairs": 27245.2,
      "step": 740,
      "student_distance_mean": 33.62534294128418,
      "teacher_distance_mean": 0.8535625696182251,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.019914195127785205,
      "weighted_relation_loss": 0.019914195127785205
    },
    {
      "action_query_batch_std": 0.5577385902404786,
      "action_query_norm": 45.25214195251465,
      "action_relation_loss": 0.01924386452883482,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.7015902712815716,
      "flow_matching_loss": 0.08579925075173378,
      "grad_norm": 0.34719592332839966,
      "h_stock_valid_norm": 97.38483810424805,
      "invalid_tail_samples": 22.9,
      "learning_rate": 4.993333333333334e-05,
      "loss": 0.105,
      "positive_teacher_pairs": 27059.8,
      "query_h_norm_ratio": 0.4646778374910355,
      "relation_loss": 0.01924386452883482,
      "selected_pairs": 27059.8,
      "step": 750,
      "student_distance_mean": 35.46399841308594,
      "teacher_distance_mean": 0.8574023485183716,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.01924386452883482,
      "weighted_relation_loss": 0.01924386452883482
    },
    {
      "action_query_batch_std": 0.5239206850528717,
      "action_query_norm": 45.25196151733398,
      "action_relation_loss": 0.019335180521011353,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.7109448082319925,
      "flow_matching_loss": 0.0872899554669857,
      "grad_norm": 0.36034095287323,
      "h_stock_valid_norm": 96.15402908325196,
      "invalid_tail_samples": 26.6,
      "learning_rate": 5.0600000000000003e-05,
      "loss": 0.1066,
      "positive_teacher_pairs": 26211.2,
      "query_h_norm_ratio": 0.4706262797117233,
      "relation_loss": 0.019335180521011353,
      "selected_pairs": 26211.2,
      "step": 760,
      "student_distance_mean": 33.18306617736816,
      "teacher_distance_mean": 0.8562582790851593,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.019335180521011353,
      "weighted_relation_loss": 0.019335180521011353
    },
    {
      "action_query_batch_std": 0.5371475398540497,
      "action_query_norm": 45.25166816711426,
      "action_relation_loss": 0.01909452471882105,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.7202993451824135,
      "flow_matching_loss": 0.08506943136453629,
      "grad_norm": 0.5093770623207092,
      "h_stock_valid_norm": 95.96199111938476,
      "invalid_tail_samples": 23.0,
      "learning_rate": 5.1266666666666675e-05,
      "loss": 0.1042,
      "positive_teacher_pairs": 27034.8,
      "query_h_norm_ratio": 0.4715652674436569,
      "relation_loss": 0.01909452471882105,
      "selected_pairs": 27034.8,
      "step": 770,
      "student_distance_mean": 34.18290138244629,
      "teacher_distance_mean": 0.8528332710266113,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.01909452471882105,
      "weighted_relation_loss": 0.01909452471882105
    },
    {
      "action_query_batch_std": 0.5144052475690841,
      "action_query_norm": 45.2509407043457,
      "action_relation_loss": 0.01913960315287113,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.7296538821328344,
      "flow_matching_loss": 0.0890475831925869,
      "grad_norm": 0.41841498017311096,
      "h_stock_valid_norm": 96.52865295410156,
      "invalid_tail_samples": 22.8,
      "learning_rate": 5.193333333333333e-05,
      "loss": 0.1082,
      "positive_teacher_pairs": 27082.2,
      "query_h_norm_ratio": 0.4687902718782425,
      "relation_loss": 0.01913960315287113,
      "selected_pairs": 27082.2,
      "step": 780,
      "student_distance_mean": 32.756714248657225,
      "teacher_distance_mean": 0.8558024168014526,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.01913960315287113,
      "weighted_relation_loss": 0.01913960315287113
    },
    {
      "action_query_batch_std": 0.4996920436620712,
      "action_query_norm": 45.24987716674805,
      "action_relation_loss": 0.019660762697458267,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.7390084190832554,
      "flow_matching_loss": 0.08727775737643242,
      "grad_norm": 0.4020824730396271,
      "h_stock_valid_norm": 96.95255889892579,
      "invalid_tail_samples": 22.9,
      "learning_rate": 5.2600000000000005e-05,
      "loss": 0.1069,
      "positive_teacher_pairs": 27058.4,
      "query_h_norm_ratio": 0.46672402918338773,
      "relation_loss": 0.019660762697458267,
      "selected_pairs": 27058.4,
      "step": 790,
      "student_distance_mean": 31.70549602508545,
      "teacher_distance_mean": 0.8593971133232117,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.019660762697458267,
      "weighted_relation_loss": 0.019660762697458267
    },
    {
      "action_query_batch_std": 0.5277751386165619,
      "action_query_norm": 45.2495777130127,
      "action_relation_loss": 0.019661208800971507,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.7483629560336763,
      "flow_matching_loss": 0.08700161650776864,
      "grad_norm": 0.3619897663593292,
      "h_stock_valid_norm": 97.16277313232422,
      "invalid_tail_samples": 22.7,
      "learning_rate": 5.326666666666666e-05,
      "loss": 0.1067,
      "positive_teacher_pairs": 27107.8,
      "query_h_norm_ratio": 0.4657104402780533,
      "relation_loss": 0.019661208800971507,
      "selected_pairs": 27107.8,
      "step": 800,
      "student_distance_mean": 33.588630294799806,
      "teacher_distance_mean": 0.8490899443626404,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.019661208800971507,
      "weighted_relation_loss": 0.019661208800971507
    },
    {
      "action_query_batch_std": 0.5402280509471893,
      "action_query_norm": 45.25002784729004,
      "action_relation_loss": 0.018827486597001553,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.7577174929840973,
      "flow_matching_loss": 0.08577365279197693,
      "grad_norm": 0.24725519120693207,
      "h_stock_valid_norm": 97.37417984008789,
      "invalid_tail_samples": 23.0,
      "learning_rate": 5.3933333333333334e-05,
      "loss": 0.1046,
      "positive_teacher_pairs": 27045.8,
      "query_h_norm_ratio": 0.4647074490785599,
      "relation_loss": 0.018827486597001553,
      "selected_pairs": 27045.8,
      "step": 810,
      "student_distance_mean": 34.27178955078125,
      "teacher_distance_mean": 0.8540050446987152,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.018827486597001553,
      "weighted_relation_loss": 0.018827486597001553
    },
    {
      "action_query_batch_std": 0.5379411697387695,
      "action_query_norm": 45.24869384765625,
      "action_relation_loss": 0.018178061209619047,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.7670720299345183,
      "flow_matching_loss": 0.08754006251692772,
      "grad_norm": 0.28390395641326904,
      "h_stock_valid_norm": 97.25664443969727,
      "invalid_tail_samples": 21.9,
      "learning_rate": 5.4600000000000006e-05,
      "loss": 0.1057,
      "positive_teacher_pairs": 27290.8,
      "query_h_norm_ratio": 0.4652534455060959,
      "relation_loss": 0.018178061209619047,
      "selected_pairs": 27290.8,
      "step": 820,
      "student_distance_mean": 34.29466705322265,
      "teacher_distance_mean": 0.8606010735034942,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.018178061209619047,
      "weighted_relation_loss": 0.018178061209619047
    },
    {
      "action_query_batch_std": 0.5000660419464111,
      "action_query_norm": 45.2482723236084,
      "action_relation_loss": 0.018275098502635957,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.7764265668849392,
      "flow_matching_loss": 0.09063779935240746,
      "grad_norm": 0.34257224202156067,
      "h_stock_valid_norm": 97.4121307373047,
      "invalid_tail_samples": 24.1,
      "learning_rate": 5.5266666666666664e-05,
      "loss": 0.1089,
      "positive_teacher_pairs": 26780.2,
      "query_h_norm_ratio": 0.46450495421886445,
      "relation_loss": 0.018275098502635957,
      "selected_pairs": 26780.2,
      "step": 830,
      "student_distance_mean": 31.76173915863037,
      "teacher_distance_mean": 0.8602522850036621,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.018275098502635957,
      "weighted_relation_loss": 0.018275098502635957
    },
    {
      "action_query_batch_std": 0.5002553492784501,
      "action_query_norm": 45.247549057006836,
      "action_relation_loss": 0.018137018755078317,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.7857811038353602,
      "flow_matching_loss": 0.08645572289824485,
      "grad_norm": 0.30785414576530457,
      "h_stock_valid_norm": 97.38591690063477,
      "invalid_tail_samples": 22.4,
      "learning_rate": 5.5933333333333335e-05,
      "loss": 0.1046,
      "positive_teacher_pairs": 27173.0,
      "query_h_norm_ratio": 0.46462497115135193,
      "relation_loss": 0.018137018755078317,
      "selected_pairs": 27173.0,
      "step": 840,
      "student_distance_mean": 31.79878215789795,
      "teacher_distance_mean": 0.8439851760864258,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.018137018755078317,
      "weighted_relation_loss": 0.018137018755078317
    },
    {
      "action_query_batch_std": 0.5405885934829712,
      "action_query_norm": 45.2474781036377,
      "action_relation_loss": 0.018539090640842915,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.7951356407857811,
      "flow_matching_loss": 0.08862527385354042,
      "grad_norm": 0.3659743666648865,
      "h_stock_valid_norm": 97.58162460327148,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.66e-05,
      "loss": 0.1072,
      "positive_teacher_pairs": 26987.0,
      "query_h_norm_ratio": 0.46369039714336396,
      "relation_loss": 0.018539090640842915,
      "selected_pairs": 26987.0,
      "step": 850,
      "student_distance_mean": 34.469049072265626,
      "teacher_distance_mean": 0.857982736825943,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.018539090640842915,
      "weighted_relation_loss": 0.018539090640842915
    },
    {
      "action_query_batch_std": 0.5540976524353027,
      "action_query_norm": 45.247823333740236,
      "action_relation_loss": 0.01947486773133278,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.804490177736202,
      "flow_matching_loss": 0.08409416899085045,
      "grad_norm": 0.28217431902885437,
      "h_stock_valid_norm": 97.88681411743164,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.726666666666667e-05,
      "loss": 0.1036,
      "positive_teacher_pairs": 26989.0,
      "query_h_norm_ratio": 0.462249282002449,
      "relation_loss": 0.01947486773133278,
      "selected_pairs": 26989.0,
      "step": 860,
      "student_distance_mean": 35.22966003417969,
      "teacher_distance_mean": 0.8604804277420044,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.01947486773133278,
      "weighted_relation_loss": 0.01947486773133278
    },
    {
      "action_query_batch_std": 0.548593407869339,
      "action_query_norm": 45.246778106689455,
      "action_relation_loss": 0.017974099889397622,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.813844714686623,
      "flow_matching_loss": 0.08834164664149284,
      "grad_norm": 0.4356127679347992,
      "h_stock_valid_norm": 97.19758453369141,
      "invalid_tail_samples": 24.5,
      "learning_rate": 5.7933333333333337e-05,
      "loss": 0.1063,
      "positive_teacher_pairs": 26689.3,
      "query_h_norm_ratio": 0.46551795601844786,
      "relation_loss": 0.017974099889397622,
      "selected_pairs": 26689.3,
      "step": 870,
      "student_distance_mean": 34.92741775512695,
      "teacher_distance_mean": 0.8650976240634918,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.017974099889397622,
      "weighted_relation_loss": 0.017974099889397622
    },
    {
      "action_query_batch_std": 0.5213864326477051,
      "action_query_norm": 45.24639701843262,
      "action_relation_loss": 0.01908685527741909,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.823199251637044,
      "flow_matching_loss": 0.08445915803313256,
      "grad_norm": 0.2878987491130829,
      "h_stock_valid_norm": 97.56967697143554,
      "invalid_tail_samples": 24.2,
      "learning_rate": 5.86e-05,
      "loss": 0.1035,
      "positive_teacher_pairs": 26760.4,
      "query_h_norm_ratio": 0.4637359231710434,
      "relation_loss": 0.01908685527741909,
      "selected_pairs": 26760.4,
      "step": 880,
      "student_distance_mean": 33.17245101928711,
      "teacher_distance_mean": 0.8410179138183593,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.01908685527741909,
      "weighted_relation_loss": 0.01908685527741909
    },
    {
      "action_query_batch_std": 0.5158852636814117,
      "action_query_norm": 45.24598731994629,
      "action_relation_loss": 0.019156492315232754,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.8325537885874649,
      "flow_matching_loss": 0.08509012833237647,
      "grad_norm": 0.38220295310020447,
      "h_stock_valid_norm": 97.55936889648437,
      "invalid_tail_samples": 23.6,
      "learning_rate": 5.926666666666667e-05,
      "loss": 0.1042,
      "positive_teacher_pairs": 26894.0,
      "query_h_norm_ratio": 0.4637837499380112,
      "relation_loss": 0.019156492315232754,
      "selected_pairs": 26894.0,
      "step": 890,
      "student_distance_mean": 32.72625675201416,
      "teacher_distance_mean": 0.8509200572967529,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.019156492315232754,
      "weighted_relation_loss": 0.019156492315232754
    },
    {
      "action_query_batch_std": 0.5047825336456299,
      "action_query_norm": 45.24475250244141,
      "action_relation_loss": 0.01915734149515629,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.8419083255378859,
      "flow_matching_loss": 0.08600143194198609,
      "grad_norm": 0.326912522315979,
      "h_stock_valid_norm": 97.26919860839844,
      "invalid_tail_samples": 21.9,
      "learning_rate": 5.9933333333333345e-05,
      "loss": 0.1052,
      "positive_teacher_pairs": 27289.2,
      "query_h_norm_ratio": 0.46515496969223025,
      "relation_loss": 0.01915734149515629,
      "selected_pairs": 27289.2,
      "step": 900,
      "student_distance_mean": 32.09141731262207,
      "teacher_distance_mean": 0.8565319061279297,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.01915734149515629,
      "weighted_relation_loss": 0.01915734149515629
    },
    {
      "action_query_batch_std": 0.5293995082378388,
      "action_query_norm": 45.245236206054685,
      "action_relation_loss": 0.018593619205057622,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.8512628624883068,
      "flow_matching_loss": 0.08443762212991715,
      "grad_norm": 0.2413804531097412,
      "h_stock_valid_norm": 97.21269912719727,
      "invalid_tail_samples": 23.6,
      "learning_rate": 6.06e-05,
      "loss": 0.103,
      "positive_teacher_pairs": 26898.3,
      "query_h_norm_ratio": 0.46542823910713194,
      "relation_loss": 0.018593619205057622,
      "selected_pairs": 26898.3,
      "step": 910,
      "student_distance_mean": 33.5850772857666,
      "teacher_distance_mean": 0.850821852684021,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.018593619205057622,
      "weighted_relation_loss": 0.018593619205057622
    },
    {
      "action_query_batch_std": 0.5351982831954956,
      "action_query_norm": 45.24458961486816,
      "action_relation_loss": 0.01838629860430956,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.8606173994387278,
      "flow_matching_loss": 0.08576432541012764,
      "grad_norm": 0.37080246210098267,
      "h_stock_valid_norm": 96.88475494384765,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.126666666666667e-05,
      "loss": 0.1042,
      "positive_teacher_pairs": 26660.5,
      "query_h_norm_ratio": 0.466995096206665,
      "relation_loss": 0.01838629860430956,
      "selected_pairs": 26660.5,
      "step": 920,
      "student_distance_mean": 33.72946128845215,
      "teacher_distance_mean": 0.8547115445137023,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.01838629860430956,
      "weighted_relation_loss": 0.01838629860430956
    },
    {
      "action_query_batch_std": 0.5499054431915283,
      "action_query_norm": 45.24406547546387,
      "action_relation_loss": 0.018618138134479524,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.8699719363891487,
      "flow_matching_loss": 0.08412151485681534,
      "grad_norm": 0.29384690523147583,
      "h_stock_valid_norm": 96.79086837768554,
      "invalid_tail_samples": 22.8,
      "learning_rate": 6.193333333333333e-05,
      "loss": 0.1027,
      "positive_teacher_pairs": 27086.8,
      "query_h_norm_ratio": 0.46744383275508883,
      "relation_loss": 0.018618138134479524,
      "selected_pairs": 27086.8,
      "step": 930,
      "student_distance_mean": 34.91274299621582,
      "teacher_distance_mean": 0.8500035166740417,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.018618138134479524,
      "weighted_relation_loss": 0.018618138134479524
    },
    {
      "action_query_batch_std": 0.5483217835426331,
      "action_query_norm": 45.24330101013184,
      "action_relation_loss": 0.018463103845715523,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.8793264733395697,
      "flow_matching_loss": 0.08421370089054107,
      "grad_norm": 0.33016321063041687,
      "h_stock_valid_norm": 96.5246597290039,
      "invalid_tail_samples": 20.5,
      "learning_rate": 6.26e-05,
      "loss": 0.1027,
      "positive_teacher_pairs": 27623.5,
      "query_h_norm_ratio": 0.4687274843454361,
      "relation_loss": 0.018463103845715523,
      "selected_pairs": 27623.5,
      "step": 940,
      "student_distance_mean": 34.71096839904785,
      "teacher_distance_mean": 0.8509825527667999,
      "valid_action_samples": 235.5,
      "weighted_auxiliary_loss": 0.018463103845715523,
      "weighted_relation_loss": 0.018463103845715523
    },
    {
      "action_query_batch_std": 0.5631768822669982,
      "action_query_norm": 45.242922973632815,
      "action_relation_loss": 0.018356503546237947,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.8886810102899907,
      "flow_matching_loss": 0.08294394388794898,
      "grad_norm": 0.2239512801170349,
      "h_stock_valid_norm": 96.36312408447266,
      "invalid_tail_samples": 22.0,
      "learning_rate": 6.326666666666667e-05,
      "loss": 0.1013,
      "positive_teacher_pairs": 27269.9,
      "query_h_norm_ratio": 0.469508096575737,
      "relation_loss": 0.018356503546237947,
      "selected_pairs": 27269.9,
      "step": 950,
      "student_distance_mean": 35.72575798034668,
      "teacher_distance_mean": 0.8667896330356598,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.018356503546237947,
      "weighted_relation_loss": 0.018356503546237947
    },
    {
      "action_query_batch_std": 0.5707748711109162,
      "action_query_norm": 45.24287338256836,
      "action_relation_loss": 0.017570472788065672,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.8980355472404116,
      "flow_matching_loss": 0.08249980732798576,
      "grad_norm": 0.34187665581703186,
      "h_stock_valid_norm": 96.71437149047851,
      "invalid_tail_samples": 25.4,
      "learning_rate": 6.393333333333333e-05,
      "loss": 0.1001,
      "positive_teacher_pairs": 26485.8,
      "query_h_norm_ratio": 0.4678002417087555,
      "relation_loss": 0.017570472788065672,
      "selected_pairs": 26485.8,
      "step": 960,
      "student_distance_mean": 36.11970252990723,
      "teacher_distance_mean": 0.860389792919159,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.017570472788065672,
      "weighted_relation_loss": 0.017570472788065672
    },
    {
      "action_query_batch_std": 0.5436256647109985,
      "action_query_norm": 45.24022789001465,
      "action_relation_loss": 0.018775215744972228,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.9073900841908326,
      "flow_matching_loss": 0.08210640251636506,
      "grad_norm": 0.3767443895339966,
      "h_stock_valid_norm": 97.37416305541993,
      "invalid_tail_samples": 25.2,
      "learning_rate": 6.460000000000001e-05,
      "loss": 0.1009,
      "positive_teacher_pairs": 26531.5,
      "query_h_norm_ratio": 0.4646047502756119,
      "relation_loss": 0.018775215744972228,
      "selected_pairs": 26531.5,
      "step": 970,
      "student_distance_mean": 34.589809799194335,
      "teacher_distance_mean": 0.8525396347045898,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.018775215744972228,
      "weighted_relation_loss": 0.018775215744972228
    },
    {
      "action_query_batch_std": 0.5515943229198456,
      "action_query_norm": 45.2394660949707,
      "action_relation_loss": 0.017462344933301212,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.9167446211412535,
      "flow_matching_loss": 0.08759313002228737,
      "grad_norm": 0.5660896897315979,
      "h_stock_valid_norm": 97.19687194824219,
      "invalid_tail_samples": 23.5,
      "learning_rate": 6.526666666666666e-05,
      "loss": 0.1051,
      "positive_teacher_pairs": 26920.1,
      "query_h_norm_ratio": 0.46544458270072936,
      "relation_loss": 0.017462344933301212,
      "selected_pairs": 26920.1,
      "step": 980,
      "student_distance_mean": 34.99107780456543,
      "teacher_distance_mean": 0.8592372298240661,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.017462344933301212,
      "weighted_relation_loss": 0.017462344933301212
    },
    {
      "action_query_batch_std": 0.5359165608882904,
      "action_query_norm": 45.23862495422363,
      "action_relation_loss": 0.017461901344358922,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.9260991580916744,
      "flow_matching_loss": 0.0836106777191162,
      "grad_norm": 0.32419753074645996,
      "h_stock_valid_norm": 96.2045799255371,
      "invalid_tail_samples": 23.1,
      "learning_rate": 6.593333333333334e-05,
      "loss": 0.1011,
      "positive_teacher_pairs": 27012.6,
      "query_h_norm_ratio": 0.470239469408989,
      "relation_loss": 0.017461901344358922,
      "selected_pairs": 27012.6,
      "step": 990,
      "student_distance_mean": 34.041103744506835,
      "teacher_distance_mean": 0.8498923063278199,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.017461901344358922,
      "weighted_relation_loss": 0.017461901344358922
    },
    {
      "action_query_batch_std": 0.5447602570056915,
      "action_query_norm": 45.23837890625,
      "action_relation_loss": 0.018352823704481123,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.9354536950420954,
      "flow_matching_loss": 0.08796347975730896,
      "grad_norm": 0.3825078010559082,
      "h_stock_valid_norm": 96.08151779174804,
      "invalid_tail_samples": 24.5,
      "learning_rate": 6.66e-05,
      "loss": 0.1063,
      "positive_teacher_pairs": 26686.1,
      "query_h_norm_ratio": 0.47083637714385984,
      "relation_loss": 0.018352823704481123,
      "selected_pairs": 26686.1,
      "step": 1000,
      "student_distance_mean": 34.519632720947264,
      "teacher_distance_mean": 0.8574318587779999,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.018352823704481123,
      "weighted_relation_loss": 0.018352823704481123
    },
    {
      "action_query_batch_std": 0.5602340459823608,
      "action_query_norm": 45.23752861022949,
      "action_relation_loss": 0.018089766427874566,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.9448082319925164,
      "flow_matching_loss": 0.085418551415205,
      "grad_norm": 0.43842703104019165,
      "h_stock_valid_norm": 97.04656600952148,
      "invalid_tail_samples": 24.1,
      "learning_rate": 6.726666666666667e-05,
      "loss": 0.1035,
      "positive_teacher_pairs": 26782.0,
      "query_h_norm_ratio": 0.46614758372306825,
      "relation_loss": 0.018089766427874566,
      "selected_pairs": 26782.0,
      "step": 1010,
      "student_distance_mean": 35.741757202148435,
      "teacher_distance_mean": 0.8578993022441864,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.018089766427874566,
      "weighted_relation_loss": 0.018089766427874566
    },
    {
      "action_query_batch_std": 0.5491454422473907,
      "action_query_norm": 45.235890579223636,
      "action_relation_loss": 0.017786141857504843,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.9541627689429373,
      "flow_matching_loss": 0.08244817927479745,
      "grad_norm": 0.4395314157009125,
      "h_stock_valid_norm": 97.39677429199219,
      "invalid_tail_samples": 22.5,
      "learning_rate": 6.793333333333334e-05,
      "loss": 0.1002,
      "positive_teacher_pairs": 27155.1,
      "query_h_norm_ratio": 0.46445406377315523,
      "relation_loss": 0.017786141857504843,
      "selected_pairs": 27155.1,
      "step": 1020,
      "student_distance_mean": 35.102339553833005,
      "teacher_distance_mean": 0.849157053232193,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.017786141857504843,
      "weighted_relation_loss": 0.017786141857504843
    },
    {
      "action_query_batch_std": 0.5426090776920318,
      "action_query_norm": 45.23607559204102,
      "action_relation_loss": 0.018452891148626804,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.9635173058933583,
      "flow_matching_loss": 0.08746427968144417,
      "grad_norm": 0.5003315806388855,
      "h_stock_valid_norm": 98.06292343139648,
      "invalid_tail_samples": 24.4,
      "learning_rate": 6.860000000000001e-05,
      "loss": 0.1059,
      "positive_teacher_pairs": 26726.0,
      "query_h_norm_ratio": 0.46129874885082245,
      "relation_loss": 0.018452891148626804,
      "selected_pairs": 26726.0,
      "step": 1030,
      "student_distance_mean": 34.4863941192627,
      "teacher_distance_mean": 0.8524686574935914,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.018452891148626804,
      "weighted_relation_loss": 0.018452891148626804
    },
    {
      "action_query_batch_std": 0.5475977063179016,
      "action_query_norm": 45.236049270629884,
      "action_relation_loss": 0.01809035111218691,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.9728718428437793,
      "flow_matching_loss": 0.08416153639554977,
      "grad_norm": 0.28561025857925415,
      "h_stock_valid_norm": 98.34277572631837,
      "invalid_tail_samples": 22.3,
      "learning_rate": 6.926666666666667e-05,
      "loss": 0.1023,
      "positive_teacher_pairs": 27193.9,
      "query_h_norm_ratio": 0.4599853754043579,
      "relation_loss": 0.01809035111218691,
      "selected_pairs": 27193.9,
      "step": 1040,
      "student_distance_mean": 34.8912712097168,
      "teacher_distance_mean": 0.8559509694576264,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.01809035111218691,
      "weighted_relation_loss": 0.01809035111218691
    },
    {
      "action_query_batch_std": 0.5506422102451325,
      "action_query_norm": 45.23401641845703,
      "action_relation_loss": 0.01816392783075571,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.9822263797942001,
      "flow_matching_loss": 0.08375498652458191,
      "grad_norm": 0.2605457007884979,
      "h_stock_valid_norm": 97.38274154663085,
      "invalid_tail_samples": 27.5,
      "learning_rate": 6.993333333333334e-05,
      "loss": 0.1019,
      "positive_teacher_pairs": 26000.8,
      "query_h_norm_ratio": 0.4645046412944794,
      "relation_loss": 0.01816392783075571,
      "selected_pairs": 26000.8,
      "step": 1050,
      "student_distance_mean": 35.030835723876955,
      "teacher_distance_mean": 0.8739325344562531,
      "valid_action_samples": 228.5,
      "weighted_auxiliary_loss": 0.01816392783075571,
      "weighted_relation_loss": 0.01816392783075571
    },
    {
      "action_query_batch_std": 0.5747785210609436,
      "action_query_norm": 45.23383140563965,
      "action_relation_loss": 0.01863120459020138,
      "auxiliary_loss_weight": 1.0,
      "epoch": 0.9915809167446211,
      "flow_matching_loss": 0.08298583626747132,
      "grad_norm": 0.2368013709783554,
      "h_stock_valid_norm": 96.8809700012207,
      "invalid_tail_samples": 20.5,
      "learning_rate": 7.06e-05,
      "loss": 0.1016,
      "positive_teacher_pairs": 27615.0,
      "query_h_norm_ratio": 0.4669073671102524,
      "relation_loss": 0.01863120459020138,
      "selected_pairs": 27615.0,
      "step": 1060,
      "student_distance_mean": 36.612939453125,
      "teacher_distance_mean": 0.8545524895191192,
      "valid_action_samples": 235.5,
      "weighted_auxiliary_loss": 0.01863120459020138,
      "weighted_relation_loss": 0.01863120459020138
    },
    {
      "action_query_batch_std": 0.5486873805522918,
      "action_query_norm": 45.23147773742676,
      "action_relation_loss": 0.017252318002283572,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.000935453695042,
      "flow_matching_loss": 0.08243878558278084,
      "grad_norm": 0.35344088077545166,
      "h_stock_valid_norm": 98.03097076416016,
      "invalid_tail_samples": 22.1,
      "learning_rate": 7.126666666666667e-05,
      "loss": 0.0997,
      "positive_teacher_pairs": 24331.3,
      "query_h_norm_ratio": 0.46140762865543367,
      "relation_loss": 0.017252318002283572,
      "selected_pairs": 24331.3,
      "step": 1070,
      "student_distance_mean": 34.97575149536133,
      "teacher_distance_mean": 0.8568814516067504,
      "valid_action_samples": 214.0,
      "weighted_auxiliary_loss": 0.017252318002283572,
      "weighted_relation_loss": 0.017252318002283572
    },
    {
      "action_query_batch_std": 0.5398070216178894,
      "action_query_norm": 45.230132293701175,
      "action_relation_loss": 0.017837955616414547,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.010289990645463,
      "flow_matching_loss": 0.08609309643507004,
      "grad_norm": 0.3351980745792389,
      "h_stock_valid_norm": 98.77430648803711,
      "invalid_tail_samples": 23.8,
      "learning_rate": 7.193333333333334e-05,
      "loss": 0.1039,
      "positive_teacher_pairs": 26854.3,
      "query_h_norm_ratio": 0.4579167068004608,
      "relation_loss": 0.017837955616414547,
      "selected_pairs": 26854.3,
      "step": 1080,
      "student_distance_mean": 34.4959171295166,
      "teacher_distance_mean": 0.8550265669822693,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.017837955616414547,
      "weighted_relation_loss": 0.017837955616414547
    },
    {
      "action_query_batch_std": 0.5588713884353638,
      "action_query_norm": 45.23105735778809,
      "action_relation_loss": 0.01795183476060629,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.019644527595884,
      "flow_matching_loss": 0.08106319978833199,
      "grad_norm": 0.2530900835990906,
      "h_stock_valid_norm": 98.42629318237304,
      "invalid_tail_samples": 24.5,
      "learning_rate": 7.26e-05,
      "loss": 0.099,
      "positive_teacher_pairs": 26685.2,
      "query_h_norm_ratio": 0.459546759724617,
      "relation_loss": 0.01795183476060629,
      "selected_pairs": 26685.2,
      "step": 1090,
      "student_distance_mean": 35.4817569732666,
      "teacher_distance_mean": 0.855253803730011,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.01795183476060629,
      "weighted_relation_loss": 0.01795183476060629
    },
    {
      "action_query_batch_std": 0.5670303404331207,
      "action_query_norm": 45.229455184936526,
      "action_relation_loss": 0.01802831459790468,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.028999064546305,
      "flow_matching_loss": 0.08848854079842568,
      "grad_norm": 0.3171290159225464,
      "h_stock_valid_norm": 97.75020370483398,
      "invalid_tail_samples": 21.9,
      "learning_rate": 7.326666666666667e-05,
      "loss": 0.1065,
      "positive_teacher_pairs": 27292.0,
      "query_h_norm_ratio": 0.46270873844623567,
      "relation_loss": 0.01802831459790468,
      "selected_pairs": 27292.0,
      "step": 1100,
      "student_distance_mean": 36.150350189208986,
      "teacher_distance_mean": 0.8531495630741119,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.01802831459790468,
      "weighted_relation_loss": 0.01802831459790468
    },
    {
      "action_query_batch_std": 0.5754216074943542,
      "action_query_norm": 45.22951126098633,
      "action_relation_loss": 0.017652670107781886,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.0383536014967258,
      "flow_matching_loss": 0.08414214923977852,
      "grad_norm": 0.33088335394859314,
      "h_stock_valid_norm": 96.89197235107422,
      "invalid_tail_samples": 25.6,
      "learning_rate": 7.393333333333333e-05,
      "loss": 0.1018,
      "positive_teacher_pairs": 26446.3,
      "query_h_norm_ratio": 0.46680656373500823,
      "relation_loss": 0.017652670107781886,
      "selected_pairs": 26446.3,
      "step": 1110,
      "student_distance_mean": 36.59592552185059,
      "teacher_distance_mean": 0.8546444058418274,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.017652670107781886,
      "weighted_relation_loss": 0.017652670107781886
    },
    {
      "action_query_batch_std": 0.5509312450885773,
      "action_query_norm": 45.228240203857425,
      "action_relation_loss": 0.017308297380805016,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.047708138447147,
      "flow_matching_loss": 0.08569902777671815,
      "grad_norm": 0.23973554372787476,
      "h_stock_valid_norm": 96.44084167480469,
      "invalid_tail_samples": 21.1,
      "learning_rate": 7.46e-05,
      "loss": 0.103,
      "positive_teacher_pairs": 27480.5,
      "query_h_norm_ratio": 0.46897694170475007,
      "relation_loss": 0.017308297380805016,
      "selected_pairs": 27480.5,
      "step": 1120,
      "student_distance_mean": 35.15576324462891,
      "teacher_distance_mean": 0.8579326033592224,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.017308297380805016,
      "weighted_relation_loss": 0.017308297380805016
    },
    {
      "action_query_batch_std": 0.5673594355583191,
      "action_query_norm": 45.2279712677002,
      "action_relation_loss": 0.017109397798776627,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.0570626753975678,
      "flow_matching_loss": 0.08263532891869545,
      "grad_norm": 0.23695158958435059,
      "h_stock_valid_norm": 95.81270141601563,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.526666666666668e-05,
      "loss": 0.0997,
      "positive_teacher_pairs": 27033.8,
      "query_h_norm_ratio": 0.4720481514930725,
      "relation_loss": 0.017109397798776627,
      "selected_pairs": 27033.8,
      "step": 1130,
      "student_distance_mean": 36.109226608276366,
      "teacher_distance_mean": 0.8693834841251373,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.017109397798776627,
      "weighted_relation_loss": 0.017109397798776627
    },
    {
      "action_query_batch_std": 0.567137885093689,
      "action_query_norm": 45.22538185119629,
      "action_relation_loss": 0.017629852611571552,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.0664172123479887,
      "flow_matching_loss": 0.08398201316595078,
      "grad_norm": 0.2882460951805115,
      "h_stock_valid_norm": 96.40371704101562,
      "invalid_tail_samples": 25.7,
      "learning_rate": 7.593333333333334e-05,
      "loss": 0.1016,
      "positive_teacher_pairs": 26409.6,
      "query_h_norm_ratio": 0.46913121938705443,
      "relation_loss": 0.017629852611571552,
      "selected_pairs": 26409.6,
      "step": 1140,
      "student_distance_mean": 36.127702713012695,
      "teacher_distance_mean": 0.8624014377593994,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.017629852611571552,
      "weighted_relation_loss": 0.017629852611571552
    },
    {
      "action_query_batch_std": 0.5646240055561066,
      "action_query_norm": 45.224547958374025,
      "action_relation_loss": 0.0177335062995553,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.0757717492984098,
      "flow_matching_loss": 0.08327563479542732,
      "grad_norm": 0.2960732579231262,
      "h_stock_valid_norm": 97.48940887451172,
      "invalid_tail_samples": 20.3,
      "learning_rate": 7.66e-05,
      "loss": 0.101,
      "positive_teacher_pairs": 27662.5,
      "query_h_norm_ratio": 0.4639056593179703,
      "relation_loss": 0.0177335062995553,
      "selected_pairs": 27662.5,
      "step": 1150,
      "student_distance_mean": 35.98193359375,
      "teacher_distance_mean": 0.8590244889259339,
      "valid_action_samples": 235.7,
      "weighted_auxiliary_loss": 0.0177335062995553,
      "weighted_relation_loss": 0.0177335062995553
    },
    {
      "action_query_batch_std": 0.5699526607990265,
      "action_query_norm": 45.223443603515626,
      "action_relation_loss": 0.017475288920104505,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.0851262862488307,
      "flow_matching_loss": 0.08312815055251122,
      "grad_norm": 0.3394313454627991,
      "h_stock_valid_norm": 98.4067596435547,
      "invalid_tail_samples": 24.8,
      "learning_rate": 7.726666666666667e-05,
      "loss": 0.1006,
      "positive_teacher_pairs": 26627.6,
      "query_h_norm_ratio": 0.45955862998962405,
      "relation_loss": 0.017475288920104505,
      "selected_pairs": 26627.6,
      "step": 1160,
      "student_distance_mean": 36.36160659790039,
      "teacher_distance_mean": 0.8621905088424683,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.017475288920104505,
      "weighted_relation_loss": 0.017475288920104505
    },
    {
      "action_query_batch_std": 0.5588680267333984,
      "action_query_norm": 45.22275886535645,
      "action_relation_loss": 0.01697111427783966,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.0944808231992516,
      "flow_matching_loss": 0.08159576579928399,
      "grad_norm": 0.24333156645298004,
      "h_stock_valid_norm": 98.10959854125977,
      "invalid_tail_samples": 23.5,
      "learning_rate": 7.793333333333333e-05,
      "loss": 0.0986,
      "positive_teacher_pairs": 26926.1,
      "query_h_norm_ratio": 0.46095053255558016,
      "relation_loss": 0.01697111427783966,
      "selected_pairs": 26926.1,
      "step": 1170,
      "student_distance_mean": 35.480277633666994,
      "teacher_distance_mean": 0.8496854722499847,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.01697111427783966,
      "weighted_relation_loss": 0.01697111427783966
    },
    {
      "action_query_batch_std": 0.548815792798996,
      "action_query_norm": 45.220176696777344,
      "action_relation_loss": 0.017148368805646897,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.1038353601496727,
      "flow_matching_loss": 0.08316718861460685,
      "grad_norm": 0.31218618154525757,
      "h_stock_valid_norm": 97.21047744750976,
      "invalid_tail_samples": 24.3,
      "learning_rate": 7.860000000000001e-05,
      "loss": 0.1003,
      "positive_teacher_pairs": 26729.7,
      "query_h_norm_ratio": 0.4651789337396622,
      "relation_loss": 0.017148368805646897,
      "selected_pairs": 26729.7,
      "step": 1180,
      "student_distance_mean": 35.00735702514648,
      "teacher_distance_mean": 0.8564692914485932,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.017148368805646897,
      "weighted_relation_loss": 0.017148368805646897
    },
    {
      "action_query_batch_std": 0.529685628414154,
      "action_query_norm": 45.21837387084961,
      "action_relation_loss": 0.01744063962250948,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.1131898971000935,
      "flow_matching_loss": 0.08388474732637405,
      "grad_norm": 0.19494327902793884,
      "h_stock_valid_norm": 97.11289443969727,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.926666666666666e-05,
      "loss": 0.1013,
      "positive_teacher_pairs": 27038.9,
      "query_h_norm_ratio": 0.4656314641237259,
      "relation_loss": 0.01744063962250948,
      "selected_pairs": 27038.9,
      "step": 1190,
      "student_distance_mean": 33.81189880371094,
      "teacher_distance_mean": 0.8424399316310882,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.01744063962250948,
      "weighted_relation_loss": 0.01744063962250948
    },
    {
      "action_query_batch_std": 0.551472932100296,
      "action_query_norm": 45.218854141235354,
      "action_relation_loss": 0.017375926300883294,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.1225444340505144,
      "flow_matching_loss": 0.08236829042434693,
      "grad_norm": 0.20861133933067322,
      "h_stock_valid_norm": 96.62366561889648,
      "invalid_tail_samples": 26.6,
      "learning_rate": 7.993333333333334e-05,
      "loss": 0.0997,
      "positive_teacher_pairs": 26204.3,
      "query_h_norm_ratio": 0.4679918110370636,
      "relation_loss": 0.017375926300883294,
      "selected_pairs": 26204.3,
      "step": 1200,
      "student_distance_mean": 35.059391021728516,
      "teacher_distance_mean": 0.8597922265529633,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.017375926300883294,
      "weighted_relation_loss": 0.017375926300883294
    },
    {
      "action_query_batch_std": 0.5603287100791932,
      "action_query_norm": 45.218078994750975,
      "action_relation_loss": 0.016867073345929386,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.1318989710009355,
      "flow_matching_loss": 0.08143971487879753,
      "grad_norm": 0.2900734543800354,
      "h_stock_valid_norm": 97.0798210144043,
      "invalid_tail_samples": 23.8,
      "learning_rate": 8.060000000000001e-05,
      "loss": 0.0983,
      "positive_teacher_pairs": 26853.9,
      "query_h_norm_ratio": 0.46579148471355436,
      "relation_loss": 0.016867073345929386,
      "selected_pairs": 26853.9,
      "step": 1210,
      "student_distance_mean": 35.4848876953125,
      "teacher_distance_mean": 0.8531014382839203,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.016867073345929386,
      "weighted_relation_loss": 0.016867073345929386
    },
    {
      "action_query_batch_std": 0.5578969955444336,
      "action_query_norm": 45.216874313354495,
      "action_relation_loss": 0.01781087750568986,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.1412535079513564,
      "flow_matching_loss": 0.0807200811803341,
      "grad_norm": 0.2425098866224289,
      "h_stock_valid_norm": 97.9141960144043,
      "invalid_tail_samples": 23.3,
      "learning_rate": 8.126666666666667e-05,
      "loss": 0.0985,
      "positive_teacher_pairs": 26970.1,
      "query_h_norm_ratio": 0.4618062645196915,
      "relation_loss": 0.01781087750568986,
      "selected_pairs": 26970.1,
      "step": 1220,
      "student_distance_mean": 35.63512344360352,
      "teacher_distance_mean": 0.8479903221130372,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.01781087750568986,
      "weighted_relation_loss": 0.01781087750568986
    },
    {
      "action_query_batch_std": 0.5544403612613678,
      "action_query_norm": 45.215857315063474,
      "action_relation_loss": 0.01852002739906311,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.1506080449017775,
      "flow_matching_loss": 0.08161494359374047,
      "grad_norm": 0.2728819251060486,
      "h_stock_valid_norm": 97.37521820068359,
      "invalid_tail_samples": 23.6,
      "learning_rate": 8.193333333333334e-05,
      "loss": 0.1001,
      "positive_teacher_pairs": 26903.1,
      "query_h_norm_ratio": 0.464349827170372,
      "relation_loss": 0.01852002739906311,
      "selected_pairs": 26903.1,
      "step": 1230,
      "student_distance_mean": 35.279395294189456,
      "teacher_distance_mean": 0.8530473589897156,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.01852002739906311,
      "weighted_relation_loss": 0.01852002739906311
    },
    {
      "action_query_batch_std": 0.5653324782848358,
      "action_query_norm": 45.21373977661133,
      "action_relation_loss": 0.016792394034564496,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.1599625818521984,
      "flow_matching_loss": 0.08382445126771927,
      "grad_norm": 0.32679280638694763,
      "h_stock_valid_norm": 97.41494140625,
      "invalid_tail_samples": 24.8,
      "learning_rate": 8.26e-05,
      "loss": 0.1006,
      "positive_teacher_pairs": 26621.1,
      "query_h_norm_ratio": 0.4641382068395615,
      "relation_loss": 0.016792394034564496,
      "selected_pairs": 26621.1,
      "step": 1240,
      "student_distance_mean": 35.873125076293945,
      "teacher_distance_mean": 0.8497156083583832,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.016792394034564496,
      "weighted_relation_loss": 0.016792394034564496
    },
    {
      "action_query_batch_std": 0.583589494228363,
      "action_query_norm": 45.21468620300293,
      "action_relation_loss": 0.017650193348526955,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.1693171188026192,
      "flow_matching_loss": 0.08199037387967109,
      "grad_norm": 0.31582823395729065,
      "h_stock_valid_norm": 97.14810180664062,
      "invalid_tail_samples": 23.1,
      "learning_rate": 8.326666666666667e-05,
      "loss": 0.0996,
      "positive_teacher_pairs": 27023.8,
      "query_h_norm_ratio": 0.46542623043060305,
      "relation_loss": 0.017650193348526955,
      "selected_pairs": 27023.8,
      "step": 1250,
      "student_distance_mean": 37.20137672424316,
      "teacher_distance_mean": 0.8551608204841614,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.017650193348526955,
      "weighted_relation_loss": 0.017650193348526955
    },
    {
      "action_query_batch_std": 0.5597941696643829,
      "action_query_norm": 45.21295166015625,
      "action_relation_loss": 0.017144352942705155,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.1786716557530403,
      "flow_matching_loss": 0.08531775400042534,
      "grad_norm": 0.2736450135707855,
      "h_stock_valid_norm": 96.75436630249024,
      "invalid_tail_samples": 23.0,
      "learning_rate": 8.393333333333335e-05,
      "loss": 0.1025,
      "positive_teacher_pairs": 27037.5,
      "query_h_norm_ratio": 0.4673002779483795,
      "relation_loss": 0.017144352942705155,
      "selected_pairs": 27037.5,
      "step": 1260,
      "student_distance_mean": 35.65769882202149,
      "teacher_distance_mean": 0.8523586928844452,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.017144352942705155,
      "weighted_relation_loss": 0.017144352942705155
    },
    {
      "action_query_batch_std": 0.5560854375362396,
      "action_query_norm": 45.2104736328125,
      "action_relation_loss": 0.01758377905935049,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.1880261927034612,
      "flow_matching_loss": 0.07994240075349808,
      "grad_norm": 0.3032025694847107,
      "h_stock_valid_norm": 97.12505340576172,
      "invalid_tail_samples": 22.9,
      "learning_rate": 8.46e-05,
      "loss": 0.0975,
      "positive_teacher_pairs": 27059.5,
      "query_h_norm_ratio": 0.46548978984355927,
      "relation_loss": 0.01758377905935049,
      "selected_pairs": 27059.5,
      "step": 1270,
      "student_distance_mean": 35.426184463500974,
      "teacher_distance_mean": 0.8493092238903046,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.01758377905935049,
      "weighted_relation_loss": 0.01758377905935049
    },
    {
      "action_query_batch_std": 0.5725816011428833,
      "action_query_norm": 45.21108436584473,
      "action_relation_loss": 0.018054373934864996,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.197380729653882,
      "flow_matching_loss": 0.0847734309732914,
      "grad_norm": 0.26315534114837646,
      "h_stock_valid_norm": 96.69620819091797,
      "invalid_tail_samples": 24.6,
      "learning_rate": 8.526666666666667e-05,
      "loss": 0.1028,
      "positive_teacher_pairs": 26659.5,
      "query_h_norm_ratio": 0.4675603061914444,
      "relation_loss": 0.018054373934864996,
      "selected_pairs": 26659.5,
      "step": 1280,
      "student_distance_mean": 36.772407913208006,
      "teacher_distance_mean": 0.8462124824523926,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.018054373934864996,
      "weighted_relation_loss": 0.018054373934864996
    },
    {
      "action_query_batch_std": 0.5434402525424957,
      "action_query_norm": 45.20773162841797,
      "action_relation_loss": 0.017121295630931854,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.2067352666043032,
      "flow_matching_loss": 0.08240850046277046,
      "grad_norm": 0.2462533861398697,
      "h_stock_valid_norm": 96.80019607543946,
      "invalid_tail_samples": 22.7,
      "learning_rate": 8.593333333333333e-05,
      "loss": 0.0995,
      "positive_teacher_pairs": 27108.8,
      "query_h_norm_ratio": 0.46702389121055604,
      "relation_loss": 0.017121295630931854,
      "selected_pairs": 27108.8,
      "step": 1290,
      "student_distance_mean": 34.510567092895506,
      "teacher_distance_mean": 0.8469069600105286,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.017121295630931854,
      "weighted_relation_loss": 0.017121295630931854
    },
    {
      "action_query_batch_std": 0.5519129693508148,
      "action_query_norm": 45.20758399963379,
      "action_relation_loss": 0.017652436904609204,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.216089803554724,
      "flow_matching_loss": 0.08006709888577461,
      "grad_norm": 0.2089613974094391,
      "h_stock_valid_norm": 96.69086608886718,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.66e-05,
      "loss": 0.0977,
      "positive_teacher_pairs": 26697.4,
      "query_h_norm_ratio": 0.46755197644233704,
      "relation_loss": 0.017652436904609204,
      "selected_pairs": 26697.4,
      "step": 1300,
      "student_distance_mean": 35.09556655883789,
      "teacher_distance_mean": 0.8607197284698487,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.017652436904609204,
      "weighted_relation_loss": 0.017652436904609204
    },
    {
      "action_query_batch_std": 0.5598015546798706,
      "action_query_norm": 45.2074031829834,
      "action_relation_loss": 0.017422140389680863,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.225444340505145,
      "flow_matching_loss": 0.08245339244604111,
      "grad_norm": 0.3255414366722107,
      "h_stock_valid_norm": 96.5659896850586,
      "invalid_tail_samples": 25.5,
      "learning_rate": 8.726666666666667e-05,
      "loss": 0.0999,
      "positive_teacher_pairs": 26462.9,
      "query_h_norm_ratio": 0.4681514471769333,
      "relation_loss": 0.017422140389680863,
      "selected_pairs": 26462.9,
      "step": 1310,
      "student_distance_mean": 35.90523414611816,
      "teacher_distance_mean": 0.8566758096218109,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.017422140389680863,
      "weighted_relation_loss": 0.017422140389680863
    },
    {
      "action_query_batch_std": 0.5674306631088257,
      "action_query_norm": 45.2063533782959,
      "action_relation_loss": 0.017181994300335644,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.234798877455566,
      "flow_matching_loss": 0.08194156438112259,
      "grad_norm": 0.30880066752433777,
      "h_stock_valid_norm": 96.22456893920898,
      "invalid_tail_samples": 22.6,
      "learning_rate": 8.793333333333333e-05,
      "loss": 0.0991,
      "positive_teacher_pairs": 27136.2,
      "query_h_norm_ratio": 0.46980506777763364,
      "relation_loss": 0.017181994300335644,
      "selected_pairs": 27136.2,
      "step": 1320,
      "student_distance_mean": 36.300428009033205,
      "teacher_distance_mean": 0.8545205950736999,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.017181994300335644,
      "weighted_relation_loss": 0.017181994300335644
    },
    {
      "action_query_batch_std": 0.5842749238014221,
      "action_query_norm": 45.207257461547854,
      "action_relation_loss": 0.017529410682618617,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.244153414405987,
      "flow_matching_loss": 0.08131160959601402,
      "grad_norm": 0.27261579036712646,
      "h_stock_valid_norm": 96.0102424621582,
      "invalid_tail_samples": 23.3,
      "learning_rate": 8.86e-05,
      "loss": 0.0988,
      "positive_teacher_pairs": 26976.3,
      "query_h_norm_ratio": 0.4708672404289246,
      "relation_loss": 0.017529410682618617,
      "selected_pairs": 26976.3,
      "step": 1330,
      "student_distance_mean": 37.151231384277345,
      "teacher_distance_mean": 0.8633258521556855,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.017529410682618617,
      "weighted_relation_loss": 0.017529410682618617
    },
    {
      "action_query_batch_std": 0.5649820268154144,
      "action_query_norm": 45.20360794067383,
      "action_relation_loss": 0.0176820021122694,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.2535079513564078,
      "flow_matching_loss": 0.0797963209450245,
      "grad_norm": 0.3587123155593872,
      "h_stock_valid_norm": 96.29706497192383,
      "invalid_tail_samples": 20.6,
      "learning_rate": 8.926666666666668e-05,
      "loss": 0.0975,
      "positive_teacher_pairs": 27592.7,
      "query_h_norm_ratio": 0.4694219946861267,
      "relation_loss": 0.0176820021122694,
      "selected_pairs": 27592.7,
      "step": 1340,
      "student_distance_mean": 36.01246452331543,
      "teacher_distance_mean": 0.8554794073104859,
      "valid_action_samples": 235.4,
      "weighted_auxiliary_loss": 0.0176820021122694,
      "weighted_relation_loss": 0.0176820021122694
    },
    {
      "action_query_batch_std": 0.5689064979553222,
      "action_query_norm": 45.20079154968262,
      "action_relation_loss": 0.01746842432767153,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.262862488306829,
      "flow_matching_loss": 0.07670473977923394,
      "grad_norm": 0.30036330223083496,
      "h_stock_valid_norm": 97.81262130737305,
      "invalid_tail_samples": 23.3,
      "learning_rate": 8.993333333333334e-05,
      "loss": 0.0942,
      "positive_teacher_pairs": 26967.6,
      "query_h_norm_ratio": 0.462125301361084,
      "relation_loss": 0.01746842432767153,
      "selected_pairs": 26967.6,
      "step": 1350,
      "student_distance_mean": 36.252515411376955,
      "teacher_distance_mean": 0.8636923372745514,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.01746842432767153,
      "weighted_relation_loss": 0.01746842432767153
    },
    {
      "action_query_batch_std": 0.5775859236717225,
      "action_query_norm": 45.20041046142578,
      "action_relation_loss": 0.016504306811839342,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.2722170252572498,
      "flow_matching_loss": 0.07922788709402084,
      "grad_norm": 0.3144865036010742,
      "h_stock_valid_norm": 99.38521194458008,
      "invalid_tail_samples": 25.0,
      "learning_rate": 9.06e-05,
      "loss": 0.0957,
      "positive_teacher_pairs": 26574.5,
      "query_h_norm_ratio": 0.45481242537498473,
      "relation_loss": 0.016504306811839342,
      "selected_pairs": 26574.5,
      "step": 1360,
      "student_distance_mean": 36.5458251953125,
      "teacher_distance_mean": 0.8618788838386535,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.016504306811839342,
      "weighted_relation_loss": 0.016504306811839342
    },
    {
      "action_query_batch_std": 0.5728660345077514,
      "action_query_norm": 45.197824478149414,
      "action_relation_loss": 0.017426449246704577,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.2815715622076707,
      "flow_matching_loss": 0.0789901740849018,
      "grad_norm": 0.2646861672401428,
      "h_stock_valid_norm": 99.7725730895996,
      "invalid_tail_samples": 25.3,
      "learning_rate": 9.126666666666667e-05,
      "loss": 0.0964,
      "positive_teacher_pairs": 26501.1,
      "query_h_norm_ratio": 0.45301336944103243,
      "relation_loss": 0.017426449246704577,
      "selected_pairs": 26501.1,
      "step": 1370,
      "student_distance_mean": 36.48808822631836,
      "teacher_distance_mean": 0.8600843906402588,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.017426449246704577,
      "weighted_relation_loss": 0.017426449246704577
    },
    {
      "action_query_batch_std": 0.5690068364143371,
      "action_query_norm": 45.196298599243164,
      "action_relation_loss": 0.016494743712246417,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.2909260991580918,
      "flow_matching_loss": 0.08260769695043564,
      "grad_norm": 0.32673531770706177,
      "h_stock_valid_norm": 99.23834533691407,
      "invalid_tail_samples": 22.0,
      "learning_rate": 9.193333333333334e-05,
      "loss": 0.0991,
      "positive_teacher_pairs": 27263.0,
      "query_h_norm_ratio": 0.4554353654384613,
      "relation_loss": 0.016494743712246417,
      "selected_pairs": 27263.0,
      "step": 1380,
      "student_distance_mean": 36.42636909484863,
      "teacher_distance_mean": 0.8605721712112426,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.016494743712246417,
      "weighted_relation_loss": 0.016494743712246417
    },
    {
      "action_query_batch_std": 0.5590060234069825,
      "action_query_norm": 45.19435920715332,
      "action_relation_loss": 0.018121513724327087,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.3002806361085126,
      "flow_matching_loss": 0.08414089977741242,
      "grad_norm": 0.3262665271759033,
      "h_stock_valid_norm": 98.0663848876953,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.260000000000001e-05,
      "loss": 0.1023,
      "positive_teacher_pairs": 26693.4,
      "query_h_norm_ratio": 0.4608682572841644,
      "relation_loss": 0.018121513724327087,
      "selected_pairs": 26693.4,
      "step": 1390,
      "student_distance_mean": 35.66914443969726,
      "teacher_distance_mean": 0.8571523606777192,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.018121513724327087,
      "weighted_relation_loss": 0.018121513724327087
    },
    {
      "action_query_batch_std": 0.5435879230499268,
      "action_query_norm": 45.19099044799805,
      "action_relation_loss": 0.0169982329942286,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.3096351730589335,
      "flow_matching_loss": 0.08369538635015487,
      "grad_norm": 0.344871461391449,
      "h_stock_valid_norm": 97.61527557373047,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.326666666666667e-05,
      "loss": 0.1007,
      "positive_teacher_pairs": 26919.3,
      "query_h_norm_ratio": 0.4629538178443909,
      "relation_loss": 0.0169982329942286,
      "selected_pairs": 26919.3,
      "step": 1400,
      "student_distance_mean": 34.698768615722656,
      "teacher_distance_mean": 0.8583309769630432,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0169982329942286,
      "weighted_relation_loss": 0.0169982329942286
    },
    {
      "action_query_batch_std": 0.556494414806366,
      "action_query_norm": 45.1918285369873,
      "action_relation_loss": 0.01728458721190691,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.3189897100093546,
      "flow_matching_loss": 0.08018163740634918,
      "grad_norm": 0.3020755648612976,
      "h_stock_valid_norm": 97.62509613037109,
      "invalid_tail_samples": 21.7,
      "learning_rate": 9.393333333333334e-05,
      "loss": 0.0975,
      "positive_teacher_pairs": 27339.2,
      "query_h_norm_ratio": 0.46291950941085813,
      "relation_loss": 0.01728458721190691,
      "selected_pairs": 27339.2,
      "step": 1410,
      "student_distance_mean": 35.48440017700195,
      "teacher_distance_mean": 0.855209469795227,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.01728458721190691,
      "weighted_relation_loss": 0.01728458721190691
    },
    {
      "action_query_batch_std": 0.5581904232501984,
      "action_query_norm": 45.19027900695801,
      "action_relation_loss": 0.017015091236680747,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.3283442469597755,
      "flow_matching_loss": 0.0813666746020317,
      "grad_norm": 0.2759583294391632,
      "h_stock_valid_norm": 97.90992431640625,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.46e-05,
      "loss": 0.0984,
      "positive_teacher_pairs": 26665.1,
      "query_h_norm_ratio": 0.4615525841712952,
      "relation_loss": 0.017015091236680747,
      "selected_pairs": 26665.1,
      "step": 1420,
      "student_distance_mean": 35.60771484375,
      "teacher_distance_mean": 0.8631888568401337,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.017015091236680747,
      "weighted_relation_loss": 0.017015091236680747
    },
    {
      "action_query_batch_std": 0.5633740603923798,
      "action_query_norm": 45.19057502746582,
      "action_relation_loss": 0.016135538183152675,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.3376987839101964,
      "flow_matching_loss": 0.08009049966931343,
      "grad_norm": 0.2671486437320709,
      "h_stock_valid_norm": 98.5038688659668,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.526666666666667e-05,
      "loss": 0.0962,
      "positive_teacher_pairs": 26660.6,
      "query_h_norm_ratio": 0.45878552496433256,
      "relation_loss": 0.016135538183152675,
      "selected_pairs": 26660.6,
      "step": 1430,
      "student_distance_mean": 35.88565559387207,
      "teacher_distance_mean": 0.8511648654937745,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.016135538183152675,
      "weighted_relation_loss": 0.016135538183152675
    },
    {
      "action_query_batch_std": 0.566498875617981,
      "action_query_norm": 45.188943481445314,
      "action_relation_loss": 0.016418681386858223,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.3470533208606175,
      "flow_matching_loss": 0.08239344954490661,
      "grad_norm": 0.3200434148311615,
      "h_stock_valid_norm": 99.0890495300293,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.593333333333334e-05,
      "loss": 0.0988,
      "positive_teacher_pairs": 26951.2,
      "query_h_norm_ratio": 0.4560456037521362,
      "relation_loss": 0.016418681386858223,
      "selected_pairs": 26951.2,
      "step": 1440,
      "student_distance_mean": 35.98691596984863,
      "teacher_distance_mean": 0.864771842956543,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.016418681386858223,
      "weighted_relation_loss": 0.016418681386858223
    },
    {
      "action_query_batch_std": 0.5599715292453766,
      "action_query_norm": 45.187008666992185,
      "action_relation_loss": 0.015407949313521385,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.3564078578110383,
      "flow_matching_loss": 0.07925329953432084,
      "grad_norm": 0.33127328753471375,
      "h_stock_valid_norm": 98.2240379333496,
      "invalid_tail_samples": 24.4,
      "learning_rate": 9.66e-05,
      "loss": 0.0947,
      "positive_teacher_pairs": 26723.7,
      "query_h_norm_ratio": 0.46004271507263184,
      "relation_loss": 0.015407949313521385,
      "selected_pairs": 26723.7,
      "step": 1450,
      "student_distance_mean": 35.77588005065918,
      "teacher_distance_mean": 0.8553390979766846,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.015407949313521385,
      "weighted_relation_loss": 0.015407949313521385
    },
    {
      "action_query_batch_std": 0.552730655670166,
      "action_query_norm": 45.183052444458006,
      "action_relation_loss": 0.016694404929876328,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.3657623947614592,
      "flow_matching_loss": 0.08099257200956345,
      "grad_norm": 0.309707373380661,
      "h_stock_valid_norm": 97.15166244506835,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.726666666666667e-05,
      "loss": 0.0977,
      "positive_teacher_pairs": 27150.1,
      "query_h_norm_ratio": 0.4650887370109558,
      "relation_loss": 0.016694404929876328,
      "selected_pairs": 27150.1,
      "step": 1460,
      "student_distance_mean": 35.225537872314455,
      "teacher_distance_mean": 0.8633797943592072,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.016694404929876328,
      "weighted_relation_loss": 0.016694404929876328
    },
    {
      "action_query_batch_std": 0.5542185187339783,
      "action_query_norm": 45.18160057067871,
      "action_relation_loss": 0.015866716112941503,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.3751169317118803,
      "flow_matching_loss": 0.07579405158758164,
      "grad_norm": 0.3150475025177002,
      "h_stock_valid_norm": 96.32727737426758,
      "invalid_tail_samples": 22.7,
      "learning_rate": 9.793333333333333e-05,
      "loss": 0.0917,
      "positive_teacher_pairs": 27109.4,
      "query_h_norm_ratio": 0.46904371976852416,
      "relation_loss": 0.015866716112941503,
      "selected_pairs": 27109.4,
      "step": 1470,
      "student_distance_mean": 35.442936706542966,
      "teacher_distance_mean": 0.8604467213153839,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.015866716112941503,
      "weighted_relation_loss": 0.015866716112941503
    },
    {
      "action_query_batch_std": 0.560349452495575,
      "action_query_norm": 45.18012008666992,
      "action_relation_loss": 0.01674315966665745,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.3844714686623012,
      "flow_matching_loss": 0.07870883867144585,
      "grad_norm": 0.2679056227207184,
      "h_stock_valid_norm": 96.59628067016601,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.86e-05,
      "loss": 0.0955,
      "positive_teacher_pairs": 26978.5,
      "query_h_norm_ratio": 0.46772499978542326,
      "relation_loss": 0.01674315966665745,
      "selected_pairs": 26978.5,
      "step": 1480,
      "student_distance_mean": 35.758820343017575,
      "teacher_distance_mean": 0.855737155675888,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.01674315966665745,
      "weighted_relation_loss": 0.01674315966665745
    },
    {
      "action_query_batch_std": 0.561488687992096,
      "action_query_norm": 45.18013458251953,
      "action_relation_loss": 0.016503927111625672,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.393826005612722,
      "flow_matching_loss": 0.08065521419048309,
      "grad_norm": 0.3699997365474701,
      "h_stock_valid_norm": 97.08663787841797,
      "invalid_tail_samples": 24.4,
      "learning_rate": 9.926666666666668e-05,
      "loss": 0.0972,
      "positive_teacher_pairs": 26717.2,
      "query_h_norm_ratio": 0.46536058485507964,
      "relation_loss": 0.016503927111625672,
      "selected_pairs": 26717.2,
      "step": 1490,
      "student_distance_mean": 35.76416778564453,
      "teacher_distance_mean": 0.8531434416770936,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.016503927111625672,
      "weighted_relation_loss": 0.016503927111625672
    },
    {
      "action_query_batch_std": 0.5403245151042938,
      "action_query_norm": 45.175804138183594,
      "action_relation_loss": 0.016957656107842924,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.4031805425631432,
      "flow_matching_loss": 0.07695383504033089,
      "grad_norm": 0.29689309000968933,
      "h_stock_valid_norm": 97.04726486206054,
      "invalid_tail_samples": 25.4,
      "learning_rate": 9.993333333333334e-05,
      "loss": 0.0939,
      "positive_teacher_pairs": 26481.1,
      "query_h_norm_ratio": 0.465507972240448,
      "relation_loss": 0.016957656107842924,
      "selected_pairs": 26481.1,
      "step": 1500,
      "student_distance_mean": 34.44171485900879,
      "teacher_distance_mean": 0.8443897724151611,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.016957656107842924,
      "weighted_relation_loss": 0.016957656107842924
    },
    {
      "action_query_batch_std": 0.5463543772697449,
      "action_query_norm": 45.17428207397461,
      "action_relation_loss": 0.016606726218014956,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.412535079513564,
      "flow_matching_loss": 0.080060476064682,
      "grad_norm": 0.3261842429637909,
      "h_stock_valid_norm": 96.30303115844727,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.999997539434007e-05,
      "loss": 0.0967,
      "positive_teacher_pairs": 26966.8,
      "query_h_norm_ratio": 0.4690904080867767,
      "relation_loss": 0.016606726218014956,
      "selected_pairs": 26966.8,
      "step": 1510,
      "student_distance_mean": 35.013692474365236,
      "teacher_distance_mean": 0.8625234723091125,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.016606726218014956,
      "weighted_relation_loss": 0.016606726218014956
    },
    {
      "action_query_batch_std": 0.5595113396644592,
      "action_query_norm": 45.175187301635745,
      "action_relation_loss": 0.015437929891049862,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.421889616463985,
      "flow_matching_loss": 0.08133587390184402,
      "grad_norm": 0.35670843720436096,
      "h_stock_valid_norm": 97.27932205200196,
      "invalid_tail_samples": 20.8,
      "learning_rate": 9.999989033776898e-05,
      "loss": 0.0968,
      "positive_teacher_pairs": 27545.6,
      "query_h_norm_ratio": 0.464408802986145,
      "relation_loss": 0.015437929891049862,
      "selected_pairs": 27545.6,
      "step": 1520,
      "student_distance_mean": 35.63232917785645,
      "teacher_distance_mean": 0.8581378996372223,
      "valid_action_samples": 235.2,
      "weighted_auxiliary_loss": 0.015437929891049862,
      "weighted_relation_loss": 0.015437929891049862
    },
    {
      "action_query_batch_std": 0.5572593569755554,
      "action_query_norm": 45.173208999633786,
      "action_relation_loss": 0.01601644130423665,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.431244153414406,
      "flow_matching_loss": 0.08034766912460327,
      "grad_norm": 0.3128598630428314,
      "h_stock_valid_norm": 98.50079193115235,
      "invalid_tail_samples": 24.4,
      "learning_rate": 9.999974452661641e-05,
      "loss": 0.0964,
      "positive_teacher_pairs": 26709.9,
      "query_h_norm_ratio": 0.4586088448762894,
      "relation_loss": 0.01601644130423665,
      "selected_pairs": 26709.9,
      "step": 1530,
      "student_distance_mean": 35.553126907348634,
      "teacher_distance_mean": 0.8593062996864319,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.01601644130423665,
      "weighted_relation_loss": 0.01601644130423665
    },
    {
      "action_query_batch_std": 0.5500857651233673,
      "action_query_norm": 45.17079238891601,
      "action_relation_loss": 0.016082242876291276,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.440598690364827,
      "flow_matching_loss": 0.07860372960567474,
      "grad_norm": 0.27613160014152527,
      "h_stock_valid_norm": 98.7691436767578,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.999953796105959e-05,
      "loss": 0.0947,
      "positive_teacher_pairs": 26944.6,
      "query_h_norm_ratio": 0.45733844935894014,
      "relation_loss": 0.016082242876291276,
      "selected_pairs": 26944.6,
      "step": 1540,
      "student_distance_mean": 35.10965461730957,
      "teacher_distance_mean": 0.855421632528305,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.016082242876291276,
      "weighted_relation_loss": 0.016082242876291276
    },
    {
      "action_query_batch_std": 0.5430603802204133,
      "action_query_norm": 45.16723442077637,
      "action_relation_loss": 0.016667178273200987,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.4499532273152478,
      "flow_matching_loss": 0.07606496587395668,
      "grad_norm": 0.2657855153083801,
      "h_stock_valid_norm": 98.43047561645508,
      "invalid_tail_samples": 26.1,
      "learning_rate": 9.999927064134949e-05,
      "loss": 0.0927,
      "positive_teacher_pairs": 26316.8,
      "query_h_norm_ratio": 0.4588815331459045,
      "relation_loss": 0.016667178273200987,
      "selected_pairs": 26316.8,
      "step": 1550,
      "student_distance_mean": 34.58920097351074,
      "teacher_distance_mean": 0.8391379117965698,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.016667178273200987,
      "weighted_relation_loss": 0.016667178273200987
    },
    {
      "action_query_batch_std": 0.5556989133358001,
      "action_query_norm": 45.167572784423825,
      "action_relation_loss": 0.016189152747392653,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.4593077642656689,
      "flow_matching_loss": 0.08098200932145119,
      "grad_norm": 0.22135211527347565,
      "h_stock_valid_norm": 97.1134635925293,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.999894256781095e-05,
      "loss": 0.0972,
      "positive_teacher_pairs": 26826.8,
      "query_h_norm_ratio": 0.46510257124900817,
      "relation_loss": 0.016189152747392653,
      "selected_pairs": 26826.8,
      "step": 1560,
      "student_distance_mean": 35.447317886352536,
      "teacher_distance_mean": 0.8633062601089477,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.016189152747392653,
      "weighted_relation_loss": 0.016189152747392653
    },
    {
      "action_query_batch_std": 0.5615025341510773,
      "action_query_norm": 45.16831016540527,
      "action_relation_loss": 0.016877874732017517,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.4686623012160898,
      "flow_matching_loss": 0.0796213686466217,
      "grad_norm": 0.273857057094574,
      "h_stock_valid_norm": 96.47675857543945,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.99985537408426e-05,
      "loss": 0.0965,
      "positive_teacher_pairs": 27079.3,
      "query_h_norm_ratio": 0.4681833446025848,
      "relation_loss": 0.016877874732017517,
      "selected_pairs": 27079.3,
      "step": 1570,
      "student_distance_mean": 35.83283996582031,
      "teacher_distance_mean": 0.8577201306819916,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.016877874732017517,
      "weighted_relation_loss": 0.016877874732017517
    },
    {
      "action_query_batch_std": 0.5734101831912994,
      "action_query_norm": 45.16952819824219,
      "action_relation_loss": 0.016571511700749397,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.4780168381665106,
      "flow_matching_loss": 0.0776443287730217,
      "grad_norm": 0.2701156735420227,
      "h_stock_valid_norm": 95.96065368652344,
      "invalid_tail_samples": 23.1,
      "learning_rate": 9.999810416091688e-05,
      "loss": 0.0942,
      "positive_teacher_pairs": 27011.6,
      "query_h_norm_ratio": 0.4707139879465103,
      "relation_loss": 0.016571511700749397,
      "selected_pairs": 27011.6,
      "step": 1580,
      "student_distance_mean": 36.704884719848636,
      "teacher_distance_mean": 0.8518498837947845,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.016571511700749397,
      "weighted_relation_loss": 0.016571511700749397
    },
    {
      "action_query_batch_std": 0.5436052739620209,
      "action_query_norm": 45.16396713256836,
      "action_relation_loss": 0.016053412202745677,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.4873713751169317,
      "flow_matching_loss": 0.07813003435730934,
      "grad_norm": 0.2619543671607971,
      "h_stock_valid_norm": 97.64468002319336,
      "invalid_tail_samples": 25.8,
      "learning_rate": 9.99975938285801e-05,
      "loss": 0.0942,
      "positive_teacher_pairs": 26389.4,
      "query_h_norm_ratio": 0.4625579625368118,
      "relation_loss": 0.016053412202745677,
      "selected_pairs": 26389.4,
      "step": 1590,
      "student_distance_mean": 34.679921340942386,
      "teacher_distance_mean": 0.8571131944656372,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.016053412202745677,
      "weighted_relation_loss": 0.016053412202745677
    },
    {
      "action_query_batch_std": 0.5385957896709442,
      "action_query_norm": 45.163143920898435,
      "action_relation_loss": 0.016461206413805485,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.4967259120673526,
      "flow_matching_loss": 0.0792177326977253,
      "grad_norm": 0.17707601189613342,
      "h_stock_valid_norm": 98.32545623779296,
      "invalid_tail_samples": 20.6,
      "learning_rate": 9.999702274445236e-05,
      "loss": 0.0957,
      "positive_teacher_pairs": 27599.6,
      "query_h_norm_ratio": 0.4593250662088394,
      "relation_loss": 0.016461206413805485,
      "selected_pairs": 27599.6,
      "step": 1600,
      "student_distance_mean": 34.46547737121582,
      "teacher_distance_mean": 0.8488584697246552,
      "valid_action_samples": 235.4,
      "weighted_auxiliary_loss": 0.016461206413805485,
      "weighted_relation_loss": 0.016461206413805485
    },
    {
      "action_query_batch_std": 0.5467988669872283,
      "action_query_norm": 45.16162185668945,
      "action_relation_loss": 0.016155829653143883,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.5060804490177735,
      "flow_matching_loss": 0.07935700565576553,
      "grad_norm": 0.2694505453109741,
      "h_stock_valid_norm": 98.85325546264649,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.999639090922756e-05,
      "loss": 0.0955,
      "positive_teacher_pairs": 26897.9,
      "query_h_norm_ratio": 0.45686439871788026,
      "relation_loss": 0.016155829653143883,
      "selected_pairs": 26897.9,
      "step": 1610,
      "student_distance_mean": 34.88966789245605,
      "teacher_distance_mean": 0.86095090508461,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.016155829653143883,
      "weighted_relation_loss": 0.016155829653143883
    },
    {
      "action_query_batch_std": 0.5609873235225677,
      "action_query_norm": 45.16334419250488,
      "action_relation_loss": 0.015466060116887092,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.5154349859681946,
      "flow_matching_loss": 0.07452155500650406,
      "grad_norm": 0.3636872172355652,
      "h_stock_valid_norm": 99.92330474853516,
      "invalid_tail_samples": 22.1,
      "learning_rate": 9.999569832367346e-05,
      "loss": 0.09,
      "positive_teacher_pairs": 27243.8,
      "query_h_norm_ratio": 0.45198273062705996,
      "relation_loss": 0.015466060116887092,
      "selected_pairs": 27243.8,
      "step": 1620,
      "student_distance_mean": 35.79200210571289,
      "teacher_distance_mean": 0.8649421334266663,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.015466060116887092,
      "weighted_relation_loss": 0.015466060116887092
    },
    {
      "action_query_batch_std": 0.5534148633480072,
      "action_query_norm": 45.1598201751709,
      "action_relation_loss": 0.015698047541081905,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.5247895229186157,
      "flow_matching_loss": 0.07835796177387237,
      "grad_norm": 0.2901126444339752,
      "h_stock_valid_norm": 98.92325286865234,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.999494498863162e-05,
      "loss": 0.0941,
      "positive_teacher_pairs": 26732.0,
      "query_h_norm_ratio": 0.4565212458372116,
      "relation_loss": 0.015698047541081905,
      "selected_pairs": 26732.0,
      "step": 1630,
      "student_distance_mean": 35.297539138793944,
      "teacher_distance_mean": 0.8628821492195129,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.015698047541081905,
      "weighted_relation_loss": 0.015698047541081905
    },
    {
      "action_query_batch_std": 0.5636074304580688,
      "action_query_norm": 45.158416748046875,
      "action_relation_loss": 0.01584204053506255,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.5341440598690363,
      "flow_matching_loss": 0.07681288495659828,
      "grad_norm": 0.17580793797969818,
      "h_stock_valid_norm": 98.09182739257812,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.99941309050174e-05,
      "loss": 0.0927,
      "positive_teacher_pairs": 27085.0,
      "query_h_norm_ratio": 0.46037155091762544,
      "relation_loss": 0.01584204053506255,
      "selected_pairs": 27085.0,
      "step": 1640,
      "student_distance_mean": 35.933097457885744,
      "teacher_distance_mean": 0.8494516849517822,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.01584204053506255,
      "weighted_relation_loss": 0.01584204053506255
    },
    {
      "action_query_batch_std": 0.5569768667221069,
      "action_query_norm": 45.15323944091797,
      "action_relation_loss": 0.016218898911029102,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.5434985968194574,
      "flow_matching_loss": 0.07859856039285659,
      "grad_norm": 0.21774724125862122,
      "h_stock_valid_norm": 98.76788101196288,
      "invalid_tail_samples": 26.1,
      "learning_rate": 9.999325607381999e-05,
      "loss": 0.0948,
      "positive_teacher_pairs": 26320.7,
      "query_h_norm_ratio": 0.4571680098772049,
      "relation_loss": 0.016218898911029102,
      "selected_pairs": 26320.7,
      "step": 1650,
      "student_distance_mean": 35.545053482055664,
      "teacher_distance_mean": 0.8485081672668457,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.016218898911029102,
      "weighted_relation_loss": 0.016218898911029102
    },
    {
      "action_query_batch_std": 0.5484546184539795,
      "action_query_norm": 45.15066337585449,
      "action_relation_loss": 0.015881291031837462,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.5528531337698785,
      "flow_matching_loss": 0.07945639342069626,
      "grad_norm": 0.1629604548215866,
      "h_stock_valid_norm": 98.98746109008789,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.999232049610238e-05,
      "loss": 0.0953,
      "positive_teacher_pairs": 26759.7,
      "query_h_norm_ratio": 0.4561282068490982,
      "relation_loss": 0.015881291031837462,
      "selected_pairs": 26759.7,
      "step": 1660,
      "student_distance_mean": 34.98189697265625,
      "teacher_distance_mean": 0.8503999054431916,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.015881291031837462,
      "weighted_relation_loss": 0.015881291031837462
    },
    {
      "action_query_batch_std": 0.560086053609848,
      "action_query_norm": 45.15127334594727,
      "action_relation_loss": 0.01667632544413209,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.5622076707202992,
      "flow_matching_loss": 0.07611207365989685,
      "grad_norm": 0.2635723650455475,
      "h_stock_valid_norm": 97.95829849243164,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.999132417300142e-05,
      "loss": 0.0928,
      "positive_teacher_pairs": 26814.8,
      "query_h_norm_ratio": 0.4609259873628616,
      "relation_loss": 0.01667632544413209,
      "selected_pairs": 26814.8,
      "step": 1670,
      "student_distance_mean": 35.72200813293457,
      "teacher_distance_mean": 0.8581283807754516,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.01667632544413209,
      "weighted_relation_loss": 0.01667632544413209
    },
    {
      "action_query_batch_std": 0.562026971578598,
      "action_query_norm": 45.149832534790036,
      "action_relation_loss": 0.016352137736976146,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.5715622076707203,
      "flow_matching_loss": 0.0778638556599617,
      "grad_norm": 0.29053860902786255,
      "h_stock_valid_norm": 98.3719970703125,
      "invalid_tail_samples": 25.0,
      "learning_rate": 9.99902671057277e-05,
      "loss": 0.0942,
      "positive_teacher_pairs": 26572.3,
      "query_h_norm_ratio": 0.4589767962694168,
      "relation_loss": 0.016352137736976146,
      "selected_pairs": 26572.3,
      "step": 1680,
      "student_distance_mean": 35.88229904174805,
      "teacher_distance_mean": 0.8616723239421844,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.016352137736976146,
      "weighted_relation_loss": 0.016352137736976146
    },
    {
      "action_query_batch_std": 0.5480457484722138,
      "action_query_norm": 45.14709701538086,
      "action_relation_loss": 0.016153331845998764,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.5809167446211414,
      "flow_matching_loss": 0.08047213479876518,
      "grad_norm": 0.3353300392627716,
      "h_stock_valid_norm": 98.54424667358398,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.998914929556569e-05,
      "loss": 0.0966,
      "positive_teacher_pairs": 26998.2,
      "query_h_norm_ratio": 0.4581439048051834,
      "relation_loss": 0.016153331845998764,
      "selected_pairs": 26998.2,
      "step": 1690,
      "student_distance_mean": 34.92493934631348,
      "teacher_distance_mean": 0.8559258937835693,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.016153331845998764,
      "weighted_relation_loss": 0.016153331845998764
    },
    {
      "action_query_batch_std": 0.5700434744358063,
      "action_query_norm": 45.148062515258786,
      "action_relation_loss": 0.015597257949411868,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.590271281571562,
      "flow_matching_loss": 0.08235836327075959,
      "grad_norm": 0.2622681260108948,
      "h_stock_valid_norm": 98.3324203491211,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.998797074387361e-05,
      "loss": 0.098,
      "positive_teacher_pairs": 26685.9,
      "query_h_norm_ratio": 0.4591439336538315,
      "relation_loss": 0.015597257949411868,
      "selected_pairs": 26685.9,
      "step": 1700,
      "student_distance_mean": 36.50908660888672,
      "teacher_distance_mean": 0.8513954699039459,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.015597257949411868,
      "weighted_relation_loss": 0.015597257949411868
    },
    {
      "action_query_batch_std": 0.5587005615234375,
      "action_query_norm": 45.14456787109375,
      "action_relation_loss": 0.016149669513106348,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.5996258185219832,
      "flow_matching_loss": 0.0787391759455204,
      "grad_norm": 0.23334939777851105,
      "h_stock_valid_norm": 97.73004760742188,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.99867314520835e-05,
      "loss": 0.0949,
      "positive_teacher_pairs": 26668.3,
      "query_h_norm_ratio": 0.4619340389966965,
      "relation_loss": 0.016149669513106348,
      "selected_pairs": 26668.3,
      "step": 1710,
      "student_distance_mean": 35.810551452636716,
      "teacher_distance_mean": 0.8508182823657989,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.016149669513106348,
      "weighted_relation_loss": 0.016149669513106348
    },
    {
      "action_query_batch_std": 0.5727374136447907,
      "action_query_norm": 45.144918060302736,
      "action_relation_loss": 0.016168240644037723,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.6089803554724043,
      "flow_matching_loss": 0.07749119475483894,
      "grad_norm": 0.34017306566238403,
      "h_stock_valid_norm": 97.7558807373047,
      "invalid_tail_samples": 25.5,
      "learning_rate": 9.998543142170126e-05,
      "loss": 0.0937,
      "positive_teacher_pairs": 26461.7,
      "query_h_norm_ratio": 0.46181808710098265,
      "relation_loss": 0.016168240644037723,
      "selected_pairs": 26461.7,
      "step": 1720,
      "student_distance_mean": 36.86399459838867,
      "teacher_distance_mean": 0.8585139513015747,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.016168240644037723,
      "weighted_relation_loss": 0.016168240644037723
    },
    {
      "action_query_batch_std": 0.5557635605335236,
      "action_query_norm": 45.13998374938965,
      "action_relation_loss": 0.015941330790519716,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.618334892422825,
      "flow_matching_loss": 0.08099274784326553,
      "grad_norm": 0.3770643174648285,
      "h_stock_valid_norm": 98.04139251708985,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.99840706543065e-05,
      "loss": 0.0969,
      "positive_teacher_pairs": 27149.7,
      "query_h_norm_ratio": 0.4604218423366547,
      "relation_loss": 0.015941330790519716,
      "selected_pairs": 27149.7,
      "step": 1730,
      "student_distance_mean": 35.503286361694336,
      "teacher_distance_mean": 0.8396815836429596,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.015941330790519716,
      "weighted_relation_loss": 0.015941330790519716
    },
    {
      "action_query_batch_std": 0.5433460354804993,
      "action_query_norm": 45.137446594238284,
      "action_relation_loss": 0.015990648046135904,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.627689429373246,
      "flow_matching_loss": 0.07791259065270424,
      "grad_norm": 0.3580467700958252,
      "h_stock_valid_norm": 97.47067794799804,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.998264915155274e-05,
      "loss": 0.0939,
      "positive_teacher_pairs": 26826.4,
      "query_h_norm_ratio": 0.4631018042564392,
      "relation_loss": 0.015990648046135904,
      "selected_pairs": 26826.4,
      "step": 1740,
      "student_distance_mean": 34.80019989013672,
      "teacher_distance_mean": 0.8534941732883453,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.015990648046135904,
      "weighted_relation_loss": 0.015990648046135904
    },
    {
      "action_query_batch_std": 0.5522192537784576,
      "action_query_norm": 45.13565864562988,
      "action_relation_loss": 0.015298975910991431,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.637043966323667,
      "flow_matching_loss": 0.07671834751963616,
      "grad_norm": 0.2660585343837738,
      "h_stock_valid_norm": 97.12141571044921,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.998116691516718e-05,
      "loss": 0.092,
      "positive_teacher_pairs": 26640.6,
      "query_h_norm_ratio": 0.464737144112587,
      "relation_loss": 0.015298975910991431,
      "selected_pairs": 26640.6,
      "step": 1750,
      "student_distance_mean": 35.475277709960935,
      "teacher_distance_mean": 0.8532587826251984,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.015298975910991431,
      "weighted_relation_loss": 0.015298975910991431
    },
    {
      "action_query_batch_std": 0.5564303576946259,
      "action_query_norm": 45.13242530822754,
      "action_relation_loss": 0.01705737803131342,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.646398503274088,
      "flow_matching_loss": 0.08000436499714851,
      "grad_norm": 0.30401676893234253,
      "h_stock_valid_norm": 97.08911819458008,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.997962394695091e-05,
      "loss": 0.0971,
      "positive_teacher_pairs": 26644.2,
      "query_h_norm_ratio": 0.46485785245895384,
      "relation_loss": 0.01705737803131342,
      "selected_pairs": 26644.2,
      "step": 1760,
      "student_distance_mean": 35.60954627990723,
      "teacher_distance_mean": 0.8516146063804626,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.01705737803131342,
      "weighted_relation_loss": 0.01705737803131342
    },
    {
      "action_query_batch_std": 0.5619522333145142,
      "action_query_norm": 45.131801223754884,
      "action_relation_loss": 0.015028716064989567,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.6557530402245089,
      "flow_matching_loss": 0.07604829892516136,
      "grad_norm": 0.21102069318294525,
      "h_stock_valid_norm": 97.59265441894532,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.997802024877875e-05,
      "loss": 0.0911,
      "positive_teacher_pairs": 26640.7,
      "query_h_norm_ratio": 0.46245417296886443,
      "relation_loss": 0.015028716064989567,
      "selected_pairs": 26640.7,
      "step": 1770,
      "student_distance_mean": 36.122650146484375,
      "teacher_distance_mean": 0.8575116336345673,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.015028716064989567,
      "weighted_relation_loss": 0.015028716064989567
    },
    {
      "action_query_batch_std": 0.5443063795566558,
      "action_query_norm": 45.12946395874023,
      "action_relation_loss": 0.015815689414739608,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.66510757717493,
      "flow_matching_loss": 0.07743917256593705,
      "grad_norm": 0.22619757056236267,
      "h_stock_valid_norm": 98.27591018676758,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.99763558225994e-05,
      "loss": 0.0933,
      "positive_teacher_pairs": 26638.9,
      "query_h_norm_ratio": 0.45921685099601744,
      "relation_loss": 0.015815689414739608,
      "selected_pairs": 26638.9,
      "step": 1780,
      "student_distance_mean": 34.76725769042969,
      "teacher_distance_mean": 0.8488313555717468,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.015815689414739608,
      "weighted_relation_loss": 0.015815689414739608
    },
    {
      "action_query_batch_std": 0.5519847810268402,
      "action_query_norm": 45.129163360595705,
      "action_relation_loss": 0.01612661061808467,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.6744621141253508,
      "flow_matching_loss": 0.07536040991544724,
      "grad_norm": 0.266723096370697,
      "h_stock_valid_norm": 98.05405120849609,
      "invalid_tail_samples": 21.7,
      "learning_rate": 9.997463067043526e-05,
      "loss": 0.0915,
      "positive_teacher_pairs": 27342.1,
      "query_h_norm_ratio": 0.46024850010871887,
      "relation_loss": 0.01612661061808467,
      "selected_pairs": 27342.1,
      "step": 1790,
      "student_distance_mean": 35.4487190246582,
      "teacher_distance_mean": 0.8521231830120086,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.01612661061808467,
      "weighted_relation_loss": 0.01612661061808467
    },
    {
      "action_query_batch_std": 0.5483396351337433,
      "action_query_norm": 45.12624969482422,
      "action_relation_loss": 0.01580517953261733,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.6838166510757717,
      "flow_matching_loss": 0.07714308239519596,
      "grad_norm": 0.20466667413711548,
      "h_stock_valid_norm": 97.8838264465332,
      "invalid_tail_samples": 22.6,
      "learning_rate": 9.997284479438253e-05,
      "loss": 0.0929,
      "positive_teacher_pairs": 27130.7,
      "query_h_norm_ratio": 0.46102002561092376,
      "relation_loss": 0.01580517953261733,
      "selected_pairs": 27130.7,
      "step": 1800,
      "student_distance_mean": 35.23682289123535,
      "teacher_distance_mean": 0.8461378514766693,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.01580517953261733,
      "weighted_relation_loss": 0.01580517953261733
    },
    {
      "action_query_batch_std": 0.5378807425498963,
      "action_query_norm": 45.11991500854492,
      "action_relation_loss": 0.015991416573524476,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.6931711880261928,
      "flow_matching_loss": 0.07749746665358544,
      "grad_norm": 0.20678451657295227,
      "h_stock_valid_norm": 98.27103958129882,
      "invalid_tail_samples": 25.4,
      "learning_rate": 9.997099819661127e-05,
      "loss": 0.0935,
      "positive_teacher_pairs": 26480.5,
      "query_h_norm_ratio": 0.4591391861438751,
      "relation_loss": 0.015991416573524476,
      "selected_pairs": 26480.5,
      "step": 1810,
      "student_distance_mean": 34.695349502563474,
      "teacher_distance_mean": 0.8522658288478852,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.015991416573524476,
      "weighted_relation_loss": 0.015991416573524476
    },
    {
      "action_query_batch_std": 0.5564488470554352,
      "action_query_norm": 45.124568176269534,
      "action_relation_loss": 0.01615749569609761,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.7025257249766137,
      "flow_matching_loss": 0.07738340273499489,
      "grad_norm": 0.223147913813591,
      "h_stock_valid_norm": 99.04293594360351,
      "invalid_tail_samples": 21.7,
      "learning_rate": 9.996909087936524e-05,
      "loss": 0.0935,
      "positive_teacher_pairs": 27339.8,
      "query_h_norm_ratio": 0.45561084151268005,
      "relation_loss": 0.01615749569609761,
      "selected_pairs": 27339.8,
      "step": 1820,
      "student_distance_mean": 35.529335021972656,
      "teacher_distance_mean": 0.8603769779205322,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.01615749569609761,
      "weighted_relation_loss": 0.01615749569609761
    },
    {
      "action_query_batch_std": 0.5659746289253235,
      "action_query_norm": 45.12551498413086,
      "action_relation_loss": 0.016110969986766577,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.7118802619270346,
      "flow_matching_loss": 0.07988027557730674,
      "grad_norm": 0.21030136942863464,
      "h_stock_valid_norm": 98.59375839233398,
      "invalid_tail_samples": 22.6,
      "learning_rate": 9.996712284496201e-05,
      "loss": 0.096,
      "positive_teacher_pairs": 27126.5,
      "query_h_norm_ratio": 0.4577066570520401,
      "relation_loss": 0.016110969986766577,
      "selected_pairs": 27126.5,
      "step": 1830,
      "student_distance_mean": 36.19895706176758,
      "teacher_distance_mean": 0.8587318360805511,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.016110969986766577,
      "weighted_relation_loss": 0.016110969986766577
    },
    {
      "action_query_batch_std": 0.5535334169864654,
      "action_query_norm": 45.12038078308105,
      "action_relation_loss": 0.015344396978616715,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.7212347988774557,
      "flow_matching_loss": 0.07548646330833435,
      "grad_norm": 0.22249197959899902,
      "h_stock_valid_norm": 96.57565689086914,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.996509409579293e-05,
      "loss": 0.0908,
      "positive_teacher_pairs": 26743.6,
      "query_h_norm_ratio": 0.4672181248664856,
      "relation_loss": 0.015344396978616715,
      "selected_pairs": 26743.6,
      "step": 1840,
      "student_distance_mean": 35.45891075134277,
      "teacher_distance_mean": 0.8635297954082489,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.015344396978616715,
      "weighted_relation_loss": 0.015344396978616715
    },
    {
      "action_query_batch_std": 0.5461754202842712,
      "action_query_norm": 45.12029838562012,
      "action_relation_loss": 0.015727725252509116,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.7305893358278766,
      "flow_matching_loss": 0.07931300848722458,
      "grad_norm": 0.16750115156173706,
      "h_stock_valid_norm": 96.46136932373047,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.996300463432312e-05,
      "loss": 0.095,
      "positive_teacher_pairs": 27202.9,
      "query_h_norm_ratio": 0.46776789128780366,
      "relation_loss": 0.015727725252509116,
      "selected_pairs": 27202.9,
      "step": 1850,
      "student_distance_mean": 34.92101936340332,
      "teacher_distance_mean": 0.8517641246318817,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.015727725252509116,
      "weighted_relation_loss": 0.015727725252509116
    },
    {
      "action_query_batch_std": 0.5536134839057922,
      "action_query_norm": 45.11448936462402,
      "action_relation_loss": 0.01644133049994707,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.7399438727782974,
      "flow_matching_loss": 0.07246873304247856,
      "grad_norm": 0.20571431517601013,
      "h_stock_valid_norm": 97.85666580200196,
      "invalid_tail_samples": 25.4,
      "learning_rate": 9.996085446309148e-05,
      "loss": 0.0889,
      "positive_teacher_pairs": 26477.1,
      "query_h_norm_ratio": 0.46103355884552,
      "relation_loss": 0.01644133049994707,
      "selected_pairs": 26477.1,
      "step": 1860,
      "student_distance_mean": 35.59827651977539,
      "teacher_distance_mean": 0.849731731414795,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.01644133049994707,
      "weighted_relation_loss": 0.01644133049994707
    },
    {
      "action_query_batch_std": 0.5580399990081787,
      "action_query_norm": 45.1145523071289,
      "action_relation_loss": 0.016021806560456753,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.7492984097287185,
      "flow_matching_loss": 0.07540042251348496,
      "grad_norm": 0.2423938512802124,
      "h_stock_valid_norm": 97.82011337280274,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.995864358471066e-05,
      "loss": 0.0914,
      "positive_teacher_pairs": 26878.5,
      "query_h_norm_ratio": 0.4612040609121323,
      "relation_loss": 0.016021806560456753,
      "selected_pairs": 26878.5,
      "step": 1870,
      "student_distance_mean": 35.82481689453125,
      "teacher_distance_mean": 0.847928935289383,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.016021806560456753,
      "weighted_relation_loss": 0.016021806560456753
    },
    {
      "action_query_batch_std": 0.5619132041931152,
      "action_query_norm": 45.114802551269534,
      "action_relation_loss": 0.015909655299037696,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.7586529466791394,
      "flow_matching_loss": 0.07435320541262627,
      "grad_norm": 0.24144026637077332,
      "h_stock_valid_norm": 97.74000625610351,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.99563720018671e-05,
      "loss": 0.0903,
      "positive_teacher_pairs": 27152.7,
      "query_h_norm_ratio": 0.4615832716226578,
      "relation_loss": 0.015909655299037696,
      "selected_pairs": 27152.7,
      "step": 1880,
      "student_distance_mean": 35.89212226867676,
      "teacher_distance_mean": 0.8646112322807312,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.015909655299037696,
      "weighted_relation_loss": 0.015909655299037696
    },
    {
      "action_query_batch_std": 0.5483502209186554,
      "action_query_norm": 45.11025466918945,
      "action_relation_loss": 0.016352304629981516,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.7680074836295603,
      "flow_matching_loss": 0.07647572681307793,
      "grad_norm": 0.23619025945663452,
      "h_stock_valid_norm": 97.61125030517579,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.995403971732098e-05,
      "loss": 0.0928,
      "positive_teacher_pairs": 27152.4,
      "query_h_norm_ratio": 0.4621442437171936,
      "relation_loss": 0.016352304629981516,
      "selected_pairs": 27152.4,
      "step": 1890,
      "student_distance_mean": 35.05602264404297,
      "teacher_distance_mean": 0.8561167418956757,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.016352304629981516,
      "weighted_relation_loss": 0.016352304629981516
    },
    {
      "action_query_batch_std": 0.5505945384502411,
      "action_query_norm": 45.109165573120116,
      "action_relation_loss": 0.016097992565482853,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.7773620205799814,
      "flow_matching_loss": 0.07238096594810486,
      "grad_norm": 0.2394661158323288,
      "h_stock_valid_norm": 96.23172302246094,
      "invalid_tail_samples": 23.0,
      "learning_rate": 9.995164673390625e-05,
      "loss": 0.0885,
      "positive_teacher_pairs": 27033.6,
      "query_h_norm_ratio": 0.4687703251838684,
      "relation_loss": 0.016097992565482853,
      "selected_pairs": 27033.6,
      "step": 1900,
      "student_distance_mean": 35.27130241394043,
      "teacher_distance_mean": 0.8510109961032868,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.016097992565482853,
      "weighted_relation_loss": 0.016097992565482853
    },
    {
      "action_query_batch_std": 0.5588025927543641,
      "action_query_norm": 45.10836906433106,
      "action_relation_loss": 0.01531189475208521,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.7867165575304023,
      "flow_matching_loss": 0.07567714899778366,
      "grad_norm": 0.29203274846076965,
      "h_stock_valid_norm": 96.5125015258789,
      "invalid_tail_samples": 25.4,
      "learning_rate": 9.994919305453059e-05,
      "loss": 0.091,
      "positive_teacher_pairs": 26486.7,
      "query_h_norm_ratio": 0.46738913357257844,
      "relation_loss": 0.01531189475208521,
      "selected_pairs": 26486.7,
      "step": 1910,
      "student_distance_mean": 35.81205101013184,
      "teacher_distance_mean": 0.8620594322681427,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.01531189475208521,
      "weighted_relation_loss": 0.01531189475208521
    },
    {
      "action_query_batch_std": 0.5518398463726044,
      "action_query_norm": 45.10373725891113,
      "action_relation_loss": 0.01499724630266428,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.7960710944808231,
      "flow_matching_loss": 0.07798553481698037,
      "grad_norm": 0.28345221281051636,
      "h_stock_valid_norm": 98.26763763427735,
      "invalid_tail_samples": 19.5,
      "learning_rate": 9.994667868217548e-05,
      "loss": 0.093,
      "positive_teacher_pairs": 27858.6,
      "query_h_norm_ratio": 0.45899394154548645,
      "relation_loss": 0.01499724630266428,
      "selected_pairs": 27858.6,
      "step": 1920,
      "student_distance_mean": 35.42411117553711,
      "teacher_distance_mean": 0.8557267785072327,
      "valid_action_samples": 236.5,
      "weighted_auxiliary_loss": 0.01499724630266428,
      "weighted_relation_loss": 0.01499724630266428
    },
    {
      "action_query_batch_std": 0.5525559902191162,
      "action_query_norm": 45.10150680541992,
      "action_relation_loss": 0.01660324651747942,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.8054256314312442,
      "flow_matching_loss": 0.07595034241676331,
      "grad_norm": 0.23535941541194916,
      "h_stock_valid_norm": 99.0602783203125,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.99441036198961e-05,
      "loss": 0.0926,
      "positive_teacher_pairs": 27152.7,
      "query_h_norm_ratio": 0.4553008109331131,
      "relation_loss": 0.01660324651747942,
      "selected_pairs": 27152.7,
      "step": 1930,
      "student_distance_mean": 35.4240047454834,
      "teacher_distance_mean": 0.8481143653392792,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.01660324651747942,
      "weighted_relation_loss": 0.01660324651747942
    },
    {
      "action_query_batch_std": 0.5458601534366607,
      "action_query_norm": 45.09855613708496,
      "action_relation_loss": 0.015531003940850496,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.8147801683816651,
      "flow_matching_loss": 0.07599896788597107,
      "grad_norm": 0.3014294505119324,
      "h_stock_valid_norm": 98.44401092529297,
      "invalid_tail_samples": 27.2,
      "learning_rate": 9.99414678708214e-05,
      "loss": 0.0915,
      "positive_teacher_pairs": 26075.9,
      "query_h_norm_ratio": 0.45812085270881653,
      "relation_loss": 0.015531003940850496,
      "selected_pairs": 26075.9,
      "step": 1940,
      "student_distance_mean": 35.05231857299805,
      "teacher_distance_mean": 0.8543880105018615,
      "valid_action_samples": 228.8,
      "weighted_auxiliary_loss": 0.015531003940850496,
      "weighted_relation_loss": 0.015531003940850496
    },
    {
      "action_query_batch_std": 0.5472501456737519,
      "action_query_norm": 45.09912185668945,
      "action_relation_loss": 0.015691119991242886,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.824134705332086,
      "flow_matching_loss": 0.07398982867598533,
      "grad_norm": 0.30071476101875305,
      "h_stock_valid_norm": 97.78571014404297,
      "invalid_tail_samples": 26.6,
      "learning_rate": 9.993877143815407e-05,
      "loss": 0.0897,
      "positive_teacher_pairs": 26204.5,
      "query_h_norm_ratio": 0.4612088829278946,
      "relation_loss": 0.015691119991242886,
      "selected_pairs": 26204.5,
      "step": 1950,
      "student_distance_mean": 35.03309440612793,
      "teacher_distance_mean": 0.8536416530609131,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.015691119991242886,
      "weighted_relation_loss": 0.015691119991242886
    },
    {
      "action_query_batch_std": 0.5488614976406098,
      "action_query_norm": 45.095313262939456,
      "action_relation_loss": 0.01609486434608698,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.833489242282507,
      "flow_matching_loss": 0.0736922137439251,
      "grad_norm": 0.2518589198589325,
      "h_stock_valid_norm": 98.79843826293946,
      "invalid_tail_samples": 25.9,
      "learning_rate": 9.993601432517053e-05,
      "loss": 0.0898,
      "positive_teacher_pairs": 26369.8,
      "query_h_norm_ratio": 0.4564428418874741,
      "relation_loss": 0.01609486434608698,
      "selected_pairs": 26369.8,
      "step": 1960,
      "student_distance_mean": 35.33509635925293,
      "teacher_distance_mean": 0.8458411753177643,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.01609486434608698,
      "weighted_relation_loss": 0.01609486434608698
    },
    {
      "action_query_batch_std": 0.5470472693443298,
      "action_query_norm": 45.0928352355957,
      "action_relation_loss": 0.015669303946197033,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.842843779232928,
      "flow_matching_loss": 0.07336855307221413,
      "grad_norm": 0.23762644827365875,
      "h_stock_valid_norm": 97.96613693237305,
      "invalid_tail_samples": 26.6,
      "learning_rate": 9.993319653522091e-05,
      "loss": 0.089,
      "positive_teacher_pairs": 26216.1,
      "query_h_norm_ratio": 0.460306379199028,
      "relation_loss": 0.015669303946197033,
      "selected_pairs": 26216.1,
      "step": 1970,
      "student_distance_mean": 35.176110458374026,
      "teacher_distance_mean": 0.8504306733608246,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.015669303946197033,
      "weighted_relation_loss": 0.015669303946197033
    },
    {
      "action_query_batch_std": 0.5567104339599609,
      "action_query_norm": 45.09538536071777,
      "action_relation_loss": 0.015352122671902179,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.8521983161833488,
      "flow_matching_loss": 0.07521693482995033,
      "grad_norm": 0.2340080738067627,
      "h_stock_valid_norm": 96.56769638061523,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.993031807172911e-05,
      "loss": 0.0906,
      "positive_teacher_pairs": 26673.4,
      "query_h_norm_ratio": 0.46698732376098634,
      "relation_loss": 0.015352122671902179,
      "selected_pairs": 26673.4,
      "step": 1980,
      "student_distance_mean": 35.9508056640625,
      "teacher_distance_mean": 0.8552246510982513,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.015352122671902179,
      "weighted_relation_loss": 0.015352122671902179
    },
    {
      "action_query_batch_std": 0.5494475901126862,
      "action_query_norm": 45.0914867401123,
      "action_relation_loss": 0.01453190939500928,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.86155285313377,
      "flow_matching_loss": 0.07508675232529641,
      "grad_norm": 0.1578192114830017,
      "h_stock_valid_norm": 95.92003479003907,
      "invalid_tail_samples": 26.2,
      "learning_rate": 9.992737893819273e-05,
      "loss": 0.0896,
      "positive_teacher_pairs": 26311.4,
      "query_h_norm_ratio": 0.47010002434253695,
      "relation_loss": 0.01453190939500928,
      "selected_pairs": 26311.4,
      "step": 1990,
      "student_distance_mean": 35.28171234130859,
      "teacher_distance_mean": 0.8612596929073334,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.01453190939500928,
      "weighted_relation_loss": 0.01453190939500928
    },
    {
      "action_query_batch_std": 0.5435617387294769,
      "action_query_norm": 45.0897403717041,
      "action_relation_loss": 0.014971380587667227,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.8709073900841908,
      "flow_matching_loss": 0.07448385059833526,
      "grad_norm": 0.2722168564796448,
      "h_stock_valid_norm": 95.53534469604492,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.992437913818312e-05,
      "loss": 0.0895,
      "positive_teacher_pairs": 26806.5,
      "query_h_norm_ratio": 0.4719745934009552,
      "relation_loss": 0.014971380587667227,
      "selected_pairs": 26806.5,
      "step": 2000,
      "student_distance_mean": 34.91555061340332,
      "teacher_distance_mean": 0.8420625746250152,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.014971380587667227,
      "weighted_relation_loss": 0.014971380587667227
    },
    {
      "action_query_batch_std": 0.5551852226257324,
      "action_query_norm": 45.091482162475586,
      "action_relation_loss": 0.015197704918682576,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.8802619270346117,
      "flow_matching_loss": 0.07519381940364837,
      "grad_norm": 0.2531036138534546,
      "h_stock_valid_norm": 97.3766258239746,
      "invalid_tail_samples": 21.5,
      "learning_rate": 9.992131867534526e-05,
      "loss": 0.0904,
      "positive_teacher_pairs": 27383.8,
      "query_h_norm_ratio": 0.46306752860546113,
      "relation_loss": 0.015197704918682576,
      "selected_pairs": 27383.8,
      "step": 2010,
      "student_distance_mean": 35.627568054199216,
      "teacher_distance_mean": 0.8498137593269348,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.015197704918682576,
      "weighted_relation_loss": 0.015197704918682576
    },
    {
      "action_query_batch_std": 0.5497061610221863,
      "action_query_norm": 45.08644752502441,
      "action_relation_loss": 0.015441721398383379,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.8896164639850328,
      "flow_matching_loss": 0.07613185942173004,
      "grad_norm": 0.22088523209095,
      "h_stock_valid_norm": 98.29954071044922,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.991819755339796e-05,
      "loss": 0.0916,
      "positive_teacher_pairs": 26844.6,
      "query_h_norm_ratio": 0.4586724370718002,
      "relation_loss": 0.015441721398383379,
      "selected_pairs": 26844.6,
      "step": 2020,
      "student_distance_mean": 35.25251350402832,
      "teacher_distance_mean": 0.8449108958244324,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.015441721398383379,
      "weighted_relation_loss": 0.015441721398383379
    },
    {
      "action_query_batch_std": 0.5484351933002471,
      "action_query_norm": 45.08515548706055,
      "action_relation_loss": 0.014773639291524887,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.8989710009354537,
      "flow_matching_loss": 0.07465183213353158,
      "grad_norm": 0.2133500576019287,
      "h_stock_valid_norm": 99.37676162719727,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.991501577613365e-05,
      "loss": 0.0894,
      "positive_teacher_pairs": 26619.7,
      "query_h_norm_ratio": 0.45368329584598543,
      "relation_loss": 0.014773639291524887,
      "selected_pairs": 26619.7,
      "step": 2030,
      "student_distance_mean": 35.25890769958496,
      "teacher_distance_mean": 0.855025976896286,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.014773639291524887,
      "weighted_relation_loss": 0.014773639291524887
    },
    {
      "action_query_batch_std": 0.5533151984214782,
      "action_query_norm": 45.08640556335449,
      "action_relation_loss": 0.014204783271998167,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.9083255378858746,
      "flow_matching_loss": 0.07535634636878967,
      "grad_norm": 0.2585781216621399,
      "h_stock_valid_norm": 100.07632522583008,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.99117733474185e-05,
      "loss": 0.0896,
      "positive_teacher_pairs": 26871.5,
      "query_h_norm_ratio": 0.4505251348018646,
      "relation_loss": 0.014204783271998167,
      "selected_pairs": 26871.5,
      "step": 2040,
      "student_distance_mean": 35.404762268066406,
      "teacher_distance_mean": 0.8520225286483765,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.014204783271998167,
      "weighted_relation_loss": 0.014204783271998167
    },
    {
      "action_query_batch_std": 0.5483445823192596,
      "action_query_norm": 45.08297500610352,
      "action_relation_loss": 0.014872339181602,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.9176800748362957,
      "flow_matching_loss": 0.07666403204202651,
      "grad_norm": 0.2708301246166229,
      "h_stock_valid_norm": 99.14708633422852,
      "invalid_tail_samples": 26.0,
      "learning_rate": 9.990847027119234e-05,
      "loss": 0.0915,
      "positive_teacher_pairs": 26341.8,
      "query_h_norm_ratio": 0.45471992194652555,
      "relation_loss": 0.014872339181602,
      "selected_pairs": 26341.8,
      "step": 2050,
      "student_distance_mean": 35.156521224975585,
      "teacher_distance_mean": 0.8660690903663635,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.014872339181602,
      "weighted_relation_loss": 0.014872339181602
    },
    {
      "action_query_batch_std": 0.5485113143920899,
      "action_query_norm": 45.07956123352051,
      "action_relation_loss": 0.014575699623674155,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.9270346117867165,
      "flow_matching_loss": 0.0713735654950142,
      "grad_norm": 0.23805172741413116,
      "h_stock_valid_norm": 98.57934036254883,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.990510655146877e-05,
      "loss": 0.0859,
      "positive_teacher_pairs": 26916.8,
      "query_h_norm_ratio": 0.4573036640882492,
      "relation_loss": 0.014575699623674155,
      "selected_pairs": 26916.8,
      "step": 2060,
      "student_distance_mean": 35.21119422912598,
      "teacher_distance_mean": 0.8665591061115265,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.014575699623674155,
      "weighted_relation_loss": 0.014575699623674155
    },
    {
      "action_query_batch_std": 0.5407563984394074,
      "action_query_norm": 45.075137329101565,
      "action_relation_loss": 0.014671579655259847,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.9363891487371374,
      "flow_matching_loss": 0.0736463040113449,
      "grad_norm": 0.26561203598976135,
      "h_stock_valid_norm": 99.68539810180664,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.990168219233496e-05,
      "loss": 0.0883,
      "positive_teacher_pairs": 27174.8,
      "query_h_norm_ratio": 0.4521757662296295,
      "relation_loss": 0.014671579655259847,
      "selected_pairs": 27174.8,
      "step": 2070,
      "student_distance_mean": 34.7742431640625,
      "teacher_distance_mean": 0.8529402673244476,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.014671579655259847,
      "weighted_relation_loss": 0.014671579655259847
    },
    {
      "action_query_batch_std": 0.5338991284370422,
      "action_query_norm": 45.073063659667966,
      "action_relation_loss": 0.015089763049036264,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.9457436856875585,
      "flow_matching_loss": 0.07316865846514702,
      "grad_norm": 0.2102736383676529,
      "h_stock_valid_norm": 99.43523483276367,
      "invalid_tail_samples": 22.9,
      "learning_rate": 9.989819719795188e-05,
      "loss": 0.0883,
      "positive_teacher_pairs": 27056.8,
      "query_h_norm_ratio": 0.4532927483320236,
      "relation_loss": 0.015089763049036264,
      "selected_pairs": 27056.8,
      "step": 2080,
      "student_distance_mean": 34.11381568908691,
      "teacher_distance_mean": 0.8489420056343079,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.015089763049036264,
      "weighted_relation_loss": 0.015089763049036264
    },
    {
      "action_query_batch_std": 0.5299155175685882,
      "action_query_norm": 45.070257568359374,
      "action_relation_loss": 0.014588798396289349,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.9550982226379794,
      "flow_matching_loss": 0.07065311297774315,
      "grad_norm": 0.24846570193767548,
      "h_stock_valid_norm": 99.53402786254883,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.989465157255412e-05,
      "loss": 0.0852,
      "positive_teacher_pairs": 26812.1,
      "query_h_norm_ratio": 0.45281645357608796,
      "relation_loss": 0.014588798396289349,
      "selected_pairs": 26812.1,
      "step": 2090,
      "student_distance_mean": 34.0838508605957,
      "teacher_distance_mean": 0.8544936001300811,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.014588798396289349,
      "weighted_relation_loss": 0.014588798396289349
    },
    {
      "action_query_batch_std": 0.5386904716491699,
      "action_query_norm": 45.069434356689456,
      "action_relation_loss": 0.015390289761126042,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.9644527595884003,
      "flow_matching_loss": 0.07485003918409347,
      "grad_norm": 0.2305055856704712,
      "h_stock_valid_norm": 98.58555145263672,
      "invalid_tail_samples": 27.3,
      "learning_rate": 9.989104532044994e-05,
      "loss": 0.0902,
      "positive_teacher_pairs": 26047.8,
      "query_h_norm_ratio": 0.45718251168727875,
      "relation_loss": 0.015390289761126042,
      "selected_pairs": 26047.8,
      "step": 2100,
      "student_distance_mean": 34.64475784301758,
      "teacher_distance_mean": 0.854183691740036,
      "valid_action_samples": 228.7,
      "weighted_auxiliary_loss": 0.015390289761126042,
      "weighted_relation_loss": 0.015390289761126042
    },
    {
      "action_query_batch_std": 0.5342781901359558,
      "action_query_norm": 45.068017196655276,
      "action_relation_loss": 0.015112135000526905,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.9738072965388214,
      "flow_matching_loss": 0.07323893159627914,
      "grad_norm": 0.29488641023635864,
      "h_stock_valid_norm": 97.1142074584961,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.988737844602128e-05,
      "loss": 0.0884,
      "positive_teacher_pairs": 27162.7,
      "query_h_norm_ratio": 0.4640747904777527,
      "relation_loss": 0.015112135000526905,
      "selected_pairs": 27162.7,
      "step": 2110,
      "student_distance_mean": 34.25485496520996,
      "teacher_distance_mean": 0.8452260613441467,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.015112135000526905,
      "weighted_relation_loss": 0.015112135000526905
    },
    {
      "action_query_batch_std": 0.5293061375617981,
      "action_query_norm": 45.065155029296875,
      "action_relation_loss": 0.014576654974371194,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.9831618334892422,
      "flow_matching_loss": 0.07332916185259819,
      "grad_norm": 0.1717747449874878,
      "h_stock_valid_norm": 96.41341629028321,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.988365095372372e-05,
      "loss": 0.0879,
      "positive_teacher_pairs": 27179.4,
      "query_h_norm_ratio": 0.4674204081296921,
      "relation_loss": 0.014576654974371194,
      "selected_pairs": 27179.4,
      "step": 2120,
      "student_distance_mean": 33.94641952514648,
      "teacher_distance_mean": 0.8593399822711945,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.014576654974371194,
      "weighted_relation_loss": 0.014576654974371194
    },
    {
      "action_query_batch_std": 0.532834130525589,
      "action_query_norm": 45.06166229248047,
      "action_relation_loss": 0.015568631049245597,
      "auxiliary_loss_weight": 1.0,
      "epoch": 1.9925163704396631,
      "flow_matching_loss": 0.07382717505097389,
      "grad_norm": 0.17367342114448547,
      "h_stock_valid_norm": 97.24866104125977,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.987986284808654e-05,
      "loss": 0.0894,
      "positive_teacher_pairs": 26916.4,
      "query_h_norm_ratio": 0.46338277161121366,
      "relation_loss": 0.015568631049245597,
      "selected_pairs": 26916.4,
      "step": 2130,
      "student_distance_mean": 34.26640853881836,
      "teacher_distance_mean": 0.8468732237815857,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.015568631049245597,
      "weighted_relation_loss": 0.015568631049245597
    },
    {
      "action_query_batch_std": 0.5338034749031066,
      "action_query_norm": 45.06399726867676,
      "action_relation_loss": 0.014487301092594862,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.001870907390084,
      "flow_matching_loss": 0.07276409417390824,
      "grad_norm": 0.18571871519088745,
      "h_stock_valid_norm": 99.15515060424805,
      "invalid_tail_samples": 20.9,
      "learning_rate": 9.987601413371264e-05,
      "loss": 0.0873,
      "positive_teacher_pairs": 24556.9,
      "query_h_norm_ratio": 0.4544933199882507,
      "relation_loss": 0.014487301092594862,
      "selected_pairs": 24556.9,
      "step": 2140,
      "student_distance_mean": 34.39793701171875,
      "teacher_distance_mean": 0.8547848343849183,
      "valid_action_samples": 215.2,
      "weighted_auxiliary_loss": 0.014487301092594862,
      "weighted_relation_loss": 0.014487301092594862
    },
    {
      "action_query_batch_std": 0.5435788750648498,
      "action_query_norm": 45.06294326782226,
      "action_relation_loss": 0.01515418291091919,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.0112254443405053,
      "flow_matching_loss": 0.07863241508603096,
      "grad_norm": 0.2508491575717926,
      "h_stock_valid_norm": 100.7697624206543,
      "invalid_tail_samples": 25.8,
      "learning_rate": 9.987210481527855e-05,
      "loss": 0.0938,
      "positive_teacher_pairs": 26400.0,
      "query_h_norm_ratio": 0.4471952706575394,
      "relation_loss": 0.01515418291091919,
      "selected_pairs": 26400.0,
      "step": 2150,
      "student_distance_mean": 35.03802604675293,
      "teacher_distance_mean": 0.8498879849910737,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.01515418291091919,
      "weighted_relation_loss": 0.01515418291091919
    },
    {
      "action_query_batch_std": 0.5469567775726318,
      "action_query_norm": 45.06437454223633,
      "action_relation_loss": 0.014943040441721678,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.020579981290926,
      "flow_matching_loss": 0.07622615247964859,
      "grad_norm": 0.30060070753097534,
      "h_stock_valid_norm": 101.44895401000977,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.98681348975345e-05,
      "loss": 0.0912,
      "positive_teacher_pairs": 26648.1,
      "query_h_norm_ratio": 0.4442099004983902,
      "relation_loss": 0.014943040441721678,
      "selected_pairs": 26648.1,
      "step": 2160,
      "student_distance_mean": 35.36110382080078,
      "teacher_distance_mean": 0.8566796183586121,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.014943040441721678,
      "weighted_relation_loss": 0.014943040441721678
    },
    {
      "action_query_batch_std": 0.528815507888794,
      "action_query_norm": 45.05535659790039,
      "action_relation_loss": 0.015982800628989935,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.029934518241347,
      "flow_matching_loss": 0.07454374432563782,
      "grad_norm": 0.1931244134902954,
      "h_stock_valid_norm": 99.61855010986328,
      "invalid_tail_samples": 22.6,
      "learning_rate": 9.986410438530427e-05,
      "loss": 0.0905,
      "positive_teacher_pairs": 27128.4,
      "query_h_norm_ratio": 0.4523115217685699,
      "relation_loss": 0.015982800628989935,
      "selected_pairs": 27128.4,
      "step": 2170,
      "student_distance_mean": 33.82038536071777,
      "teacher_distance_mean": 0.8485345363616943,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.015982800628989935,
      "weighted_relation_loss": 0.015982800628989935
    },
    {
      "action_query_batch_std": 0.56060009598732,
      "action_query_norm": 45.06503028869629,
      "action_relation_loss": 0.014640658348798751,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.039289055191768,
      "flow_matching_loss": 0.07153118774294853,
      "grad_norm": 0.2748093903064728,
      "h_stock_valid_norm": 96.52616500854492,
      "invalid_tail_samples": 21.9,
      "learning_rate": 9.986001328348534e-05,
      "loss": 0.0862,
      "positive_teacher_pairs": 27291.6,
      "query_h_norm_ratio": 0.46690634191036223,
      "relation_loss": 0.014640658348798751,
      "selected_pairs": 27291.6,
      "step": 2180,
      "student_distance_mean": 36.17037086486816,
      "teacher_distance_mean": 0.8569011330604553,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.014640658348798751,
      "weighted_relation_loss": 0.014640658348798751
    },
    {
      "action_query_batch_std": 0.5284090280532837,
      "action_query_norm": 45.051861953735354,
      "action_relation_loss": 0.014641400519758463,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.048643592142189,
      "flow_matching_loss": 0.06981255300343037,
      "grad_norm": 0.2478969544172287,
      "h_stock_valid_norm": 95.3136993408203,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.985586159704878e-05,
      "loss": 0.0845,
      "positive_teacher_pairs": 26761.1,
      "query_h_norm_ratio": 0.47267302572727204,
      "relation_loss": 0.014641400519758463,
      "selected_pairs": 26761.1,
      "step": 2190,
      "student_distance_mean": 33.87995567321777,
      "teacher_distance_mean": 0.8435621798038483,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.014641400519758463,
      "weighted_relation_loss": 0.014641400519758463
    },
    {
      "action_query_batch_std": 0.5247767210006714,
      "action_query_norm": 45.05009078979492,
      "action_relation_loss": 0.015109460428357124,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.05799812909261,
      "flow_matching_loss": 0.07349278181791305,
      "grad_norm": 0.18224571645259857,
      "h_stock_valid_norm": 95.35381240844727,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.985164933103929e-05,
      "loss": 0.0886,
      "positive_teacher_pairs": 26971.8,
      "query_h_norm_ratio": 0.4724599480628967,
      "relation_loss": 0.015109460428357124,
      "selected_pairs": 26971.8,
      "step": 2200,
      "student_distance_mean": 33.524767303466795,
      "teacher_distance_mean": 0.860341477394104,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.015109460428357124,
      "weighted_relation_loss": 0.015109460428357124
    },
    {
      "action_query_batch_std": 0.5373083770275116,
      "action_query_norm": 45.05346870422363,
      "action_relation_loss": 0.014848451595753431,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.067352666043031,
      "flow_matching_loss": 0.07206239178776741,
      "grad_norm": 0.25980591773986816,
      "h_stock_valid_norm": 96.95889205932617,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.984737649057513e-05,
      "loss": 0.0869,
      "positive_teacher_pairs": 26928.6,
      "query_h_norm_ratio": 0.46472187638282775,
      "relation_loss": 0.014848451595753431,
      "selected_pairs": 26928.6,
      "step": 2210,
      "student_distance_mean": 34.35475730895996,
      "teacher_distance_mean": 0.8448000907897949,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.014848451595753431,
      "weighted_relation_loss": 0.014848451595753431
    },
    {
      "action_query_batch_std": 0.5386975228786468,
      "action_query_norm": 45.0485424041748,
      "action_relation_loss": 0.015268094837665558,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.0767072029934517,
      "flow_matching_loss": 0.07429247051477432,
      "grad_norm": 0.1686604619026184,
      "h_stock_valid_norm": 99.1676643371582,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.984304308084827e-05,
      "loss": 0.0896,
      "positive_teacher_pairs": 26822.9,
      "query_h_norm_ratio": 0.45426989197731016,
      "relation_loss": 0.015268094837665558,
      "selected_pairs": 26822.9,
      "step": 2220,
      "student_distance_mean": 34.73730812072754,
      "teacher_distance_mean": 0.8476587057113647,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.015268094837665558,
      "weighted_relation_loss": 0.015268094837665558
    },
    {
      "action_query_batch_std": 0.5435874462127686,
      "action_query_norm": 45.04932174682617,
      "action_relation_loss": 0.015302989166229963,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.086061739943873,
      "flow_matching_loss": 0.07446924299001693,
      "grad_norm": 0.20004470646381378,
      "h_stock_valid_norm": 100.27709579467773,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.983864910712416e-05,
      "loss": 0.0898,
      "positive_teacher_pairs": 27156.8,
      "query_h_norm_ratio": 0.44925314486026763,
      "relation_loss": 0.015302989166229963,
      "selected_pairs": 27156.8,
      "step": 2230,
      "student_distance_mean": 35.002964401245116,
      "teacher_distance_mean": 0.8543000400066376,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.015302989166229963,
      "weighted_relation_loss": 0.015302989166229963
    },
    {
      "action_query_batch_std": 0.55200954079628,
      "action_query_norm": 45.051137924194336,
      "action_relation_loss": 0.01492713913321495,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.095416276894294,
      "flow_matching_loss": 0.07390820011496543,
      "grad_norm": 0.2376556396484375,
      "h_stock_valid_norm": 99.70981521606446,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.98341945747419e-05,
      "loss": 0.0888,
      "positive_teacher_pairs": 27174.6,
      "query_h_norm_ratio": 0.4518264323472977,
      "relation_loss": 0.01492713913321495,
      "selected_pairs": 27174.6,
      "step": 2240,
      "student_distance_mean": 35.2955680847168,
      "teacher_distance_mean": 0.8554051637649536,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.01492713913321495,
      "weighted_relation_loss": 0.01492713913321495
    },
    {
      "action_query_batch_std": 0.5480197012424469,
      "action_query_norm": 45.04834251403808,
      "action_relation_loss": 0.014596345089375972,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.1047708138447145,
      "flow_matching_loss": 0.07577087581157685,
      "grad_norm": 0.18415221571922302,
      "h_stock_valid_norm": 98.86891403198243,
      "invalid_tail_samples": 22.6,
      "learning_rate": 9.98296794891142e-05,
      "loss": 0.0904,
      "positive_teacher_pairs": 27131.5,
      "query_h_norm_ratio": 0.4556490808725357,
      "relation_loss": 0.014596345089375972,
      "selected_pairs": 27131.5,
      "step": 2250,
      "student_distance_mean": 35.07907943725586,
      "teacher_distance_mean": 0.8553018629550934,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.014596345089375972,
      "weighted_relation_loss": 0.014596345089375972
    },
    {
      "action_query_batch_std": 0.5512958407402039,
      "action_query_norm": 45.044887924194335,
      "action_relation_loss": 0.015190644469112159,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.1141253507951356,
      "flow_matching_loss": 0.07738973349332809,
      "grad_norm": 0.23867417871952057,
      "h_stock_valid_norm": 98.3607307434082,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.982510385572725e-05,
      "loss": 0.0926,
      "positive_teacher_pairs": 26824.4,
      "query_h_norm_ratio": 0.45795924961566925,
      "relation_loss": 0.015190644469112159,
      "selected_pairs": 26824.4,
      "step": 2260,
      "student_distance_mean": 35.43435173034668,
      "teacher_distance_mean": 0.8567930519580841,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.015190644469112159,
      "weighted_relation_loss": 0.015190644469112159
    },
    {
      "action_query_batch_std": 0.5429916679859161,
      "action_query_norm": 45.038685989379886,
      "action_relation_loss": 0.014709832705557346,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.1234798877455567,
      "flow_matching_loss": 0.07727788835763931,
      "grad_norm": 0.28714603185653687,
      "h_stock_valid_norm": 98.11468658447265,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.982046768014094e-05,
      "loss": 0.092,
      "positive_teacher_pairs": 26792.6,
      "query_h_norm_ratio": 0.4590467602014542,
      "relation_loss": 0.014709832705557346,
      "selected_pairs": 26792.6,
      "step": 2270,
      "student_distance_mean": 35.105933380126956,
      "teacher_distance_mean": 0.8519338190555572,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.014709832705557346,
      "weighted_relation_loss": 0.014709832705557346
    },
    {
      "action_query_batch_std": 0.5437731862068176,
      "action_query_norm": 45.03940696716309,
      "action_relation_loss": 0.014049762208014727,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.1328344246959774,
      "flow_matching_loss": 0.07075061053037643,
      "grad_norm": 0.21141524612903595,
      "h_stock_valid_norm": 98.04840545654297,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.981577096798863e-05,
      "loss": 0.0848,
      "positive_teacher_pairs": 26942.0,
      "query_h_norm_ratio": 0.4593623459339142,
      "relation_loss": 0.014049762208014727,
      "selected_pairs": 26942.0,
      "step": 2280,
      "student_distance_mean": 35.010331344604495,
      "teacher_distance_mean": 0.8580209195613862,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.014049762208014727,
      "weighted_relation_loss": 0.014049762208014727
    },
    {
      "action_query_batch_std": 0.5529908835887909,
      "action_query_norm": 45.036359024047854,
      "action_relation_loss": 0.013692011777311563,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.1421889616463985,
      "flow_matching_loss": 0.07084423452615737,
      "grad_norm": 0.19982559978961945,
      "h_stock_valid_norm": 98.29246520996094,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.981101372497727e-05,
      "loss": 0.0845,
      "positive_teacher_pairs": 26789.0,
      "query_h_norm_ratio": 0.45820059478282926,
      "relation_loss": 0.013692011777311563,
      "selected_pairs": 26789.0,
      "step": 2290,
      "student_distance_mean": 35.69244804382324,
      "teacher_distance_mean": 0.860910427570343,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.013692011777311563,
      "weighted_relation_loss": 0.013692011777311563
    },
    {
      "action_query_batch_std": 0.5208597958087922,
      "action_query_norm": 45.02371788024902,
      "action_relation_loss": 0.014187246933579445,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.1515434985968196,
      "flow_matching_loss": 0.07046948745846748,
      "grad_norm": 0.16036678850650787,
      "h_stock_valid_norm": 97.37362365722656,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.980619595688737e-05,
      "loss": 0.0847,
      "positive_teacher_pairs": 26996.1,
      "query_h_norm_ratio": 0.4623838573694229,
      "relation_loss": 0.014187246933579445,
      "selected_pairs": 26996.1,
      "step": 2300,
      "student_distance_mean": 33.49867897033691,
      "teacher_distance_mean": 0.8455359697341919,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.014187246933579445,
      "weighted_relation_loss": 0.014187246933579445
    },
    {
      "action_query_batch_std": 0.5383338034152985,
      "action_query_norm": 45.028872680664065,
      "action_relation_loss": 0.0143289128318429,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.1608980355472402,
      "flow_matching_loss": 0.07227809876203536,
      "grad_norm": 0.25756585597991943,
      "h_stock_valid_norm": 97.97790756225587,
      "invalid_tail_samples": 25.7,
      "learning_rate": 9.980131766957295e-05,
      "loss": 0.0866,
      "positive_teacher_pairs": 26411.7,
      "query_h_norm_ratio": 0.4595878481864929,
      "relation_loss": 0.0143289128318429,
      "selected_pairs": 26411.7,
      "step": 2310,
      "student_distance_mean": 34.6583137512207,
      "teacher_distance_mean": 0.8515179216861725,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0143289128318429,
      "weighted_relation_loss": 0.0143289128318429
    },
    {
      "action_query_batch_std": 0.5357353031635285,
      "action_query_norm": 45.02619171142578,
      "action_relation_loss": 0.014329015277326107,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.1702525724976613,
      "flow_matching_loss": 0.07455218806862832,
      "grad_norm": 0.3551977276802063,
      "h_stock_valid_norm": 98.04889602661133,
      "invalid_tail_samples": 25.5,
      "learning_rate": 9.979637886896163e-05,
      "loss": 0.0889,
      "positive_teacher_pairs": 26457.8,
      "query_h_norm_ratio": 0.45922642946243286,
      "relation_loss": 0.014329015277326107,
      "selected_pairs": 26457.8,
      "step": 2320,
      "student_distance_mean": 34.50885353088379,
      "teacher_distance_mean": 0.8499108672142028,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.014329015277326107,
      "weighted_relation_loss": 0.014329015277326107
    },
    {
      "action_query_batch_std": 0.5360078096389771,
      "action_query_norm": 45.022605895996094,
      "action_relation_loss": 0.014093887247145175,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.1796071094480824,
      "flow_matching_loss": 0.07273503988981248,
      "grad_norm": 0.20847131311893463,
      "h_stock_valid_norm": 97.53401336669921,
      "invalid_tail_samples": 26.3,
      "learning_rate": 9.979137956105447e-05,
      "loss": 0.0868,
      "positive_teacher_pairs": 26272.8,
      "query_h_norm_ratio": 0.461612731218338,
      "relation_loss": 0.014093887247145175,
      "selected_pairs": 26272.8,
      "step": 2330,
      "student_distance_mean": 34.531317138671874,
      "teacher_distance_mean": 0.8552037656307221,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.014093887247145175,
      "weighted_relation_loss": 0.014093887247145175
    },
    {
      "action_query_batch_std": 0.5454417824745178,
      "action_query_norm": 45.02914924621582,
      "action_relation_loss": 0.015127704292535783,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.188961646398503,
      "flow_matching_loss": 0.07375730872154236,
      "grad_norm": 0.17173802852630615,
      "h_stock_valid_norm": 98.09366760253906,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.978631975192613e-05,
      "loss": 0.0889,
      "positive_teacher_pairs": 27174.2,
      "query_h_norm_ratio": 0.45904751420021056,
      "relation_loss": 0.015127704292535783,
      "selected_pairs": 27174.2,
      "step": 2340,
      "student_distance_mean": 35.05769729614258,
      "teacher_distance_mean": 0.8592521190643311,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.015127704292535783,
      "weighted_relation_loss": 0.015127704292535783
    },
    {
      "action_query_batch_std": 0.5466865301132202,
      "action_query_norm": 45.02519226074219,
      "action_relation_loss": 0.014501327835023402,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.198316183348924,
      "flow_matching_loss": 0.06912590377032757,
      "grad_norm": 0.17305950820446014,
      "h_stock_valid_norm": 97.2699691772461,
      "invalid_tail_samples": 22.7,
      "learning_rate": 9.978119944772475e-05,
      "loss": 0.0836,
      "positive_teacher_pairs": 27102.6,
      "query_h_norm_ratio": 0.46290162205696106,
      "relation_loss": 0.014501327835023402,
      "selected_pairs": 27102.6,
      "step": 2350,
      "student_distance_mean": 35.282324981689456,
      "teacher_distance_mean": 0.8545251369476319,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.014501327835023402,
      "weighted_relation_loss": 0.014501327835023402
    },
    {
      "action_query_batch_std": 0.5446608006954193,
      "action_query_norm": 45.021267318725585,
      "action_relation_loss": 0.014652265794575214,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.2076707202993453,
      "flow_matching_loss": 0.07283585593104362,
      "grad_norm": 0.27693402767181396,
      "h_stock_valid_norm": 96.8580810546875,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.977601865467197e-05,
      "loss": 0.0875,
      "positive_teacher_pairs": 26735.8,
      "query_h_norm_ratio": 0.4648212164640427,
      "relation_loss": 0.014652265794575214,
      "selected_pairs": 26735.8,
      "step": 2360,
      "student_distance_mean": 35.07824974060058,
      "teacher_distance_mean": 0.8605015873908997,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.014652265794575214,
      "weighted_relation_loss": 0.014652265794575214
    },
    {
      "action_query_batch_std": 0.5416324019432068,
      "action_query_norm": 45.01958236694336,
      "action_relation_loss": 0.014360902272164822,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.217025257249766,
      "flow_matching_loss": 0.0734777256846428,
      "grad_norm": 0.1938718557357788,
      "h_stock_valid_norm": 97.27743682861328,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.977077737906297e-05,
      "loss": 0.0878,
      "positive_teacher_pairs": 27174.6,
      "query_h_norm_ratio": 0.4627998977899551,
      "relation_loss": 0.014360902272164822,
      "selected_pairs": 27174.6,
      "step": 2370,
      "student_distance_mean": 34.77820472717285,
      "teacher_distance_mean": 0.8627536714076995,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.014360902272164822,
      "weighted_relation_loss": 0.014360902272164822
    },
    {
      "action_query_batch_std": 0.5392952978610992,
      "action_query_norm": 45.01603775024414,
      "action_relation_loss": 0.014391688909381629,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.226379794200187,
      "flow_matching_loss": 0.07211971506476403,
      "grad_norm": 0.20955204963684082,
      "h_stock_valid_norm": 97.81043319702148,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.976547562726636e-05,
      "loss": 0.0865,
      "positive_teacher_pairs": 26759.0,
      "query_h_norm_ratio": 0.4602489650249481,
      "relation_loss": 0.014391688909381629,
      "selected_pairs": 26759.0,
      "step": 2380,
      "student_distance_mean": 34.43628044128418,
      "teacher_distance_mean": 0.8607861697673798,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.014391688909381629,
      "weighted_relation_loss": 0.014391688909381629
    },
    {
      "action_query_batch_std": 0.5320995509624481,
      "action_query_norm": 45.0104305267334,
      "action_relation_loss": 0.013955207634717226,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.235734331150608,
      "flow_matching_loss": 0.06948502138257026,
      "grad_norm": 0.21778610348701477,
      "h_stock_valid_norm": 98.83356018066407,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.976011340572429e-05,
      "loss": 0.0834,
      "positive_teacher_pairs": 27153.0,
      "query_h_norm_ratio": 0.4554196625947952,
      "relation_loss": 0.013955207634717226,
      "selected_pairs": 27153.0,
      "step": 2390,
      "student_distance_mean": 34.3466079711914,
      "teacher_distance_mean": 0.856106412410736,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.013955207634717226,
      "weighted_relation_loss": 0.013955207634717226
    },
    {
      "action_query_batch_std": 0.5398782074451447,
      "action_query_norm": 45.01350631713867,
      "action_relation_loss": 0.014434392470866442,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.245088868101029,
      "flow_matching_loss": 0.07207055538892745,
      "grad_norm": 0.18930421769618988,
      "h_stock_valid_norm": 98.14362030029297,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.975469072095237e-05,
      "loss": 0.0865,
      "positive_teacher_pairs": 26781.3,
      "query_h_norm_ratio": 0.4586562305688858,
      "relation_loss": 0.014434392470866442,
      "selected_pairs": 26781.3,
      "step": 2400,
      "student_distance_mean": 34.87620964050293,
      "teacher_distance_mean": 0.8586649656295776,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.014434392470866442,
      "weighted_relation_loss": 0.014434392470866442
    },
    {
      "action_query_batch_std": 0.5308482229709626,
      "action_query_norm": 45.010444259643556,
      "action_relation_loss": 0.013383717089891434,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.25444340505145,
      "flow_matching_loss": 0.07131172120571136,
      "grad_norm": 0.23785732686519623,
      "h_stock_valid_norm": 98.28689651489258,
      "invalid_tail_samples": 25.6,
      "learning_rate": 9.974920757953965e-05,
      "loss": 0.0847,
      "positive_teacher_pairs": 26433.6,
      "query_h_norm_ratio": 0.45795606076717377,
      "relation_loss": 0.013383717089891434,
      "selected_pairs": 26433.6,
      "step": 2410,
      "student_distance_mean": 34.01305656433105,
      "teacher_distance_mean": 0.8620654046535492,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.013383717089891434,
      "weighted_relation_loss": 0.013383717089891434
    },
    {
      "action_query_batch_std": 0.5294762551784515,
      "action_query_norm": 45.01093940734863,
      "action_relation_loss": 0.01342815551906824,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.263797942001871,
      "flow_matching_loss": 0.07056865356862545,
      "grad_norm": 0.2501314878463745,
      "h_stock_valid_norm": 99.02057876586915,
      "invalid_tail_samples": 22.9,
      "learning_rate": 9.97436639881487e-05,
      "loss": 0.084,
      "positive_teacher_pairs": 27059.9,
      "query_h_norm_ratio": 0.4545630842447281,
      "relation_loss": 0.01342815551906824,
      "selected_pairs": 27059.9,
      "step": 2420,
      "student_distance_mean": 34.08959274291992,
      "teacher_distance_mean": 0.864580774307251,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.01342815551906824,
      "weighted_relation_loss": 0.01342815551906824
    },
    {
      "action_query_batch_std": 0.5270541548728943,
      "action_query_norm": 45.00642318725586,
      "action_relation_loss": 0.014650676678866149,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.2731524789522917,
      "flow_matching_loss": 0.07152440398931503,
      "grad_norm": 0.15286631882190704,
      "h_stock_valid_norm": 98.5124526977539,
      "invalid_tail_samples": 25.2,
      "learning_rate": 9.973805995351545e-05,
      "loss": 0.0862,
      "positive_teacher_pairs": 26532.1,
      "query_h_norm_ratio": 0.4568616896867752,
      "relation_loss": 0.014650676678866149,
      "selected_pairs": 26532.1,
      "step": 2430,
      "student_distance_mean": 33.8962345123291,
      "teacher_distance_mean": 0.8517261981964112,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.014650676678866149,
      "weighted_relation_loss": 0.014650676678866149
    },
    {
      "action_query_batch_std": 0.519207113981247,
      "action_query_norm": 45.00374526977539,
      "action_relation_loss": 0.01400346476584673,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.2825070159027128,
      "flow_matching_loss": 0.07323931530117989,
      "grad_norm": 0.1624373346567154,
      "h_stock_valid_norm": 98.63621673583984,
      "invalid_tail_samples": 25.3,
      "learning_rate": 9.973239548244939e-05,
      "loss": 0.0872,
      "positive_teacher_pairs": 26501.9,
      "query_h_norm_ratio": 0.45626198053359984,
      "relation_loss": 0.01400346476584673,
      "selected_pairs": 26501.9,
      "step": 2440,
      "student_distance_mean": 33.41511840820313,
      "teacher_distance_mean": 0.8615609765052795,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.01400346476584673,
      "weighted_relation_loss": 0.01400346476584673
    },
    {
      "action_query_batch_std": 0.5323634088039398,
      "action_query_norm": 45.00724487304687,
      "action_relation_loss": 0.014488351345062257,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.291861552853134,
      "flow_matching_loss": 0.07241590917110444,
      "grad_norm": 0.17795610427856445,
      "h_stock_valid_norm": 98.02946243286132,
      "invalid_tail_samples": 26.0,
      "learning_rate": 9.972667058183333e-05,
      "loss": 0.0869,
      "positive_teacher_pairs": 26342.6,
      "query_h_norm_ratio": 0.4591405987739563,
      "relation_loss": 0.014488351345062257,
      "selected_pairs": 26342.6,
      "step": 2450,
      "student_distance_mean": 34.22012825012207,
      "teacher_distance_mean": 0.8545688092708588,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.014488351345062257,
      "weighted_relation_loss": 0.014488351345062257
    },
    {
      "action_query_batch_std": 0.5276206195354461,
      "action_query_norm": 45.005852127075194,
      "action_relation_loss": 0.014730379730463029,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.301216089803555,
      "flow_matching_loss": 0.07010644413530827,
      "grad_norm": 0.2336643487215042,
      "h_stock_valid_norm": 96.61166763305664,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.972088525862362e-05,
      "loss": 0.0848,
      "positive_teacher_pairs": 26660.3,
      "query_h_norm_ratio": 0.4658482939004898,
      "relation_loss": 0.014730379730463029,
      "selected_pairs": 26660.3,
      "step": 2460,
      "student_distance_mean": 34.09126014709473,
      "teacher_distance_mean": 0.8584686934947967,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.014730379730463029,
      "weighted_relation_loss": 0.014730379730463029
    },
    {
      "action_query_batch_std": 0.5027299165725708,
      "action_query_norm": 44.98799705505371,
      "action_relation_loss": 0.014680150523781776,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.3105706267539756,
      "flow_matching_loss": 0.0720736425369978,
      "grad_norm": 0.2336701601743698,
      "h_stock_valid_norm": 95.64825286865235,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.971503951984995e-05,
      "loss": 0.0868,
      "positive_teacher_pairs": 26997.4,
      "query_h_norm_ratio": 0.47035124003887174,
      "relation_loss": 0.014680150523781776,
      "selected_pairs": 26997.4,
      "step": 2470,
      "student_distance_mean": 32.293818664550784,
      "teacher_distance_mean": 0.8527155101299286,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.014680150523781776,
      "weighted_relation_loss": 0.014680150523781776
    },
    {
      "action_query_batch_std": 0.5223251640796661,
      "action_query_norm": 44.99622650146485,
      "action_relation_loss": 0.014527548290789127,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.3199251637043967,
      "flow_matching_loss": 0.07298678867518901,
      "grad_norm": 0.26978567242622375,
      "h_stock_valid_norm": 96.5224739074707,
      "invalid_tail_samples": 26.1,
      "learning_rate": 9.970913337261543e-05,
      "loss": 0.0875,
      "positive_teacher_pairs": 26327.1,
      "query_h_norm_ratio": 0.46617652773857116,
      "relation_loss": 0.014527548290789127,
      "selected_pairs": 26327.1,
      "step": 2480,
      "student_distance_mean": 33.60118465423584,
      "teacher_distance_mean": 0.8549225211143494,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.014527548290789127,
      "weighted_relation_loss": 0.014527548290789127
    },
    {
      "action_query_batch_std": 0.5402735054492951,
      "action_query_norm": 44.9967758178711,
      "action_relation_loss": 0.014761329162865877,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.3292797006548174,
      "flow_matching_loss": 0.0715439073741436,
      "grad_norm": 0.17770755290985107,
      "h_stock_valid_norm": 96.12255935668945,
      "invalid_tail_samples": 25.1,
      "learning_rate": 9.97031668240966e-05,
      "loss": 0.0863,
      "positive_teacher_pairs": 26544.0,
      "query_h_norm_ratio": 0.46812055706977845,
      "relation_loss": 0.014761329162865877,
      "selected_pairs": 26544.0,
      "step": 2490,
      "student_distance_mean": 34.866044998168945,
      "teacher_distance_mean": 0.8507690489292145,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.014761329162865877,
      "weighted_relation_loss": 0.014761329162865877
    },
    {
      "action_query_batch_std": 0.5191121101379395,
      "action_query_norm": 44.99186973571777,
      "action_relation_loss": 0.014867476746439933,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.3386342376052385,
      "flow_matching_loss": 0.07028223127126694,
      "grad_norm": 0.18717871606349945,
      "h_stock_valid_norm": 96.60944213867188,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.969713988154339e-05,
      "loss": 0.0851,
      "positive_teacher_pairs": 26850.5,
      "query_h_norm_ratio": 0.4657116085290909,
      "relation_loss": 0.014867476746439933,
      "selected_pairs": 26850.5,
      "step": 2500,
      "student_distance_mean": 33.40879516601562,
      "teacher_distance_mean": 0.846074104309082,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.014867476746439933,
      "weighted_relation_loss": 0.014867476746439933
    },
    {
      "action_query_batch_std": 0.5263739764690399,
      "action_query_norm": 44.994881057739256,
      "action_relation_loss": 0.014573226310312748,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.3479887745556596,
      "flow_matching_loss": 0.06812541037797928,
      "grad_norm": 0.17788831889629364,
      "h_stock_valid_norm": 96.50085754394532,
      "invalid_tail_samples": 21.9,
      "learning_rate": 9.969105255227906e-05,
      "loss": 0.0827,
      "positive_teacher_pairs": 27292.8,
      "query_h_norm_ratio": 0.4662674754858017,
      "relation_loss": 0.014573226310312748,
      "selected_pairs": 27292.8,
      "step": 2510,
      "student_distance_mean": 33.582461547851565,
      "teacher_distance_mean": 0.8498829066753387,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.014573226310312748,
      "weighted_relation_loss": 0.014573226310312748
    },
    {
      "action_query_batch_std": 0.5282586753368378,
      "action_query_norm": 44.99141311645508,
      "action_relation_loss": 0.014456233195960521,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.3573433115060807,
      "flow_matching_loss": 0.06819333024322986,
      "grad_norm": 0.226392462849617,
      "h_stock_valid_norm": 97.6255012512207,
      "invalid_tail_samples": 26.4,
      "learning_rate": 9.968490484370035e-05,
      "loss": 0.0826,
      "positive_teacher_pairs": 26249.9,
      "query_h_norm_ratio": 0.4608683466911316,
      "relation_loss": 0.014456233195960521,
      "selected_pairs": 26249.9,
      "step": 2520,
      "student_distance_mean": 34.02829818725586,
      "teacher_distance_mean": 0.8544618487358093,
      "valid_action_samples": 229.6,
      "weighted_auxiliary_loss": 0.014456233195960521,
      "weighted_relation_loss": 0.014456233195960521
    },
    {
      "action_query_batch_std": 0.5191096425056457,
      "action_query_norm": 44.98422470092773,
      "action_relation_loss": 0.014104842115193606,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.3666978484565013,
      "flow_matching_loss": 0.07230008691549301,
      "grad_norm": 0.19956322014331818,
      "h_stock_valid_norm": 98.6590347290039,
      "invalid_tail_samples": 23.0,
      "learning_rate": 9.967869676327726e-05,
      "loss": 0.0864,
      "positive_teacher_pairs": 27034.6,
      "query_h_norm_ratio": 0.4559669613838196,
      "relation_loss": 0.014104842115193606,
      "selected_pairs": 27034.6,
      "step": 2530,
      "student_distance_mean": 33.567423248291014,
      "teacher_distance_mean": 0.8492948114871979,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.014104842115193606,
      "weighted_relation_loss": 0.014104842115193606
    },
    {
      "action_query_batch_std": 0.5326960980892181,
      "action_query_norm": 44.99080810546875,
      "action_relation_loss": 0.014007226470857859,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.3760523854069224,
      "flow_matching_loss": 0.06649584695696831,
      "grad_norm": 0.24630317091941833,
      "h_stock_valid_norm": 99.55000915527344,
      "invalid_tail_samples": 23.1,
      "learning_rate": 9.96724283185532e-05,
      "loss": 0.0805,
      "positive_teacher_pairs": 27012.2,
      "query_h_norm_ratio": 0.4519439786672592,
      "relation_loss": 0.014007226470857859,
      "selected_pairs": 27012.2,
      "step": 2540,
      "student_distance_mean": 34.29173469543457,
      "teacher_distance_mean": 0.8606425881385803,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.014007226470857859,
      "weighted_relation_loss": 0.014007226470857859
    },
    {
      "action_query_batch_std": 0.5254208028316498,
      "action_query_norm": 44.98701171875,
      "action_relation_loss": 0.014120172429829835,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.385406922357343,
      "flow_matching_loss": 0.07001191824674606,
      "grad_norm": 0.2142256796360016,
      "h_stock_valid_norm": 99.41521530151367,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.966609951714494e-05,
      "loss": 0.0841,
      "positive_teacher_pairs": 27196.4,
      "query_h_norm_ratio": 0.4525209963321686,
      "relation_loss": 0.014120172429829835,
      "selected_pairs": 27196.4,
      "step": 2550,
      "student_distance_mean": 33.910404968261716,
      "teacher_distance_mean": 0.8591488540172577,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.014120172429829835,
      "weighted_relation_loss": 0.014120172429829835
    },
    {
      "action_query_batch_std": 0.5150444030761718,
      "action_query_norm": 44.98327865600586,
      "action_relation_loss": 0.01365848546847701,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.394761459307764,
      "flow_matching_loss": 0.06742743477225303,
      "grad_norm": 0.12880772352218628,
      "h_stock_valid_norm": 98.40343170166015,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.965971036674255e-05,
      "loss": 0.0811,
      "positive_teacher_pairs": 27084.3,
      "query_h_norm_ratio": 0.4571425884962082,
      "relation_loss": 0.01365848546847701,
      "selected_pairs": 27084.3,
      "step": 2560,
      "student_distance_mean": 33.21008949279785,
      "teacher_distance_mean": 0.8535837709903717,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.01365848546847701,
      "weighted_relation_loss": 0.01365848546847701
    },
    {
      "action_query_batch_std": 0.5253719985485077,
      "action_query_norm": 44.98601951599121,
      "action_relation_loss": 0.01449537929147482,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.4041159962581853,
      "flow_matching_loss": 0.07007079310715199,
      "grad_norm": 0.18523496389389038,
      "h_stock_valid_norm": 98.11652908325195,
      "invalid_tail_samples": 25.4,
      "learning_rate": 9.965326087510947e-05,
      "loss": 0.0846,
      "positive_teacher_pairs": 26484.4,
      "query_h_norm_ratio": 0.45849876701831815,
      "relation_loss": 0.01449537929147482,
      "selected_pairs": 26484.4,
      "step": 2570,
      "student_distance_mean": 33.983730697631835,
      "teacher_distance_mean": 0.8589366555213929,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.01449537929147482,
      "weighted_relation_loss": 0.01449537929147482
    },
    {
      "action_query_batch_std": 0.5276558101177216,
      "action_query_norm": 44.98659057617188,
      "action_relation_loss": 0.013975481409579516,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.4134705332086064,
      "flow_matching_loss": 0.07112245634198189,
      "grad_norm": 0.239820197224617,
      "h_stock_valid_norm": 98.65158309936524,
      "invalid_tail_samples": 21.9,
      "learning_rate": 9.964675105008244e-05,
      "loss": 0.0851,
      "positive_teacher_pairs": 27292.2,
      "query_h_norm_ratio": 0.4560199022293091,
      "relation_loss": 0.013975481409579516,
      "selected_pairs": 27292.2,
      "step": 2580,
      "student_distance_mean": 33.95537490844727,
      "teacher_distance_mean": 0.8554266571998597,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.013975481409579516,
      "weighted_relation_loss": 0.013975481409579516
    },
    {
      "action_query_batch_std": 0.5123036235570908,
      "action_query_norm": 44.975474166870114,
      "action_relation_loss": 0.01460400763899088,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.422825070159027,
      "flow_matching_loss": 0.06996185183525086,
      "grad_norm": 0.17964965105056763,
      "h_stock_valid_norm": 98.72765197753907,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.964018089957147e-05,
      "loss": 0.0846,
      "positive_teacher_pairs": 26940.8,
      "query_h_norm_ratio": 0.45555699467658994,
      "relation_loss": 0.01460400763899088,
      "selected_pairs": 26940.8,
      "step": 2590,
      "student_distance_mean": 33.04977684020996,
      "teacher_distance_mean": 0.856326824426651,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.01460400763899088,
      "weighted_relation_loss": 0.01460400763899088
    },
    {
      "action_query_batch_std": 0.5391653060913086,
      "action_query_norm": 44.98270378112793,
      "action_relation_loss": 0.013551090750843287,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.432179607109448,
      "flow_matching_loss": 0.07102518305182456,
      "grad_norm": 0.17319336533546448,
      "h_stock_valid_norm": 98.76160354614258,
      "invalid_tail_samples": 21.0,
      "learning_rate": 9.963355043155996e-05,
      "loss": 0.0846,
      "positive_teacher_pairs": 27510.0,
      "query_h_norm_ratio": 0.4554703742265701,
      "relation_loss": 0.013551090750843287,
      "selected_pairs": 27510.0,
      "step": 2600,
      "student_distance_mean": 34.87480926513672,
      "teacher_distance_mean": 0.8668508231639862,
      "valid_action_samples": 235.0,
      "weighted_auxiliary_loss": 0.013551090750843287,
      "weighted_relation_loss": 0.013551090750843287
    },
    {
      "action_query_batch_std": 0.5213906943798066,
      "action_query_norm": 44.9705623626709,
      "action_relation_loss": 0.014479231741279363,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.441534144059869,
      "flow_matching_loss": 0.06955143883824348,
      "grad_norm": 0.1658855527639389,
      "h_stock_valid_norm": 97.66883773803711,
      "invalid_tail_samples": 23.0,
      "learning_rate": 9.962685965410455e-05,
      "loss": 0.084,
      "positive_teacher_pairs": 27034.6,
      "query_h_norm_ratio": 0.46044660210609434,
      "relation_loss": 0.014479231741279363,
      "selected_pairs": 27034.6,
      "step": 2610,
      "student_distance_mean": 33.554771995544435,
      "teacher_distance_mean": 0.8477510035037994,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.014479231741279363,
      "weighted_relation_loss": 0.014479231741279363
    },
    {
      "action_query_batch_std": 0.5237074375152588,
      "action_query_norm": 44.97193565368652,
      "action_relation_loss": 0.014338617213070393,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.45088868101029,
      "flow_matching_loss": 0.07152433916926385,
      "grad_norm": 0.13926953077316284,
      "h_stock_valid_norm": 97.8393669128418,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.962010857533514e-05,
      "loss": 0.0859,
      "positive_teacher_pairs": 27198.1,
      "query_h_norm_ratio": 0.4596529334783554,
      "relation_loss": 0.014338617213070393,
      "selected_pairs": 27198.1,
      "step": 2620,
      "student_distance_mean": 33.653171920776366,
      "teacher_distance_mean": 0.8506654500961304,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.014338617213070393,
      "weighted_relation_loss": 0.014338617213070393
    },
    {
      "action_query_batch_std": 0.5344342291355133,
      "action_query_norm": 44.97263145446777,
      "action_relation_loss": 0.014859134890139103,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.460243217960711,
      "flow_matching_loss": 0.0733422040939331,
      "grad_norm": 0.19333143532276154,
      "h_stock_valid_norm": 98.36821060180664,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.961329720345493e-05,
      "loss": 0.0882,
      "positive_teacher_pairs": 26640.0,
      "query_h_norm_ratio": 0.45719471871852874,
      "relation_loss": 0.014859134890139103,
      "selected_pairs": 26640.0,
      "step": 2630,
      "student_distance_mean": 34.4285774230957,
      "teacher_distance_mean": 0.85375816822052,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.014859134890139103,
      "weighted_relation_loss": 0.014859134890139103
    },
    {
      "action_query_batch_std": 0.5203177750110626,
      "action_query_norm": 44.96912117004395,
      "action_relation_loss": 0.013928186241537332,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.469597754911132,
      "flow_matching_loss": 0.0716071866452694,
      "grad_norm": 0.183080792427063,
      "h_stock_valid_norm": 97.84042739868164,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.96064255467404e-05,
      "loss": 0.0855,
      "positive_teacher_pairs": 27201.6,
      "query_h_norm_ratio": 0.4596253246068954,
      "relation_loss": 0.013928186241537332,
      "selected_pairs": 27201.6,
      "step": 2640,
      "student_distance_mean": 33.53423957824707,
      "teacher_distance_mean": 0.85857293009758,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.013928186241537332,
      "weighted_relation_loss": 0.013928186241537332
    },
    {
      "action_query_batch_std": 0.5151338338851928,
      "action_query_norm": 44.965732192993165,
      "action_relation_loss": 0.013730292953550816,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.4789522918615527,
      "flow_matching_loss": 0.0701284408569336,
      "grad_norm": 0.20941922068595886,
      "h_stock_valid_norm": 96.92270050048828,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.959949361354126e-05,
      "loss": 0.0839,
      "positive_teacher_pairs": 27217.7,
      "query_h_norm_ratio": 0.46394130289554597,
      "relation_loss": 0.013730292953550816,
      "selected_pairs": 27217.7,
      "step": 2650,
      "student_distance_mean": 33.17489356994629,
      "teacher_distance_mean": 0.8559734761714936,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.013730292953550816,
      "weighted_relation_loss": 0.013730292953550816
    },
    {
      "action_query_batch_std": 0.5236985146999359,
      "action_query_norm": 44.969153213500974,
      "action_relation_loss": 0.01456519030034542,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.488306828811974,
      "flow_matching_loss": 0.07127912566065789,
      "grad_norm": 0.19809621572494507,
      "h_stock_valid_norm": 97.44062194824218,
      "invalid_tail_samples": 26.3,
      "learning_rate": 9.959250141228045e-05,
      "loss": 0.0858,
      "positive_teacher_pairs": 26270.9,
      "query_h_norm_ratio": 0.46151495575904844,
      "relation_loss": 0.01456519030034542,
      "selected_pairs": 26270.9,
      "step": 2660,
      "student_distance_mean": 33.82782974243164,
      "teacher_distance_mean": 0.861570131778717,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.01456519030034542,
      "weighted_relation_loss": 0.01456519030034542
    },
    {
      "action_query_batch_std": 0.5230765700340271,
      "action_query_norm": 44.96651000976563,
      "action_relation_loss": 0.013795719761401416,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.497661365762395,
      "flow_matching_loss": 0.07038623318076134,
      "grad_norm": 0.18311887979507446,
      "h_stock_valid_norm": 97.80803680419922,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.958544895145414e-05,
      "loss": 0.0842,
      "positive_teacher_pairs": 26920.5,
      "query_h_norm_ratio": 0.4597461402416229,
      "relation_loss": 0.013795719761401416,
      "selected_pairs": 26920.5,
      "step": 2670,
      "student_distance_mean": 33.876610946655276,
      "teacher_distance_mean": 0.8573259472846985,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.013795719761401416,
      "weighted_relation_loss": 0.013795719761401416
    },
    {
      "action_query_batch_std": 0.5221839249134064,
      "action_query_norm": 44.959893798828126,
      "action_relation_loss": 0.01413703178986907,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.5070159027128156,
      "flow_matching_loss": 0.06837076023221016,
      "grad_norm": 0.15198782086372375,
      "h_stock_valid_norm": 97.34588317871093,
      "invalid_tail_samples": 22.1,
      "learning_rate": 9.957833623963177e-05,
      "loss": 0.0825,
      "positive_teacher_pairs": 27243.8,
      "query_h_norm_ratio": 0.4618603765964508,
      "relation_loss": 0.01413703178986907,
      "selected_pairs": 27243.8,
      "step": 2680,
      "student_distance_mean": 33.72426071166992,
      "teacher_distance_mean": 0.8494499027729034,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.01413703178986907,
      "weighted_relation_loss": 0.01413703178986907
    },
    {
      "action_query_batch_std": 0.5269406974315644,
      "action_query_norm": 44.96151161193848,
      "action_relation_loss": 0.013281045760959388,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.5163704396632367,
      "flow_matching_loss": 0.07025043219327927,
      "grad_norm": 0.19246789813041687,
      "h_stock_valid_norm": 97.76280288696289,
      "invalid_tail_samples": 21.3,
      "learning_rate": 9.957116328545593e-05,
      "loss": 0.0835,
      "positive_teacher_pairs": 27427.9,
      "query_h_norm_ratio": 0.45990603864192964,
      "relation_loss": 0.013281045760959388,
      "selected_pairs": 27427.9,
      "step": 2690,
      "student_distance_mean": 34.06693649291992,
      "teacher_distance_mean": 0.8611037373542786,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.013281045760959388,
      "weighted_relation_loss": 0.013281045760959388
    },
    {
      "action_query_batch_std": 0.5269034683704377,
      "action_query_norm": 44.95949440002441,
      "action_relation_loss": 0.01376337744295597,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.525724976613658,
      "flow_matching_loss": 0.06853741630911828,
      "grad_norm": 0.18575628101825714,
      "h_stock_valid_norm": 98.27131805419921,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.956393009764244e-05,
      "loss": 0.0823,
      "positive_teacher_pairs": 26687.0,
      "query_h_norm_ratio": 0.4575066417455673,
      "relation_loss": 0.01376337744295597,
      "selected_pairs": 26687.0,
      "step": 2700,
      "student_distance_mean": 33.9832260131836,
      "teacher_distance_mean": 0.852560305595398,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.01376337744295597,
      "weighted_relation_loss": 0.01376337744295597
    },
    {
      "action_query_batch_std": 0.5272536635398865,
      "action_query_norm": 44.960703659057614,
      "action_relation_loss": 0.013280925061553716,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.5350795135640785,
      "flow_matching_loss": 0.06857458092272281,
      "grad_norm": 0.13227421045303345,
      "h_stock_valid_norm": 98.576123046875,
      "invalid_tail_samples": 21.9,
      "learning_rate": 9.955663668498032e-05,
      "loss": 0.0819,
      "positive_teacher_pairs": 27298.0,
      "query_h_norm_ratio": 0.456105899810791,
      "relation_loss": 0.013280925061553716,
      "selected_pairs": 27298.0,
      "step": 2710,
      "student_distance_mean": 34.066281127929685,
      "teacher_distance_mean": 0.8563046932220459,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.013280925061553716,
      "weighted_relation_loss": 0.013280925061553716
    },
    {
      "action_query_batch_std": 0.5227085590362549,
      "action_query_norm": 44.95834732055664,
      "action_relation_loss": 0.013453980442136526,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.5444340505144996,
      "flow_matching_loss": 0.07017523720860482,
      "grad_norm": 0.15004168450832367,
      "h_stock_valid_norm": 97.4652946472168,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.954928305633173e-05,
      "loss": 0.0836,
      "positive_teacher_pairs": 26898.4,
      "query_h_norm_ratio": 0.4612855941057205,
      "relation_loss": 0.013453980442136526,
      "selected_pairs": 26898.4,
      "step": 2720,
      "student_distance_mean": 33.717720413208006,
      "teacher_distance_mean": 0.8555899739265442,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.013453980442136526,
      "weighted_relation_loss": 0.013453980442136526
    },
    {
      "action_query_batch_std": 0.5311644852161408,
      "action_query_norm": 44.95956115722656,
      "action_relation_loss": 0.013897085189819336,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.55378858746492,
      "flow_matching_loss": 0.06648710556328297,
      "grad_norm": 0.1290183961391449,
      "h_stock_valid_norm": 96.56250228881837,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.954186922063204e-05,
      "loss": 0.0804,
      "positive_teacher_pairs": 26890.7,
      "query_h_norm_ratio": 0.4656039416790009,
      "relation_loss": 0.013897085189819336,
      "selected_pairs": 26890.7,
      "step": 2730,
      "student_distance_mean": 34.42406387329102,
      "teacher_distance_mean": 0.8625764906406402,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.013897085189819336,
      "weighted_relation_loss": 0.013897085189819336
    },
    {
      "action_query_batch_std": 0.5258657187223434,
      "action_query_norm": 44.95692825317383,
      "action_relation_loss": 0.01387627087533474,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.5631431244153413,
      "flow_matching_loss": 0.06873654387891293,
      "grad_norm": 0.18585117161273956,
      "h_stock_valid_norm": 97.44204177856446,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.953439518688974e-05,
      "loss": 0.0826,
      "positive_teacher_pairs": 26826.6,
      "query_h_norm_ratio": 0.4613738596439362,
      "relation_loss": 0.01387627087533474,
      "selected_pairs": 26826.6,
      "step": 2740,
      "student_distance_mean": 33.77003555297851,
      "teacher_distance_mean": 0.8528412163257599,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.01387627087533474,
      "weighted_relation_loss": 0.01387627087533474
    },
    {
      "action_query_batch_std": 0.517137485742569,
      "action_query_norm": 44.95158042907715,
      "action_relation_loss": 0.01414315365254879,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.5724976613657624,
      "flow_matching_loss": 0.07002858780324459,
      "grad_norm": 0.16255661845207214,
      "h_stock_valid_norm": 98.12372817993165,
      "invalid_tail_samples": 25.6,
      "learning_rate": 9.952686096418652e-05,
      "loss": 0.0842,
      "positive_teacher_pairs": 26439.8,
      "query_h_norm_ratio": 0.458116689324379,
      "relation_loss": 0.01414315365254879,
      "selected_pairs": 26439.8,
      "step": 2750,
      "student_distance_mean": 33.47667388916015,
      "teacher_distance_mean": 0.8604168832302094,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.01414315365254879,
      "weighted_relation_loss": 0.01414315365254879
    },
    {
      "action_query_batch_std": 0.5065968602895736,
      "action_query_norm": 44.94336204528808,
      "action_relation_loss": 0.013660945557057857,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.5818521983161835,
      "flow_matching_loss": 0.0689772304147482,
      "grad_norm": 0.2259402871131897,
      "h_stock_valid_norm": 97.96034545898438,
      "invalid_tail_samples": 26.6,
      "learning_rate": 9.951926656167714e-05,
      "loss": 0.0826,
      "positive_teacher_pairs": 26209.9,
      "query_h_norm_ratio": 0.458796015381813,
      "relation_loss": 0.013660945557057857,
      "selected_pairs": 26209.9,
      "step": 2760,
      "student_distance_mean": 32.76034240722656,
      "teacher_distance_mean": 0.8552681982517243,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.013660945557057857,
      "weighted_relation_loss": 0.013660945557057857
    },
    {
      "action_query_batch_std": 0.5116191446781159,
      "action_query_norm": 44.94525413513183,
      "action_relation_loss": 0.013766745012253523,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.591206735266604,
      "flow_matching_loss": 0.06819511726498603,
      "grad_norm": 0.2295798659324646,
      "h_stock_valid_norm": 97.80144500732422,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.951161198858953e-05,
      "loss": 0.082,
      "positive_teacher_pairs": 27171.9,
      "query_h_norm_ratio": 0.45955912470817567,
      "relation_loss": 0.013766745012253523,
      "selected_pairs": 27171.9,
      "step": 2770,
      "student_distance_mean": 33.05007190704346,
      "teacher_distance_mean": 0.854138720035553,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.013766745012253523,
      "weighted_relation_loss": 0.013766745012253523
    },
    {
      "action_query_batch_std": 0.51794154047966,
      "action_query_norm": 44.94400901794434,
      "action_relation_loss": 0.014027019962668419,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.6005612722170253,
      "flow_matching_loss": 0.07007392942905426,
      "grad_norm": 0.19671311974525452,
      "h_stock_valid_norm": 98.08050308227538,
      "invalid_tail_samples": 25.7,
      "learning_rate": 9.950389725422471e-05,
      "loss": 0.0841,
      "positive_teacher_pairs": 26412.3,
      "query_h_norm_ratio": 0.4582381367683411,
      "relation_loss": 0.014027019962668419,
      "selected_pairs": 26412.3,
      "step": 2780,
      "student_distance_mean": 33.51111068725586,
      "teacher_distance_mean": 0.8623916268348694,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.014027019962668419,
      "weighted_relation_loss": 0.014027019962668419
    },
    {
      "action_query_batch_std": 0.5139567434787751,
      "action_query_norm": 44.9402889251709,
      "action_relation_loss": 0.013725713081657886,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.6099158091674464,
      "flow_matching_loss": 0.06867646798491478,
      "grad_norm": 0.1644914448261261,
      "h_stock_valid_norm": 97.78435440063477,
      "invalid_tail_samples": 25.4,
      "learning_rate": 9.949612236795682e-05,
      "loss": 0.0824,
      "positive_teacher_pairs": 26475.2,
      "query_h_norm_ratio": 0.45958923995494844,
      "relation_loss": 0.013725713081657886,
      "selected_pairs": 26475.2,
      "step": 2790,
      "student_distance_mean": 33.30557708740234,
      "teacher_distance_mean": 0.8532963216304779,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.013725713081657886,
      "weighted_relation_loss": 0.013725713081657886
    },
    {
      "action_query_batch_std": 0.5146173924207688,
      "action_query_norm": 44.93969993591308,
      "action_relation_loss": 0.013578308839350938,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.619270346117867,
      "flow_matching_loss": 0.07056206166744232,
      "grad_norm": 0.1883975714445114,
      "h_stock_valid_norm": 98.44184265136718,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.948828733923305e-05,
      "loss": 0.0841,
      "positive_teacher_pairs": 26981.5,
      "query_h_norm_ratio": 0.45651257038116455,
      "relation_loss": 0.013578308839350938,
      "selected_pairs": 26981.5,
      "step": 2800,
      "student_distance_mean": 33.139149856567386,
      "teacher_distance_mean": 0.8529873847961426,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.013578308839350938,
      "weighted_relation_loss": 0.013578308839350938
    },
    {
      "action_query_batch_std": 0.5054312944412231,
      "action_query_norm": 44.93507461547851,
      "action_relation_loss": 0.014011885412037373,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.628624883068288,
      "flow_matching_loss": 0.07016324028372764,
      "grad_norm": 0.16687516868114471,
      "h_stock_valid_norm": 98.15010299682618,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.948039217757374e-05,
      "loss": 0.0842,
      "positive_teacher_pairs": 26777.2,
      "query_h_norm_ratio": 0.45782383978366853,
      "relation_loss": 0.014011885412037373,
      "selected_pairs": 26777.2,
      "step": 2810,
      "student_distance_mean": 32.543016815185545,
      "teacher_distance_mean": 0.8505622625350953,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.014011885412037373,
      "weighted_relation_loss": 0.014011885412037373
    },
    {
      "action_query_batch_std": 0.5107902348041534,
      "action_query_norm": 44.9363094329834,
      "action_relation_loss": 0.013102278672158718,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.637979420018709,
      "flow_matching_loss": 0.06703254617750645,
      "grad_norm": 0.13639071583747864,
      "h_stock_valid_norm": 99.22295989990235,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.947243689257225e-05,
      "loss": 0.0801,
      "positive_teacher_pairs": 26851.5,
      "query_h_norm_ratio": 0.45288575887680055,
      "relation_loss": 0.013102278672158718,
      "selected_pairs": 26851.5,
      "step": 2820,
      "student_distance_mean": 33.029852867126465,
      "teacher_distance_mean": 0.842516976594925,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.013102278672158718,
      "weighted_relation_loss": 0.013102278672158718
    },
    {
      "action_query_batch_std": 0.5032577961683273,
      "action_query_norm": 44.93341255187988,
      "action_relation_loss": 0.01320116026327014,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.64733395696913,
      "flow_matching_loss": 0.06614294312894345,
      "grad_norm": 0.18125447630882263,
      "h_stock_valid_norm": 99.94681091308594,
      "invalid_tail_samples": 22.7,
      "learning_rate": 9.946442149389497e-05,
      "loss": 0.0793,
      "positive_teacher_pairs": 27103.5,
      "query_h_norm_ratio": 0.44957624971866605,
      "relation_loss": 0.01320116026327014,
      "selected_pairs": 27103.5,
      "step": 2830,
      "student_distance_mean": 32.423850440979,
      "teacher_distance_mean": 0.8510023772716522,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.01320116026327014,
      "weighted_relation_loss": 0.01320116026327014
    },
    {
      "action_query_batch_std": 0.5263487160205841,
      "action_query_norm": 44.94385681152344,
      "action_relation_loss": 0.012949439976364375,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.656688493919551,
      "flow_matching_loss": 0.0693357266485691,
      "grad_norm": 0.22712579369544983,
      "h_stock_valid_norm": 99.81591720581055,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.945634599128139e-05,
      "loss": 0.0823,
      "positive_teacher_pairs": 26966.8,
      "query_h_norm_ratio": 0.45027120113372804,
      "relation_loss": 0.012949439976364375,
      "selected_pairs": 26966.8,
      "step": 2840,
      "student_distance_mean": 33.87794227600098,
      "teacher_distance_mean": 0.8638333439826965,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.012949439976364375,
      "weighted_relation_loss": 0.012949439976364375
    },
    {
      "action_query_batch_std": 0.5099302768707276,
      "action_query_norm": 44.930524826049805,
      "action_relation_loss": 0.013648154214024544,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.666043030869972,
      "flow_matching_loss": 0.07158740237355232,
      "grad_norm": 0.15885786712169647,
      "h_stock_valid_norm": 99.71252670288087,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.944821039454402e-05,
      "loss": 0.0852,
      "positive_teacher_pairs": 26851.4,
      "query_h_norm_ratio": 0.450604048371315,
      "relation_loss": 0.013648154214024544,
      "selected_pairs": 26851.4,
      "step": 2850,
      "student_distance_mean": 32.91957778930664,
      "teacher_distance_mean": 0.8455385982990264,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.013648154214024544,
      "weighted_relation_loss": 0.013648154214024544
    },
    {
      "action_query_batch_std": 0.5145128846168519,
      "action_query_norm": 44.930687713623044,
      "action_relation_loss": 0.01348797669634223,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.6753975678203927,
      "flow_matching_loss": 0.06930367462337017,
      "grad_norm": 0.23114646971225739,
      "h_stock_valid_norm": 99.13667297363281,
      "invalid_tail_samples": 25.7,
      "learning_rate": 9.944001471356835e-05,
      "loss": 0.0828,
      "positive_teacher_pairs": 26412.8,
      "query_h_norm_ratio": 0.45323866605758667,
      "relation_loss": 0.01348797669634223,
      "selected_pairs": 26412.8,
      "step": 2860,
      "student_distance_mean": 33.185952758789064,
      "teacher_distance_mean": 0.8566486120224,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.01348797669634223,
      "weighted_relation_loss": 0.01348797669634223
    },
    {
      "action_query_batch_std": 0.5286009967327118,
      "action_query_norm": 44.9388542175293,
      "action_relation_loss": 0.013291440811008215,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.684752104770814,
      "flow_matching_loss": 0.06904960013926029,
      "grad_norm": 0.22274984419345856,
      "h_stock_valid_norm": 97.9406524658203,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.94317589583129e-05,
      "loss": 0.0823,
      "positive_teacher_pairs": 26855.5,
      "query_h_norm_ratio": 0.4588433146476746,
      "relation_loss": 0.013291440811008215,
      "selected_pairs": 26855.5,
      "step": 2870,
      "student_distance_mean": 34.264854431152344,
      "teacher_distance_mean": 0.8563012778759003,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.013291440811008215,
      "weighted_relation_loss": 0.013291440811008215
    },
    {
      "action_query_batch_std": 0.5094676733016967,
      "action_query_norm": 44.928175354003905,
      "action_relation_loss": 0.01418433990329504,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.694106641721235,
      "flow_matching_loss": 0.07039403691887855,
      "grad_norm": 0.20475280284881592,
      "h_stock_valid_norm": 97.50317153930663,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.942344313880922e-05,
      "loss": 0.0846,
      "positive_teacher_pairs": 26990.4,
      "query_h_norm_ratio": 0.4607916116714478,
      "relation_loss": 0.01418433990329504,
      "selected_pairs": 26990.4,
      "step": 2880,
      "student_distance_mean": 33.03733673095703,
      "teacher_distance_mean": 0.8522884905338287,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.01418433990329504,
      "weighted_relation_loss": 0.01418433990329504
    },
    {
      "action_query_batch_std": 0.4961988300085068,
      "action_query_norm": 44.919598770141604,
      "action_relation_loss": 0.014434367138892411,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.703461178671656,
      "flow_matching_loss": 0.07241085097193718,
      "grad_norm": 0.25807663798332214,
      "h_stock_valid_norm": 97.81616516113282,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.941506726516179e-05,
      "loss": 0.0868,
      "positive_teacher_pairs": 26854.9,
      "query_h_norm_ratio": 0.45923411250114443,
      "relation_loss": 0.014434367138892411,
      "selected_pairs": 26854.9,
      "step": 2890,
      "student_distance_mean": 32.061242294311526,
      "teacher_distance_mean": 0.850574254989624,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.014434367138892411,
      "weighted_relation_loss": 0.014434367138892411
    },
    {
      "action_query_batch_std": 0.5161303818225861,
      "action_query_norm": 44.92675437927246,
      "action_relation_loss": 0.013670721929520369,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.7128157156220767,
      "flow_matching_loss": 0.07070008069276809,
      "grad_norm": 0.15011636912822723,
      "h_stock_valid_norm": 99.60641784667969,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.94066313475481e-05,
      "loss": 0.0844,
      "positive_teacher_pairs": 27153.3,
      "query_h_norm_ratio": 0.45105675160884856,
      "relation_loss": 0.013670721929520369,
      "selected_pairs": 27153.3,
      "step": 2900,
      "student_distance_mean": 33.48236904144287,
      "teacher_distance_mean": 0.8560537457466125,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.013670721929520369,
      "weighted_relation_loss": 0.013670721929520369
    },
    {
      "action_query_batch_std": 0.5160129964351654,
      "action_query_norm": 44.9250072479248,
      "action_relation_loss": 0.014553059358149767,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.722170252572498,
      "flow_matching_loss": 0.07355380579829215,
      "grad_norm": 0.16633586585521698,
      "h_stock_valid_norm": 99.56985015869141,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.939813539621857e-05,
      "loss": 0.0881,
      "positive_teacher_pairs": 27087.3,
      "query_h_norm_ratio": 0.4512064278125763,
      "relation_loss": 0.014553059358149767,
      "selected_pairs": 27087.3,
      "step": 2910,
      "student_distance_mean": 33.31915130615234,
      "teacher_distance_mean": 0.8467543125152588,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.014553059358149767,
      "weighted_relation_loss": 0.014553059358149767
    },
    {
      "action_query_batch_std": 0.4984804421663284,
      "action_query_norm": 44.9151912689209,
      "action_relation_loss": 0.014805332385003566,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.7315247895229184,
      "flow_matching_loss": 0.06975805312395096,
      "grad_norm": 0.19565537571907043,
      "h_stock_valid_norm": 98.17511978149415,
      "invalid_tail_samples": 22.1,
      "learning_rate": 9.93895794214966e-05,
      "loss": 0.0846,
      "positive_teacher_pairs": 27243.4,
      "query_h_norm_ratio": 0.45750783383846283,
      "relation_loss": 0.014805332385003566,
      "selected_pairs": 27243.4,
      "step": 2920,
      "student_distance_mean": 32.19730339050293,
      "teacher_distance_mean": 0.8537183940410614,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.014805332385003566,
      "weighted_relation_loss": 0.014805332385003566
    },
    {
      "action_query_batch_std": 0.5086945533752442,
      "action_query_norm": 44.92140274047851,
      "action_relation_loss": 0.014275475684553385,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.7408793264733395,
      "flow_matching_loss": 0.07169291526079177,
      "grad_norm": 0.27798232436180115,
      "h_stock_valid_norm": 98.56640548706055,
      "invalid_tail_samples": 20.9,
      "learning_rate": 9.938096343377852e-05,
      "loss": 0.086,
      "positive_teacher_pairs": 27533.1,
      "query_h_norm_ratio": 0.4557621628046036,
      "relation_loss": 0.014275475684553385,
      "selected_pairs": 27533.1,
      "step": 2930,
      "student_distance_mean": 32.95906143188476,
      "teacher_distance_mean": 0.856350588798523,
      "valid_action_samples": 235.1,
      "weighted_auxiliary_loss": 0.014275475684553385,
      "weighted_relation_loss": 0.014275475684553385
    },
    {
      "action_query_batch_std": 0.5256537795066833,
      "action_query_norm": 44.93158493041992,
      "action_relation_loss": 0.013905771914869547,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.7502338634237606,
      "flow_matching_loss": 0.06950115785002708,
      "grad_norm": 0.2046710103750229,
      "h_stock_valid_norm": 99.36817016601563,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.937228744353353e-05,
      "loss": 0.0834,
      "positive_teacher_pairs": 27085.0,
      "query_h_norm_ratio": 0.45218091607093813,
      "relation_loss": 0.013905771914869547,
      "selected_pairs": 27085.0,
      "step": 2940,
      "student_distance_mean": 34.0982120513916,
      "teacher_distance_mean": 0.8527253091335296,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.013905771914869547,
      "weighted_relation_loss": 0.013905771914869547
    },
    {
      "action_query_batch_std": 0.512484323978424,
      "action_query_norm": 44.92419624328613,
      "action_relation_loss": 0.014281293749809265,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.7595884003741817,
      "flow_matching_loss": 0.07000356242060661,
      "grad_norm": 0.2219133973121643,
      "h_stock_valid_norm": 100.03080368041992,
      "invalid_tail_samples": 24.4,
      "learning_rate": 9.936355146130379e-05,
      "loss": 0.0843,
      "positive_teacher_pairs": 26713.4,
      "query_h_norm_ratio": 0.4491082191467285,
      "relation_loss": 0.014281293749809265,
      "selected_pairs": 26713.4,
      "step": 2950,
      "student_distance_mean": 33.2643985748291,
      "teacher_distance_mean": 0.8461726844310761,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.014281293749809265,
      "weighted_relation_loss": 0.014281293749809265
    },
    {
      "action_query_batch_std": 0.517788952589035,
      "action_query_norm": 44.92147521972656,
      "action_relation_loss": 0.013509240839630366,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.7689429373246024,
      "flow_matching_loss": 0.07178704664111138,
      "grad_norm": 0.19297493994235992,
      "h_stock_valid_norm": 99.43506164550782,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.935475549770435e-05,
      "loss": 0.0853,
      "positive_teacher_pairs": 26845.7,
      "query_h_norm_ratio": 0.4517723351716995,
      "relation_loss": 0.013509240839630366,
      "selected_pairs": 26845.7,
      "step": 2960,
      "student_distance_mean": 33.430220413208005,
      "teacher_distance_mean": 0.8578133165836335,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.013509240839630366,
      "weighted_relation_loss": 0.013509240839630366
    },
    {
      "action_query_batch_std": 0.506830695271492,
      "action_query_norm": 44.91680908203125,
      "action_relation_loss": 0.014055781438946724,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.7782974742750235,
      "flow_matching_loss": 0.07060576677322387,
      "grad_norm": 0.25860080122947693,
      "h_stock_valid_norm": 99.94717788696289,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.934589956342315e-05,
      "loss": 0.0847,
      "positive_teacher_pairs": 26941.4,
      "query_h_norm_ratio": 0.44941616654396055,
      "relation_loss": 0.014055781438946724,
      "selected_pairs": 26941.4,
      "step": 2970,
      "student_distance_mean": 32.72253932952881,
      "teacher_distance_mean": 0.8586684823036194,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.014055781438946724,
      "weighted_relation_loss": 0.014055781438946724
    },
    {
      "action_query_batch_std": 0.5157438218593597,
      "action_query_norm": 44.921966552734375,
      "action_relation_loss": 0.01344051556661725,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.787652011225444,
      "flow_matching_loss": 0.06902246363461018,
      "grad_norm": 0.1791498214006424,
      "h_stock_valid_norm": 100.32160339355468,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.933698366922093e-05,
      "loss": 0.0825,
      "positive_teacher_pairs": 26626.7,
      "query_h_norm_ratio": 0.4477863997220993,
      "relation_loss": 0.01344051556661725,
      "selected_pairs": 26626.7,
      "step": 2980,
      "student_distance_mean": 33.21568756103515,
      "teacher_distance_mean": 0.8479609966278077,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.01344051556661725,
      "weighted_relation_loss": 0.01344051556661725
    },
    {
      "action_query_batch_std": 0.5110470741987229,
      "action_query_norm": 44.91651153564453,
      "action_relation_loss": 0.014323856309056282,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.7970065481758652,
      "flow_matching_loss": 0.07208288609981536,
      "grad_norm": 0.20761433243751526,
      "h_stock_valid_norm": 98.36570816040039,
      "invalid_tail_samples": 22.9,
      "learning_rate": 9.93280078259314e-05,
      "loss": 0.0864,
      "positive_teacher_pairs": 27068.2,
      "query_h_norm_ratio": 0.4566322475671768,
      "relation_loss": 0.014323856309056282,
      "selected_pairs": 27068.2,
      "step": 2990,
      "student_distance_mean": 32.88340663909912,
      "teacher_distance_mean": 0.8427201211452484,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.014323856309056282,
      "weighted_relation_loss": 0.014323856309056282
    },
    {
      "action_query_batch_std": 0.5177294492721558,
      "action_query_norm": 44.91712875366211,
      "action_relation_loss": 0.013370869494974614,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.8063610851262863,
      "flow_matching_loss": 0.06911516711115837,
      "grad_norm": 0.19790661334991455,
      "h_stock_valid_norm": 97.77490615844727,
      "invalid_tail_samples": 22.6,
      "learning_rate": 9.931897204446104e-05,
      "loss": 0.0825,
      "positive_teacher_pairs": 27126.1,
      "query_h_norm_ratio": 0.45939541757106783,
      "relation_loss": 0.013370869494974614,
      "selected_pairs": 27126.1,
      "step": 3000,
      "student_distance_mean": 33.404448699951175,
      "teacher_distance_mean": 0.8580991208553315,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.013370869494974614,
      "weighted_relation_loss": 0.013370869494974614
    },
    {
      "action_query_batch_std": 0.50795878469944,
      "action_query_norm": 44.90946731567383,
      "action_relation_loss": 0.012892813142389058,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.8157156220767074,
      "flow_matching_loss": 0.06714792735874653,
      "grad_norm": 0.1957341432571411,
      "h_stock_valid_norm": 97.51736450195312,
      "invalid_tail_samples": 25.0,
      "learning_rate": 9.930987633578915e-05,
      "loss": 0.08,
      "positive_teacher_pairs": 26575.7,
      "query_h_norm_ratio": 0.46053149700164797,
      "relation_loss": 0.012892813142389058,
      "selected_pairs": 26575.7,
      "step": 3010,
      "student_distance_mean": 33.00892562866211,
      "teacher_distance_mean": 0.8693503141403198,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.012892813142389058,
      "weighted_relation_loss": 0.012892813142389058
    },
    {
      "action_query_batch_std": 0.5143459111452102,
      "action_query_norm": 44.9106330871582,
      "action_relation_loss": 0.013620256539434194,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.825070159027128,
      "flow_matching_loss": 0.0672314252704382,
      "grad_norm": 0.1578538715839386,
      "h_stock_valid_norm": 96.78516235351563,
      "invalid_tail_samples": 26.2,
      "learning_rate": 9.93007207109679e-05,
      "loss": 0.0809,
      "positive_teacher_pairs": 26312.2,
      "query_h_norm_ratio": 0.46402761340141296,
      "relation_loss": 0.013620256539434194,
      "selected_pairs": 26312.2,
      "step": 3020,
      "student_distance_mean": 33.40130100250244,
      "teacher_distance_mean": 0.8579997062683106,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.013620256539434194,
      "weighted_relation_loss": 0.013620256539434194
    },
    {
      "action_query_batch_std": 0.5162977755069733,
      "action_query_norm": 44.90593223571777,
      "action_relation_loss": 0.01337090926244855,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.834424695977549,
      "flow_matching_loss": 0.06979236379265785,
      "grad_norm": 0.13878926634788513,
      "h_stock_valid_norm": 96.14203414916992,
      "invalid_tail_samples": 24.9,
      "learning_rate": 9.929150518112224e-05,
      "loss": 0.0832,
      "positive_teacher_pairs": 26597.0,
      "query_h_norm_ratio": 0.4670808851718903,
      "relation_loss": 0.01337090926244855,
      "selected_pairs": 26597.0,
      "step": 3030,
      "student_distance_mean": 33.39684791564942,
      "teacher_distance_mean": 0.8588039755821228,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.01337090926244855,
      "weighted_relation_loss": 0.01337090926244855
    },
    {
      "action_query_batch_std": 0.5115345120429993,
      "action_query_norm": 44.90266761779785,
      "action_relation_loss": 0.013598822150379419,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.84377923292797,
      "flow_matching_loss": 0.06808359399437905,
      "grad_norm": 0.18998251855373383,
      "h_stock_valid_norm": 97.15196533203125,
      "invalid_tail_samples": 21.4,
      "learning_rate": 9.928222975744991e-05,
      "loss": 0.0817,
      "positive_teacher_pairs": 27404.9,
      "query_h_norm_ratio": 0.4621980220079422,
      "relation_loss": 0.013598822150379419,
      "selected_pairs": 27404.9,
      "step": 3040,
      "student_distance_mean": 33.088668251037596,
      "teacher_distance_mean": 0.8489076197147369,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.013598822150379419,
      "weighted_relation_loss": 0.013598822150379419
    },
    {
      "action_query_batch_std": 0.5079443156719208,
      "action_query_norm": 44.900042343139646,
      "action_relation_loss": 0.012871298287063838,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.853133769878391,
      "flow_matching_loss": 0.06798850446939468,
      "grad_norm": 0.19808156788349152,
      "h_stock_valid_norm": 98.9036994934082,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.92728944512214e-05,
      "loss": 0.0809,
      "positive_teacher_pairs": 26829.9,
      "query_h_norm_ratio": 0.4539863109588623,
      "relation_loss": 0.012871298287063838,
      "selected_pairs": 26829.9,
      "step": 3050,
      "student_distance_mean": 32.97987403869629,
      "teacher_distance_mean": 0.8596603393554687,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.012871298287063838,
      "weighted_relation_loss": 0.012871298287063838
    },
    {
      "action_query_batch_std": 0.5173075020313262,
      "action_query_norm": 44.90119171142578,
      "action_relation_loss": 0.013220659270882607,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.862488306828812,
      "flow_matching_loss": 0.07007485032081603,
      "grad_norm": 0.17749615013599396,
      "h_stock_valid_norm": 99.89677047729492,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.926349927378001e-05,
      "loss": 0.0833,
      "positive_teacher_pairs": 26735.0,
      "query_h_norm_ratio": 0.4494808316230774,
      "relation_loss": 0.013220659270882607,
      "selected_pairs": 26735.0,
      "step": 3060,
      "student_distance_mean": 33.44879684448242,
      "teacher_distance_mean": 0.8537364482879639,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.013220659270882607,
      "weighted_relation_loss": 0.013220659270882607
    },
    {
      "action_query_batch_std": 0.5269245088100434,
      "action_query_norm": 44.908600616455075,
      "action_relation_loss": 0.012957001198083162,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.871842843779233,
      "flow_matching_loss": 0.06570721305906772,
      "grad_norm": 0.1682761013507843,
      "h_stock_valid_norm": 100.06877670288085,
      "invalid_tail_samples": 25.3,
      "learning_rate": 9.925404423654174e-05,
      "loss": 0.0787,
      "positive_teacher_pairs": 26500.2,
      "query_h_norm_ratio": 0.4487800657749176,
      "relation_loss": 0.012957001198083162,
      "selected_pairs": 26500.2,
      "step": 3070,
      "student_distance_mean": 33.963941955566405,
      "teacher_distance_mean": 0.8580091118812561,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.012957001198083162,
      "weighted_relation_loss": 0.012957001198083162
    },
    {
      "action_query_batch_std": 0.5158925175666809,
      "action_query_norm": 44.8972900390625,
      "action_relation_loss": 0.012825367972254753,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.881197380729654,
      "flow_matching_loss": 0.06855330243706703,
      "grad_norm": 0.2179158329963684,
      "h_stock_valid_norm": 98.7873435974121,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.924452935099537e-05,
      "loss": 0.0814,
      "positive_teacher_pairs": 26760.8,
      "query_h_norm_ratio": 0.45449085533618927,
      "relation_loss": 0.012825367972254753,
      "selected_pairs": 26760.8,
      "step": 3080,
      "student_distance_mean": 33.37177619934082,
      "teacher_distance_mean": 0.8580680906772613,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.012825367972254753,
      "weighted_relation_loss": 0.012825367972254753
    },
    {
      "action_query_batch_std": 0.509573808312416,
      "action_query_norm": 44.89397926330567,
      "action_relation_loss": 0.013682141713798047,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.890551917680075,
      "flow_matching_loss": 0.06797556281089782,
      "grad_norm": 0.21539902687072754,
      "h_stock_valid_norm": 97.87737121582032,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.92349546287024e-05,
      "loss": 0.0817,
      "positive_teacher_pairs": 26761.6,
      "query_h_norm_ratio": 0.45868050754070283,
      "relation_loss": 0.013682141713798047,
      "selected_pairs": 26761.6,
      "step": 3090,
      "student_distance_mean": 33.04845542907715,
      "teacher_distance_mean": 0.8541542410850524,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.013682141713798047,
      "weighted_relation_loss": 0.013682141713798047
    },
    {
      "action_query_batch_std": 0.5144344508647919,
      "action_query_norm": 44.89557991027832,
      "action_relation_loss": 0.013163060881197453,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.8999064546304956,
      "flow_matching_loss": 0.06492077000439167,
      "grad_norm": 0.16831941902637482,
      "h_stock_valid_norm": 97.84642486572265,
      "invalid_tail_samples": 24.9,
      "learning_rate": 9.9225320081297e-05,
      "loss": 0.0781,
      "positive_teacher_pairs": 26596.9,
      "query_h_norm_ratio": 0.4588382303714752,
      "relation_loss": 0.013163060881197453,
      "selected_pairs": 26596.9,
      "step": 3100,
      "student_distance_mean": 33.26681632995606,
      "teacher_distance_mean": 0.8619672238826752,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.013163060881197453,
      "weighted_relation_loss": 0.013163060881197453
    },
    {
      "action_query_batch_std": 0.52239208817482,
      "action_query_norm": 44.89726638793945,
      "action_relation_loss": 0.013771420530974864,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.9092609915809167,
      "flow_matching_loss": 0.06569888852536679,
      "grad_norm": 0.21236686408519745,
      "h_stock_valid_norm": 97.9341552734375,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.921562572048606e-05,
      "loss": 0.0795,
      "positive_teacher_pairs": 26825.8,
      "query_h_norm_ratio": 0.45844662487506865,
      "relation_loss": 0.013771420530974864,
      "selected_pairs": 26825.8,
      "step": 3110,
      "student_distance_mean": 33.60151824951172,
      "teacher_distance_mean": 0.8547627389431,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.013771420530974864,
      "weighted_relation_loss": 0.013771420530974864
    },
    {
      "action_query_batch_std": 0.5125573098659515,
      "action_query_norm": 44.889220809936525,
      "action_relation_loss": 0.01310277935117483,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.9186155285313378,
      "flow_matching_loss": 0.06718738041818142,
      "grad_norm": 0.2119545340538025,
      "h_stock_valid_norm": 98.8200180053711,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.920587155804913e-05,
      "loss": 0.0803,
      "positive_teacher_pairs": 26923.6,
      "query_h_norm_ratio": 0.45425857305526735,
      "relation_loss": 0.01310277935117483,
      "selected_pairs": 26923.6,
      "step": 3120,
      "student_distance_mean": 33.15993194580078,
      "teacher_distance_mean": 0.8582189977169037,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.01310277935117483,
      "weighted_relation_loss": 0.01310277935117483
    },
    {
      "action_query_batch_std": 0.5024055927991867,
      "action_query_norm": 44.87825393676758,
      "action_relation_loss": 0.013309977110475301,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.927970065481759,
      "flow_matching_loss": 0.06691591925919056,
      "grad_norm": 0.23980475962162018,
      "h_stock_valid_norm": 99.32564315795898,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.919605760583845e-05,
      "loss": 0.0802,
      "positive_teacher_pairs": 27149.6,
      "query_h_norm_ratio": 0.4518322914838791,
      "relation_loss": 0.013309977110475301,
      "selected_pairs": 27149.6,
      "step": 3130,
      "student_distance_mean": 32.46031265258789,
      "teacher_distance_mean": 0.8549654245376587,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.013309977110475301,
      "weighted_relation_loss": 0.013309977110475301
    },
    {
      "action_query_batch_std": 0.5152635872364044,
      "action_query_norm": 44.88677825927734,
      "action_relation_loss": 0.013705204986035825,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.9373246024321795,
      "flow_matching_loss": 0.06474780961871147,
      "grad_norm": 0.1687912791967392,
      "h_stock_valid_norm": 100.48455429077148,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.91861838757789e-05,
      "loss": 0.0785,
      "positive_teacher_pairs": 26799.0,
      "query_h_norm_ratio": 0.446706148982048,
      "relation_loss": 0.013705204986035825,
      "selected_pairs": 26799.0,
      "step": 3140,
      "student_distance_mean": 33.45582618713379,
      "teacher_distance_mean": 0.848838883638382,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.013705204986035825,
      "weighted_relation_loss": 0.013705204986035825
    },
    {
      "action_query_batch_std": 0.5190266788005828,
      "action_query_norm": 44.89033393859863,
      "action_relation_loss": 0.013219289109110832,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.9466791393826006,
      "flow_matching_loss": 0.06657949946820736,
      "grad_norm": 0.24972665309906006,
      "h_stock_valid_norm": 100.3550895690918,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.917625037986798e-05,
      "loss": 0.0798,
      "positive_teacher_pairs": 26776.1,
      "query_h_norm_ratio": 0.44731830060482025,
      "relation_loss": 0.013219289109110832,
      "selected_pairs": 26776.1,
      "step": 3150,
      "student_distance_mean": 33.576969146728516,
      "teacher_distance_mean": 0.8568839013576508,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.013219289109110832,
      "weighted_relation_loss": 0.013219289109110832
    },
    {
      "action_query_batch_std": 0.5112395018339158,
      "action_query_norm": 44.88386344909668,
      "action_relation_loss": 0.013723545242100954,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.9560336763330213,
      "flow_matching_loss": 0.06455855630338192,
      "grad_norm": 0.1803719848394394,
      "h_stock_valid_norm": 98.6931640625,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.916625713017583e-05,
      "loss": 0.0783,
      "positive_teacher_pairs": 26776.4,
      "query_h_norm_ratio": 0.4548049598932266,
      "relation_loss": 0.013723545242100954,
      "selected_pairs": 26776.4,
      "step": 3160,
      "student_distance_mean": 33.118309783935544,
      "teacher_distance_mean": 0.8513672113418579,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.013723545242100954,
      "weighted_relation_loss": 0.013723545242100954
    },
    {
      "action_query_batch_std": 0.5087974399328232,
      "action_query_norm": 44.880228424072264,
      "action_relation_loss": 0.013652540370821952,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.9653882132834424,
      "flow_matching_loss": 0.06926971040666104,
      "grad_norm": 0.19559107720851898,
      "h_stock_valid_norm": 96.66586380004883,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.915620413884519e-05,
      "loss": 0.0829,
      "positive_teacher_pairs": 26784.3,
      "query_h_norm_ratio": 0.4642867803573608,
      "relation_loss": 0.013652540370821952,
      "selected_pairs": 26784.3,
      "step": 3170,
      "student_distance_mean": 32.93012161254883,
      "teacher_distance_mean": 0.8535195410251617,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.013652540370821952,
      "weighted_relation_loss": 0.013652540370821952
    },
    {
      "action_query_batch_std": 0.513643878698349,
      "action_query_norm": 44.88322486877441,
      "action_relation_loss": 0.013539712596684694,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.9747427502338635,
      "flow_matching_loss": 0.0665833257138729,
      "grad_norm": 0.18189433217048645,
      "h_stock_valid_norm": 96.22232055664062,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.914609141809139e-05,
      "loss": 0.0801,
      "positive_teacher_pairs": 26921.9,
      "query_h_norm_ratio": 0.4664590984582901,
      "relation_loss": 0.013539712596684694,
      "selected_pairs": 26921.9,
      "step": 3180,
      "student_distance_mean": 33.29696464538574,
      "teacher_distance_mean": 0.8482001960277558,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.013539712596684694,
      "weighted_relation_loss": 0.013539712596684694
    },
    {
      "action_query_batch_std": 0.5233478128910065,
      "action_query_norm": 44.88669204711914,
      "action_relation_loss": 0.013786360155791044,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.9840972871842846,
      "flow_matching_loss": 0.06978115923702717,
      "grad_norm": 0.17468620836734772,
      "h_stock_valid_norm": 97.84409713745117,
      "invalid_tail_samples": 22.1,
      "learning_rate": 9.913591898020235e-05,
      "loss": 0.0836,
      "positive_teacher_pairs": 27247.1,
      "query_h_norm_ratio": 0.4587644934654236,
      "relation_loss": 0.013786360155791044,
      "selected_pairs": 27247.1,
      "step": 3190,
      "student_distance_mean": 33.82390213012695,
      "teacher_distance_mean": 0.8595376431941986,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.013786360155791044,
      "weighted_relation_loss": 0.013786360155791044
    },
    {
      "action_query_batch_std": 0.5090958118438721,
      "action_query_norm": 44.87705192565918,
      "action_relation_loss": 0.013746511377394199,
      "auxiliary_loss_weight": 1.0,
      "epoch": 2.9934518241347052,
      "flow_matching_loss": 0.0710079587996006,
      "grad_norm": 0.18402600288391113,
      "h_stock_valid_norm": 98.41669692993165,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.912568683753853e-05,
      "loss": 0.0848,
      "positive_teacher_pairs": 26684.8,
      "query_h_norm_ratio": 0.4559928923845291,
      "relation_loss": 0.013746511377394199,
      "selected_pairs": 26684.8,
      "step": 3200,
      "student_distance_mean": 33.056797409057616,
      "teacher_distance_mean": 0.8425423264503479,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.013746511377394199,
      "weighted_relation_loss": 0.013746511377394199
    },
    {
      "action_query_batch_std": 0.5210449278354645,
      "action_query_norm": 44.882070541381836,
      "action_relation_loss": 0.013581257220357657,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.0028063610851263,
      "flow_matching_loss": 0.06834439225494862,
      "grad_norm": 0.1733049899339676,
      "h_stock_valid_norm": 99.8143798828125,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.911539500253295e-05,
      "loss": 0.0819,
      "positive_teacher_pairs": 23666.0,
      "query_h_norm_ratio": 0.4496948540210724,
      "relation_loss": 0.013581257220357657,
      "selected_pairs": 23666.0,
      "step": 3210,
      "student_distance_mean": 33.83275871276855,
      "teacher_distance_mean": 0.8711952090263366,
      "valid_action_samples": 211.5,
      "weighted_auxiliary_loss": 0.013581257220357657,
      "weighted_relation_loss": 0.013581257220357657
    },
    {
      "action_query_batch_std": 0.5062190115451812,
      "action_query_norm": 44.8684024810791,
      "action_relation_loss": 0.01342840502038598,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.0121608980355474,
      "flow_matching_loss": 0.06781311593949794,
      "grad_norm": 0.1954505890607834,
      "h_stock_valid_norm": 101.45729522705078,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.910504348769118e-05,
      "loss": 0.0812,
      "positive_teacher_pairs": 26881.7,
      "query_h_norm_ratio": 0.4422429442405701,
      "relation_loss": 0.01342840502038598,
      "selected_pairs": 26881.7,
      "step": 3220,
      "student_distance_mean": 32.815216064453125,
      "teacher_distance_mean": 0.8603328227996826,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.01342840502038598,
      "weighted_relation_loss": 0.01342840502038598
    },
    {
      "action_query_batch_std": 0.517514967918396,
      "action_query_norm": 44.87234306335449,
      "action_relation_loss": 0.013306539319455624,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.021515434985968,
      "flow_matching_loss": 0.06656410209834576,
      "grad_norm": 0.17905090749263763,
      "h_stock_valid_norm": 101.00079193115235,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.909463230559127e-05,
      "loss": 0.0799,
      "positive_teacher_pairs": 27205.0,
      "query_h_norm_ratio": 0.4442827582359314,
      "relation_loss": 0.013306539319455624,
      "selected_pairs": 27205.0,
      "step": 3230,
      "student_distance_mean": 33.40293426513672,
      "teacher_distance_mean": 0.852055960893631,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.013306539319455624,
      "weighted_relation_loss": 0.013306539319455624
    },
    {
      "action_query_batch_std": 0.5161337107419968,
      "action_query_norm": 44.8702392578125,
      "action_relation_loss": 0.01329807536676526,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.030869971936389,
      "flow_matching_loss": 0.06763473823666573,
      "grad_norm": 0.2322627454996109,
      "h_stock_valid_norm": 99.70625915527344,
      "invalid_tail_samples": 22.1,
      "learning_rate": 9.908416146888376e-05,
      "loss": 0.0809,
      "positive_teacher_pairs": 27251.6,
      "query_h_norm_ratio": 0.450050950050354,
      "relation_loss": 0.01329807536676526,
      "selected_pairs": 27251.6,
      "step": 3240,
      "student_distance_mean": 33.37382011413574,
      "teacher_distance_mean": 0.8468757092952728,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.01329807536676526,
      "weighted_relation_loss": 0.01329807536676526
    },
    {
      "action_query_batch_std": 0.515308392047882,
      "action_query_norm": 44.867607498168944,
      "action_relation_loss": 0.013853865303099155,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.0402245088868103,
      "flow_matching_loss": 0.06708218418061733,
      "grad_norm": 0.23984794318675995,
      "h_stock_valid_norm": 97.84249496459961,
      "invalid_tail_samples": 26.0,
      "learning_rate": 9.907363099029175e-05,
      "loss": 0.0809,
      "positive_teacher_pairs": 26344.2,
      "query_h_norm_ratio": 0.4585763931274414,
      "relation_loss": 0.013853865303099155,
      "selected_pairs": 26344.2,
      "step": 3250,
      "student_distance_mean": 33.28035755157471,
      "teacher_distance_mean": 0.8584413230419159,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.013853865303099155,
      "weighted_relation_loss": 0.013853865303099155
    },
    {
      "action_query_batch_std": 0.5299156606197357,
      "action_query_norm": 44.877645874023436,
      "action_relation_loss": 0.012419957388192415,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.049579045837231,
      "flow_matching_loss": 0.06591284610331058,
      "grad_norm": 0.16592708230018616,
      "h_stock_valid_norm": 98.26368637084961,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.906304088261073e-05,
      "loss": 0.0783,
      "positive_teacher_pairs": 27233.9,
      "query_h_norm_ratio": 0.456722104549408,
      "relation_loss": 0.012419957388192415,
      "selected_pairs": 27233.9,
      "step": 3260,
      "student_distance_mean": 34.2536205291748,
      "teacher_distance_mean": 0.8638660430908203,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.012419957388192415,
      "weighted_relation_loss": 0.012419957388192415
    },
    {
      "action_query_batch_std": 0.5173866808414459,
      "action_query_norm": 44.868014907836915,
      "action_relation_loss": 0.013137732073664666,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.058933582787652,
      "flow_matching_loss": 0.06786725893616677,
      "grad_norm": 0.15208810567855835,
      "h_stock_valid_norm": 98.70737915039062,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.905239115870872e-05,
      "loss": 0.081,
      "positive_teacher_pairs": 26854.1,
      "query_h_norm_ratio": 0.4545659124851227,
      "relation_loss": 0.013137732073664666,
      "selected_pairs": 26854.1,
      "step": 3270,
      "student_distance_mean": 33.396205520629884,
      "teacher_distance_mean": 0.8606993377208709,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.013137732073664666,
      "weighted_relation_loss": 0.013137732073664666
    },
    {
      "action_query_batch_std": 0.5208124160766602,
      "action_query_norm": 44.870314025878905,
      "action_relation_loss": 0.01339796120300889,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.068288119738073,
      "flow_matching_loss": 0.06923269182443618,
      "grad_norm": 0.14290525019168854,
      "h_stock_valid_norm": 98.43113555908204,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.90416818315261e-05,
      "loss": 0.0826,
      "positive_teacher_pairs": 26877.9,
      "query_h_norm_ratio": 0.45585917532444,
      "relation_loss": 0.01339796120300889,
      "selected_pairs": 26877.9,
      "step": 3280,
      "student_distance_mean": 33.67946090698242,
      "teacher_distance_mean": 0.8533940613269806,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.01339796120300889,
      "weighted_relation_loss": 0.01339796120300889
    },
    {
      "action_query_batch_std": 0.5181260824203491,
      "action_query_norm": 44.866865158081055,
      "action_relation_loss": 0.012625083420425653,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.077642656688494,
      "flow_matching_loss": 0.0659210193902254,
      "grad_norm": 0.17549562454223633,
      "h_stock_valid_norm": 98.67531814575196,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.903091291407573e-05,
      "loss": 0.0785,
      "positive_teacher_pairs": 27170.2,
      "query_h_norm_ratio": 0.454696848988533,
      "relation_loss": 0.012625083420425653,
      "selected_pairs": 27170.2,
      "step": 3290,
      "student_distance_mean": 33.5160415649414,
      "teacher_distance_mean": 0.8505284786224365,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.012625083420425653,
      "weighted_relation_loss": 0.012625083420425653
    },
    {
      "action_query_batch_std": 0.5076682001352311,
      "action_query_norm": 44.85648155212402,
      "action_relation_loss": 0.012681588903069497,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.086997193638915,
      "flow_matching_loss": 0.06705792620778084,
      "grad_norm": 0.1297617405653,
      "h_stock_valid_norm": 99.31215896606446,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.902008441944286e-05,
      "loss": 0.0797,
      "positive_teacher_pairs": 26972.9,
      "query_h_norm_ratio": 0.4516758233308792,
      "relation_loss": 0.012681588903069497,
      "selected_pairs": 26972.9,
      "step": 3300,
      "student_distance_mean": 33.04341907501221,
      "teacher_distance_mean": 0.8576529145240783,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.012681588903069497,
      "weighted_relation_loss": 0.012681588903069497
    },
    {
      "action_query_batch_std": 0.4956227481365204,
      "action_query_norm": 44.84522171020508,
      "action_relation_loss": 0.012468622252345084,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.096351730589336,
      "flow_matching_loss": 0.06392119005322457,
      "grad_norm": 0.16827915608882904,
      "h_stock_valid_norm": 99.52605285644532,
      "invalid_tail_samples": 26.5,
      "learning_rate": 9.900919636078512e-05,
      "loss": 0.0764,
      "positive_teacher_pairs": 26237.4,
      "query_h_norm_ratio": 0.4505900353193283,
      "relation_loss": 0.012468622252345084,
      "selected_pairs": 26237.4,
      "step": 3310,
      "student_distance_mean": 32.254266548156735,
      "teacher_distance_mean": 0.8495611548423767,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.012468622252345084,
      "weighted_relation_loss": 0.012468622252345084
    },
    {
      "action_query_batch_std": 0.5099608123302459,
      "action_query_norm": 44.852356338500975,
      "action_relation_loss": 0.013596051838248968,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.1057062675397566,
      "flow_matching_loss": 0.06599277667701245,
      "grad_norm": 0.12546513974666595,
      "h_stock_valid_norm": 99.90172576904297,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.899824875133255e-05,
      "loss": 0.0796,
      "positive_teacher_pairs": 26895.5,
      "query_h_norm_ratio": 0.44896776378154757,
      "relation_loss": 0.013596051838248968,
      "selected_pairs": 26895.5,
      "step": 3320,
      "student_distance_mean": 32.94673461914063,
      "teacher_distance_mean": 0.859937185049057,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.013596051838248968,
      "weighted_relation_loss": 0.013596051838248968
    },
    {
      "action_query_batch_std": 0.5061692297458649,
      "action_query_norm": 44.85001335144043,
      "action_relation_loss": 0.012491501867771149,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.1150608044901777,
      "flow_matching_loss": 0.06565787978470325,
      "grad_norm": 0.13509105145931244,
      "h_stock_valid_norm": 98.77723083496093,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.898724160438749e-05,
      "loss": 0.0781,
      "positive_teacher_pairs": 26834.6,
      "query_h_norm_ratio": 0.45406619012355803,
      "relation_loss": 0.012491501867771149,
      "selected_pairs": 26834.6,
      "step": 3330,
      "student_distance_mean": 32.792525100708005,
      "teacher_distance_mean": 0.855983418226242,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.012491501867771149,
      "weighted_relation_loss": 0.012491501867771149
    },
    {
      "action_query_batch_std": 0.4896200120449066,
      "action_query_norm": 44.83704452514648,
      "action_relation_loss": 0.012868446204811335,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.124415341440599,
      "flow_matching_loss": 0.06671848632395268,
      "grad_norm": 0.21789494156837463,
      "h_stock_valid_norm": 97.26913375854492,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.89761749333247e-05,
      "loss": 0.0796,
      "positive_teacher_pairs": 27175.2,
      "query_h_norm_ratio": 0.4609676420688629,
      "relation_loss": 0.012868446204811335,
      "selected_pairs": 27175.2,
      "step": 3340,
      "student_distance_mean": 31.827914810180665,
      "teacher_distance_mean": 0.8491860032081604,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.012868446204811335,
      "weighted_relation_loss": 0.012868446204811335
    },
    {
      "action_query_batch_std": 0.49566497206687926,
      "action_query_norm": 44.83794898986817,
      "action_relation_loss": 0.012338275648653507,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.1337698783910195,
      "flow_matching_loss": 0.06815672293305397,
      "grad_norm": 0.17399738729000092,
      "h_stock_valid_norm": 97.63404159545898,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.896504875159122e-05,
      "loss": 0.0805,
      "positive_teacher_pairs": 27089.8,
      "query_h_norm_ratio": 0.45924916565418245,
      "relation_loss": 0.012338275648653507,
      "selected_pairs": 27089.8,
      "step": 3350,
      "student_distance_mean": 32.1144681930542,
      "teacher_distance_mean": 0.84996777176857,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.012338275648653507,
      "weighted_relation_loss": 0.012338275648653507
    },
    {
      "action_query_batch_std": 0.506155526638031,
      "action_query_norm": 44.844655227661136,
      "action_relation_loss": 0.012507654540240765,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.1431244153414406,
      "flow_matching_loss": 0.06956948861479759,
      "grad_norm": 0.1594768464565277,
      "h_stock_valid_norm": 98.64067459106445,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.89538630727064e-05,
      "loss": 0.0821,
      "positive_teacher_pairs": 27230.4,
      "query_h_norm_ratio": 0.45463171899318694,
      "relation_loss": 0.012507654540240765,
      "selected_pairs": 27230.4,
      "step": 3360,
      "student_distance_mean": 32.7463623046875,
      "teacher_distance_mean": 0.8571313560009003,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.012507654540240765,
      "weighted_relation_loss": 0.012507654540240765
    },
    {
      "action_query_batch_std": 0.5054249137639999,
      "action_query_norm": 44.84410400390625,
      "action_relation_loss": 0.01276294095441699,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.1524789522918617,
      "flow_matching_loss": 0.06771559603512287,
      "grad_norm": 0.19414597749710083,
      "h_stock_valid_norm": 99.47249832153321,
      "invalid_tail_samples": 26.3,
      "learning_rate": 9.894261791026189e-05,
      "loss": 0.0805,
      "positive_teacher_pairs": 26273.9,
      "query_h_norm_ratio": 0.45082347095012665,
      "relation_loss": 0.01276294095441699,
      "selected_pairs": 26273.9,
      "step": 3370,
      "student_distance_mean": 32.66902179718018,
      "teacher_distance_mean": 0.8493212580680847,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.01276294095441699,
      "weighted_relation_loss": 0.01276294095441699
    },
    {
      "action_query_batch_std": 0.5065421044826508,
      "action_query_norm": 44.84460029602051,
      "action_relation_loss": 0.012160942424088717,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.1618334892422824,
      "flow_matching_loss": 0.06603635922074318,
      "grad_norm": 0.20807935297489166,
      "h_stock_valid_norm": 99.33045120239258,
      "invalid_tail_samples": 21.5,
      "learning_rate": 9.893131327792165e-05,
      "loss": 0.0782,
      "positive_teacher_pairs": 27385.8,
      "query_h_norm_ratio": 0.45147387981414794,
      "relation_loss": 0.012160942424088717,
      "selected_pairs": 27385.8,
      "step": 3380,
      "student_distance_mean": 32.96933174133301,
      "teacher_distance_mean": 0.8537519752979279,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.012160942424088717,
      "weighted_relation_loss": 0.012160942424088717
    },
    {
      "action_query_batch_std": 0.5026831597089767,
      "action_query_norm": 44.842838287353516,
      "action_relation_loss": 0.013157035503536463,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.1711880261927035,
      "flow_matching_loss": 0.06629006788134575,
      "grad_norm": 0.2342747002840042,
      "h_stock_valid_norm": 98.9029426574707,
      "invalid_tail_samples": 23.1,
      "learning_rate": 9.891994918942182e-05,
      "loss": 0.0794,
      "positive_teacher_pairs": 27008.0,
      "query_h_norm_ratio": 0.4534059077501297,
      "relation_loss": 0.013157035503536463,
      "selected_pairs": 27008.0,
      "step": 3390,
      "student_distance_mean": 32.665893745422366,
      "teacher_distance_mean": 0.8435348927974701,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.013157035503536463,
      "weighted_relation_loss": 0.013157035503536463
    },
    {
      "action_query_batch_std": 0.5146339535713196,
      "action_query_norm": 44.84969711303711,
      "action_relation_loss": 0.012196247465908527,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.1805425631431246,
      "flow_matching_loss": 0.06836279183626175,
      "grad_norm": 0.16849909722805023,
      "h_stock_valid_norm": 98.4238899230957,
      "invalid_tail_samples": 23.1,
      "learning_rate": 9.890852565857092e-05,
      "loss": 0.0806,
      "positive_teacher_pairs": 27017.3,
      "query_h_norm_ratio": 0.4556866198778152,
      "relation_loss": 0.012196247465908527,
      "selected_pairs": 27017.3,
      "step": 3400,
      "student_distance_mean": 33.26848850250244,
      "teacher_distance_mean": 0.8525424063205719,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.012196247465908527,
      "weighted_relation_loss": 0.012196247465908527
    },
    {
      "action_query_batch_std": 0.5058676481246949,
      "action_query_norm": 44.838704681396486,
      "action_relation_loss": 0.012230671476572752,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.189897100093545,
      "flow_matching_loss": 0.06440422162413598,
      "grad_norm": 0.1606782078742981,
      "h_stock_valid_norm": 98.0839126586914,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.889704269924954e-05,
      "loss": 0.0766,
      "positive_teacher_pairs": 27201.0,
      "query_h_norm_ratio": 0.45716781318187716,
      "relation_loss": 0.012230671476572752,
      "selected_pairs": 27201.0,
      "step": 3410,
      "student_distance_mean": 32.71620388031006,
      "teacher_distance_mean": 0.8470631837844849,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.012230671476572752,
      "weighted_relation_loss": 0.012230671476572752
    },
    {
      "action_query_batch_std": 0.5008967012166977,
      "action_query_norm": 44.83229370117188,
      "action_relation_loss": 0.01258381325751543,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.1992516370439663,
      "flow_matching_loss": 0.06545821838080883,
      "grad_norm": 0.14648647606372833,
      "h_stock_valid_norm": 97.56851730346679,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.888550032541059e-05,
      "loss": 0.078,
      "positive_teacher_pairs": 26783.7,
      "query_h_norm_ratio": 0.4595008432865143,
      "relation_loss": 0.01258381325751543,
      "selected_pairs": 26783.7,
      "step": 3420,
      "student_distance_mean": 32.480329513549805,
      "teacher_distance_mean": 0.8597241044044495,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.01258381325751543,
      "weighted_relation_loss": 0.01258381325751543
    },
    {
      "action_query_batch_std": 0.5117923438549041,
      "action_query_norm": 44.84153022766113,
      "action_relation_loss": 0.013033248577266931,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.2086061739943874,
      "flow_matching_loss": 0.0667386680841446,
      "grad_norm": 0.19509895145893097,
      "h_stock_valid_norm": 97.76771697998046,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.887389855107916e-05,
      "loss": 0.0798,
      "positive_teacher_pairs": 26970.5,
      "query_h_norm_ratio": 0.4586618274450302,
      "relation_loss": 0.013033248577266931,
      "selected_pairs": 26970.5,
      "step": 3430,
      "student_distance_mean": 33.29692459106445,
      "teacher_distance_mean": 0.8554693102836609,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.013033248577266931,
      "weighted_relation_loss": 0.013033248577266931
    },
    {
      "action_query_batch_std": 0.5118651419878006,
      "action_query_norm": 44.8410587310791,
      "action_relation_loss": 0.01238656770437956,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.217960710944808,
      "flow_matching_loss": 0.06632857955992222,
      "grad_norm": 0.16938981413841248,
      "h_stock_valid_norm": 97.6088981628418,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.886223739035248e-05,
      "loss": 0.0787,
      "positive_teacher_pairs": 26622.7,
      "query_h_norm_ratio": 0.459397566318512,
      "relation_loss": 0.01238656770437956,
      "selected_pairs": 26622.7,
      "step": 3440,
      "student_distance_mean": 33.264982032775876,
      "teacher_distance_mean": 0.8550926983356476,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.01238656770437956,
      "weighted_relation_loss": 0.01238656770437956
    },
    {
      "action_query_batch_std": 0.49799594283103943,
      "action_query_norm": 44.829532623291016,
      "action_relation_loss": 0.012745926715433597,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.227315247895229,
      "flow_matching_loss": 0.06590913385152816,
      "grad_norm": 0.1669769138097763,
      "h_stock_valid_norm": 98.5540771484375,
      "invalid_tail_samples": 25.9,
      "learning_rate": 9.885051685739997e-05,
      "loss": 0.0787,
      "positive_teacher_pairs": 26367.4,
      "query_h_norm_ratio": 0.4548888772726059,
      "relation_loss": 0.012745926715433597,
      "selected_pairs": 26367.4,
      "step": 3450,
      "student_distance_mean": 32.28721599578857,
      "teacher_distance_mean": 0.8480401992797851,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.012745926715433597,
      "weighted_relation_loss": 0.012745926715433597
    },
    {
      "action_query_batch_std": 0.5085396766662598,
      "action_query_norm": 44.834913635253905,
      "action_relation_loss": 0.012621988542377949,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.2366697848456503,
      "flow_matching_loss": 0.06345022395253182,
      "grad_norm": 0.16788612306118011,
      "h_stock_valid_norm": 99.87663879394532,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.883873696646316e-05,
      "loss": 0.0761,
      "positive_teacher_pairs": 27223.7,
      "query_h_norm_ratio": 0.4489103138446808,
      "relation_loss": 0.012621988542377949,
      "selected_pairs": 27223.7,
      "step": 3460,
      "student_distance_mean": 32.80252914428711,
      "teacher_distance_mean": 0.847366988658905,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.012621988542377949,
      "weighted_relation_loss": 0.012621988542377949
    },
    {
      "action_query_batch_std": 0.5076210826635361,
      "action_query_norm": 44.829515838623045,
      "action_relation_loss": 0.012090552970767022,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.246024321796071,
      "flow_matching_loss": 0.06481264270842076,
      "grad_norm": 0.20257094502449036,
      "h_stock_valid_norm": 98.6068733215332,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.882689773185575e-05,
      "loss": 0.0769,
      "positive_teacher_pairs": 26931.8,
      "query_h_norm_ratio": 0.45464012026786804,
      "relation_loss": 0.012090552970767022,
      "selected_pairs": 26931.8,
      "step": 3470,
      "student_distance_mean": 32.812690544128415,
      "teacher_distance_mean": 0.8628615856170654,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.012090552970767022,
      "weighted_relation_loss": 0.012090552970767022
    },
    {
      "action_query_batch_std": 0.5019003331661225,
      "action_query_norm": 44.82595329284668,
      "action_relation_loss": 0.01277344934642315,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.255378858746492,
      "flow_matching_loss": 0.0662090502679348,
      "grad_norm": 0.2041889727115631,
      "h_stock_valid_norm": 97.80864181518555,
      "invalid_tail_samples": 23.0,
      "learning_rate": 9.881499916796353e-05,
      "loss": 0.079,
      "positive_teacher_pairs": 27037.1,
      "query_h_norm_ratio": 0.45830526649951936,
      "relation_loss": 0.01277344934642315,
      "selected_pairs": 27037.1,
      "step": 3480,
      "student_distance_mean": 32.456806373596194,
      "teacher_distance_mean": 0.8491572678089142,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.01277344934642315,
      "weighted_relation_loss": 0.01277344934642315
    },
    {
      "action_query_batch_std": 0.5122037142515182,
      "action_query_norm": 44.83481864929199,
      "action_relation_loss": 0.012097228970378637,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.264733395696913,
      "flow_matching_loss": 0.06451640836894512,
      "grad_norm": 0.21785122156143188,
      "h_stock_valid_norm": 98.85410995483399,
      "invalid_tail_samples": 25.0,
      "learning_rate": 9.880304128924434e-05,
      "loss": 0.0766,
      "positive_teacher_pairs": 26569.1,
      "query_h_norm_ratio": 0.45354841351509095,
      "relation_loss": 0.012097228970378637,
      "selected_pairs": 26569.1,
      "step": 3490,
      "student_distance_mean": 33.08220024108887,
      "teacher_distance_mean": 0.8524312853813172,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.012097228970378637,
      "weighted_relation_loss": 0.012097228970378637
    },
    {
      "action_query_batch_std": 0.5035680294036865,
      "action_query_norm": 44.824263381958005,
      "action_relation_loss": 0.012492317240685225,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.2740879326473338,
      "flow_matching_loss": 0.06757698580622673,
      "grad_norm": 0.2237052619457245,
      "h_stock_valid_norm": 98.64139862060547,
      "invalid_tail_samples": 27.1,
      "learning_rate": 9.879102411022817e-05,
      "loss": 0.0801,
      "positive_teacher_pairs": 26096.5,
      "query_h_norm_ratio": 0.45441880226135256,
      "relation_loss": 0.012492317240685225,
      "selected_pairs": 26096.5,
      "step": 3500,
      "student_distance_mean": 32.81528282165527,
      "teacher_distance_mean": 0.8566227793693543,
      "valid_action_samples": 228.9,
      "weighted_auxiliary_loss": 0.012492317240685225,
      "weighted_relation_loss": 0.012492317240685225
    },
    {
      "action_query_batch_std": 0.4973651707172394,
      "action_query_norm": 44.821532821655275,
      "action_relation_loss": 0.01212446829304099,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.283442469597755,
      "flow_matching_loss": 0.06367933303117752,
      "grad_norm": 0.17837722599506378,
      "h_stock_valid_norm": 98.61347351074218,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.877894764551703e-05,
      "loss": 0.0758,
      "positive_teacher_pairs": 27153.3,
      "query_h_norm_ratio": 0.45451884865760805,
      "relation_loss": 0.01212446829304099,
      "selected_pairs": 27153.3,
      "step": 3510,
      "student_distance_mean": 32.18064651489258,
      "teacher_distance_mean": 0.850921905040741,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.01212446829304099,
      "weighted_relation_loss": 0.01212446829304099
    },
    {
      "action_query_batch_std": 0.5048966318368912,
      "action_query_norm": 44.82093734741211,
      "action_relation_loss": 0.01253509558737278,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.292797006548176,
      "flow_matching_loss": 0.0649195708334446,
      "grad_norm": 0.22508656978607178,
      "h_stock_valid_norm": 98.3462028503418,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.876681190978494e-05,
      "loss": 0.0775,
      "positive_teacher_pairs": 26804.7,
      "query_h_norm_ratio": 0.4557522773742676,
      "relation_loss": 0.01253509558737278,
      "selected_pairs": 26804.7,
      "step": 3520,
      "student_distance_mean": 33.001441383361815,
      "teacher_distance_mean": 0.8571419894695282,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.01253509558737278,
      "weighted_relation_loss": 0.01253509558737278
    },
    {
      "action_query_batch_std": 0.496637237071991,
      "action_query_norm": 44.814633178710935,
      "action_relation_loss": 0.012830225564539433,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.3021515434985966,
      "flow_matching_loss": 0.06611145846545696,
      "grad_norm": 0.15277725458145142,
      "h_stock_valid_norm": 97.72845687866212,
      "invalid_tail_samples": 27.0,
      "learning_rate": 9.875461691777797e-05,
      "loss": 0.0789,
      "positive_teacher_pairs": 26123.5,
      "query_h_norm_ratio": 0.45856676399707796,
      "relation_loss": 0.012830225564539433,
      "selected_pairs": 26123.5,
      "step": 3530,
      "student_distance_mean": 32.365123176574706,
      "teacher_distance_mean": 0.8495124757289887,
      "valid_action_samples": 229.0,
      "weighted_auxiliary_loss": 0.012830225564539433,
      "weighted_relation_loss": 0.012830225564539433
    },
    {
      "action_query_batch_std": 0.4997890293598175,
      "action_query_norm": 44.81747741699219,
      "action_relation_loss": 0.012643952388316393,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.3115060804490177,
      "flow_matching_loss": 0.06544643938541413,
      "grad_norm": 0.1939007043838501,
      "h_stock_valid_norm": 97.55257644653321,
      "invalid_tail_samples": 25.0,
      "learning_rate": 9.874236268431417e-05,
      "loss": 0.0781,
      "positive_teacher_pairs": 26579.6,
      "query_h_norm_ratio": 0.45942570865154264,
      "relation_loss": 0.012643952388316393,
      "selected_pairs": 26579.6,
      "step": 3540,
      "student_distance_mean": 32.50272064208984,
      "teacher_distance_mean": 0.8535836279392243,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.012643952388316393,
      "weighted_relation_loss": 0.012643952388316393
    },
    {
      "action_query_batch_std": 0.49796323776245116,
      "action_query_norm": 44.81388053894043,
      "action_relation_loss": 0.012522973027080298,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.320860617399439,
      "flow_matching_loss": 0.06336755491793156,
      "grad_norm": 0.170602485537529,
      "h_stock_valid_norm": 98.334033203125,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.873004922428361e-05,
      "loss": 0.0759,
      "positive_teacher_pairs": 26803.9,
      "query_h_norm_ratio": 0.45573248863220217,
      "relation_loss": 0.012522973027080298,
      "selected_pairs": 26803.9,
      "step": 3550,
      "student_distance_mean": 32.29709262847901,
      "teacher_distance_mean": 0.8448196709156036,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.012522973027080298,
      "weighted_relation_loss": 0.012522973027080298
    },
    {
      "action_query_batch_std": 0.5009707391262055,
      "action_query_norm": 44.81199493408203,
      "action_relation_loss": 0.01186894355341792,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.33021515434986,
      "flow_matching_loss": 0.06356335617601871,
      "grad_norm": 0.1607057750225067,
      "h_stock_valid_norm": 98.1215690612793,
      "invalid_tail_samples": 25.3,
      "learning_rate": 9.871767655264829e-05,
      "loss": 0.0754,
      "positive_teacher_pairs": 26502.2,
      "query_h_norm_ratio": 0.45669967532157896,
      "relation_loss": 0.01186894355341792,
      "selected_pairs": 26502.2,
      "step": 3560,
      "student_distance_mean": 32.53197975158692,
      "teacher_distance_mean": 0.8650893807411194,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.01186894355341792,
      "weighted_relation_loss": 0.01186894355341792
    },
    {
      "action_query_batch_std": 0.49410092532634736,
      "action_query_norm": 44.80797653198242,
      "action_relation_loss": 0.012094830255955457,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.3395696913002806,
      "flow_matching_loss": 0.06482175402343274,
      "grad_norm": 0.12321729958057404,
      "h_stock_valid_norm": 98.13626480102539,
      "invalid_tail_samples": 20.3,
      "learning_rate": 9.87052446844422e-05,
      "loss": 0.0769,
      "positive_teacher_pairs": 27664.0,
      "query_h_norm_ratio": 0.45659506916999815,
      "relation_loss": 0.012094830255955457,
      "selected_pairs": 27664.0,
      "step": 3570,
      "student_distance_mean": 32.08930263519287,
      "teacher_distance_mean": 0.8562560498714447,
      "valid_action_samples": 235.7,
      "weighted_auxiliary_loss": 0.012094830255955457,
      "weighted_relation_loss": 0.012094830255955457
    },
    {
      "action_query_batch_std": 0.5044446110725402,
      "action_query_norm": 44.81128158569336,
      "action_relation_loss": 0.012194901332259178,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.3489242282507017,
      "flow_matching_loss": 0.06849560588598251,
      "grad_norm": 0.1521085500717163,
      "h_stock_valid_norm": 97.87093505859374,
      "invalid_tail_samples": 20.6,
      "learning_rate": 9.869275363477122e-05,
      "loss": 0.0807,
      "positive_teacher_pairs": 27594.8,
      "query_h_norm_ratio": 0.4578662544488907,
      "relation_loss": 0.012194901332259178,
      "selected_pairs": 27594.8,
      "step": 3580,
      "student_distance_mean": 32.607235717773435,
      "teacher_distance_mean": 0.8611296951770783,
      "valid_action_samples": 235.4,
      "weighted_auxiliary_loss": 0.012194901332259178,
      "weighted_relation_loss": 0.012194901332259178
    },
    {
      "action_query_batch_std": 0.5129182159900665,
      "action_query_norm": 44.81862373352051,
      "action_relation_loss": 0.012497267685830593,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.3582787652011223,
      "flow_matching_loss": 0.06394922249019146,
      "grad_norm": 0.13241393864154816,
      "h_stock_valid_norm": 97.84585494995117,
      "invalid_tail_samples": 20.9,
      "learning_rate": 9.868020341881312e-05,
      "loss": 0.0764,
      "positive_teacher_pairs": 27530.4,
      "query_h_norm_ratio": 0.4580544322729111,
      "relation_loss": 0.012497267685830593,
      "selected_pairs": 27530.4,
      "step": 3590,
      "student_distance_mean": 33.1215482711792,
      "teacher_distance_mean": 0.8567508161067963,
      "valid_action_samples": 235.1,
      "weighted_auxiliary_loss": 0.012497267685830593,
      "weighted_relation_loss": 0.012497267685830593
    },
    {
      "action_query_batch_std": 0.5006458938121796,
      "action_query_norm": 44.807289123535156,
      "action_relation_loss": 0.012370216194540262,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.3676333021515434,
      "flow_matching_loss": 0.0669910717755556,
      "grad_norm": 0.225408136844635,
      "h_stock_valid_norm": 97.88192443847656,
      "invalid_tail_samples": 21.8,
      "learning_rate": 9.866759405181765e-05,
      "loss": 0.0794,
      "positive_teacher_pairs": 27323.8,
      "query_h_norm_ratio": 0.4577755868434906,
      "relation_loss": 0.012370216194540262,
      "selected_pairs": 27323.8,
      "step": 3600,
      "student_distance_mean": 32.41963920593262,
      "teacher_distance_mean": 0.8557819366455078,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.012370216194540262,
      "weighted_relation_loss": 0.012370216194540262
    },
    {
      "action_query_batch_std": 0.4980050832033157,
      "action_query_norm": 44.806932830810545,
      "action_relation_loss": 0.012668316345661878,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.3769878391019645,
      "flow_matching_loss": 0.0649641688913107,
      "grad_norm": 0.2771157920360565,
      "h_stock_valid_norm": 98.30689163208008,
      "invalid_tail_samples": 25.9,
      "learning_rate": 9.865492554910633e-05,
      "loss": 0.0776,
      "positive_teacher_pairs": 26366.8,
      "query_h_norm_ratio": 0.4557895451784134,
      "relation_loss": 0.012668316345661878,
      "selected_pairs": 26366.8,
      "step": 3610,
      "student_distance_mean": 32.458466148376466,
      "teacher_distance_mean": 0.8546540439128876,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.012668316345661878,
      "weighted_relation_loss": 0.012668316345661878
    },
    {
      "action_query_batch_std": 0.5041222989559173,
      "action_query_norm": 44.80606536865234,
      "action_relation_loss": 0.013074235431849957,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.3863423760523856,
      "flow_matching_loss": 0.06493264734745026,
      "grad_norm": 0.2393505722284317,
      "h_stock_valid_norm": 98.04418640136718,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.864219792607262e-05,
      "loss": 0.078,
      "positive_teacher_pairs": 26994.0,
      "query_h_norm_ratio": 0.4570008635520935,
      "relation_loss": 0.013074235431849957,
      "selected_pairs": 26994.0,
      "step": 3620,
      "student_distance_mean": 32.73797740936279,
      "teacher_distance_mean": 0.861390370130539,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.013074235431849957,
      "weighted_relation_loss": 0.013074235431849957
    },
    {
      "action_query_batch_std": 0.4904329627752304,
      "action_query_norm": 44.795284271240234,
      "action_relation_loss": 0.012436020746827125,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.3956969130028063,
      "flow_matching_loss": 0.06439851149916649,
      "grad_norm": 0.19064001739025116,
      "h_stock_valid_norm": 97.27951889038086,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.862941119818177e-05,
      "loss": 0.0768,
      "positive_teacher_pairs": 27164.5,
      "query_h_norm_ratio": 0.46048844456672666,
      "relation_loss": 0.012436020746827125,
      "selected_pairs": 27164.5,
      "step": 3630,
      "student_distance_mean": 31.72595958709717,
      "teacher_distance_mean": 0.8485789775848389,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.012436020746827125,
      "weighted_relation_loss": 0.012436020746827125
    },
    {
      "action_query_batch_std": 0.4934786707162857,
      "action_query_norm": 44.79739875793457,
      "action_relation_loss": 0.012703343015164136,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.4050514499532274,
      "flow_matching_loss": 0.06638289801776409,
      "grad_norm": 0.1464066058397293,
      "h_stock_valid_norm": 96.86415786743164,
      "invalid_tail_samples": 21.6,
      "learning_rate": 9.861656538097086e-05,
      "loss": 0.0791,
      "positive_teacher_pairs": 27365.8,
      "query_h_norm_ratio": 0.46247977018356323,
      "relation_loss": 0.012703343015164136,
      "selected_pairs": 27365.8,
      "step": 3640,
      "student_distance_mean": 32.00673637390137,
      "teacher_distance_mean": 0.8466773867607117,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.012703343015164136,
      "weighted_relation_loss": 0.012703343015164136
    },
    {
      "action_query_batch_std": 0.503039824962616,
      "action_query_norm": 44.79855499267578,
      "action_relation_loss": 0.012943998537957668,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.414405986903648,
      "flow_matching_loss": 0.06473740451037883,
      "grad_norm": 0.16475477814674377,
      "h_stock_valid_norm": 97.36906280517579,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.860366049004877e-05,
      "loss": 0.0777,
      "positive_teacher_pairs": 26963.9,
      "query_h_norm_ratio": 0.4600940614938736,
      "relation_loss": 0.012943998537957668,
      "selected_pairs": 26963.9,
      "step": 3650,
      "student_distance_mean": 32.87341461181641,
      "teacher_distance_mean": 0.8561480700969696,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.012943998537957668,
      "weighted_relation_loss": 0.012943998537957668
    },
    {
      "action_query_batch_std": 0.5030219852924347,
      "action_query_norm": 44.79931716918945,
      "action_relation_loss": 0.012355739902704953,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.423760523854069,
      "flow_matching_loss": 0.0652408417314291,
      "grad_norm": 0.19839277863502502,
      "h_stock_valid_norm": 98.05226516723633,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.859069654109615e-05,
      "loss": 0.0776,
      "positive_teacher_pairs": 26942.7,
      "query_h_norm_ratio": 0.4568979412317276,
      "relation_loss": 0.012355739902704953,
      "selected_pairs": 26942.7,
      "step": 3660,
      "student_distance_mean": 32.68153095245361,
      "teacher_distance_mean": 0.8544001579284668,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.012355739902704953,
      "weighted_relation_loss": 0.012355739902704953
    },
    {
      "action_query_batch_std": 0.4998233139514923,
      "action_query_norm": 44.78957061767578,
      "action_relation_loss": 0.01309727570042014,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.4331150608044902,
      "flow_matching_loss": 0.0630221951752901,
      "grad_norm": 0.26230689883232117,
      "h_stock_valid_norm": 97.49541473388672,
      "invalid_tail_samples": 25.1,
      "learning_rate": 9.857767354986545e-05,
      "loss": 0.0761,
      "positive_teacher_pairs": 26559.0,
      "query_h_norm_ratio": 0.45941172540187836,
      "relation_loss": 0.01309727570042014,
      "selected_pairs": 26559.0,
      "step": 3670,
      "student_distance_mean": 32.44059505462646,
      "teacher_distance_mean": 0.861267375946045,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.01309727570042014,
      "weighted_relation_loss": 0.01309727570042014
    },
    {
      "action_query_batch_std": 0.49066882133483886,
      "action_query_norm": 44.77815895080566,
      "action_relation_loss": 0.012805395107716323,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.4424695977549113,
      "flow_matching_loss": 0.06414306424558162,
      "grad_norm": 0.2623308300971985,
      "h_stock_valid_norm": 96.86018905639648,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.856459153218078e-05,
      "loss": 0.0769,
      "positive_teacher_pairs": 26618.7,
      "query_h_norm_ratio": 0.46229877769947053,
      "relation_loss": 0.012805395107716323,
      "selected_pairs": 26618.7,
      "step": 3680,
      "student_distance_mean": 31.964279556274413,
      "teacher_distance_mean": 0.8439852118492126,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.012805395107716323,
      "weighted_relation_loss": 0.012805395107716323
    },
    {
      "action_query_batch_std": 0.49755253195762633,
      "action_query_norm": 44.78201065063477,
      "action_relation_loss": 0.013147445581853389,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.451824134705332,
      "flow_matching_loss": 0.06467083990573883,
      "grad_norm": 0.20372620224952698,
      "h_stock_valid_norm": 97.12646865844727,
      "invalid_tail_samples": 25.5,
      "learning_rate": 9.855145050393808e-05,
      "loss": 0.0778,
      "positive_teacher_pairs": 26457.6,
      "query_h_norm_ratio": 0.46107394993305206,
      "relation_loss": 0.013147445581853389,
      "selected_pairs": 26457.6,
      "step": 3690,
      "student_distance_mean": 32.2555477142334,
      "teacher_distance_mean": 0.8572499692440033,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.013147445581853389,
      "weighted_relation_loss": 0.013147445581853389
    },
    {
      "action_query_batch_std": 0.5057856947183609,
      "action_query_norm": 44.78556938171387,
      "action_relation_loss": 0.012620511651039123,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.461178671655753,
      "flow_matching_loss": 0.06472216583788395,
      "grad_norm": 0.14871491491794586,
      "h_stock_valid_norm": 97.46234588623047,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.85382504811049e-05,
      "loss": 0.0773,
      "positive_teacher_pairs": 26831.6,
      "query_h_norm_ratio": 0.45952342450618744,
      "relation_loss": 0.012620511651039123,
      "selected_pairs": 26831.6,
      "step": 3700,
      "student_distance_mean": 32.96997623443603,
      "teacher_distance_mean": 0.8528892695903778,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.012620511651039123,
      "weighted_relation_loss": 0.012620511651039123
    },
    {
      "action_query_batch_std": 0.5062130361795425,
      "action_query_norm": 44.79176406860351,
      "action_relation_loss": 0.012707559019327163,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.4705332086061738,
      "flow_matching_loss": 0.06566533520817756,
      "grad_norm": 0.15645736455917358,
      "h_stock_valid_norm": 97.50680084228516,
      "invalid_tail_samples": 22.1,
      "learning_rate": 9.852499147972054e-05,
      "loss": 0.0784,
      "positive_teacher_pairs": 27253.7,
      "query_h_norm_ratio": 0.45937432050704957,
      "relation_loss": 0.012707559019327163,
      "selected_pairs": 27253.7,
      "step": 3710,
      "student_distance_mean": 32.847947692871095,
      "teacher_distance_mean": 0.8499199032783509,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.012707559019327163,
      "weighted_relation_loss": 0.012707559019327163
    },
    {
      "action_query_batch_std": 0.5052340000867843,
      "action_query_norm": 44.78567199707031,
      "action_relation_loss": 0.011614924296736717,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.479887745556595,
      "flow_matching_loss": 0.06524085961282253,
      "grad_norm": 0.18559493124485016,
      "h_stock_valid_norm": 97.6681739807129,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.851167351589592e-05,
      "loss": 0.0769,
      "positive_teacher_pairs": 26944.6,
      "query_h_norm_ratio": 0.458550962805748,
      "relation_loss": 0.011614924296736717,
      "selected_pairs": 26944.6,
      "step": 3720,
      "student_distance_mean": 32.6501989364624,
      "teacher_distance_mean": 0.86103395819664,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.011614924296736717,
      "weighted_relation_loss": 0.011614924296736717
    },
    {
      "action_query_batch_std": 0.5147742509841919,
      "action_query_norm": 44.79120292663574,
      "action_relation_loss": 0.011765156500041484,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.489242282507016,
      "flow_matching_loss": 0.06703385673463344,
      "grad_norm": 0.18515408039093018,
      "h_stock_valid_norm": 98.04786071777343,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.849829660581363e-05,
      "loss": 0.0788,
      "positive_teacher_pairs": 26756.8,
      "query_h_norm_ratio": 0.4568368226289749,
      "relation_loss": 0.011765156500041484,
      "selected_pairs": 26756.8,
      "step": 3730,
      "student_distance_mean": 33.27156867980957,
      "teacher_distance_mean": 0.8642394661903381,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.011765156500041484,
      "weighted_relation_loss": 0.011765156500041484
    },
    {
      "action_query_batch_std": 0.5128304600715637,
      "action_query_norm": 44.788096237182614,
      "action_relation_loss": 0.012825441174209118,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.498596819457437,
      "flow_matching_loss": 0.06735937744379043,
      "grad_norm": 0.14494825899600983,
      "h_stock_valid_norm": 96.91286773681641,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.848486076572787e-05,
      "loss": 0.0802,
      "positive_teacher_pairs": 26853.2,
      "query_h_norm_ratio": 0.4621626824140549,
      "relation_loss": 0.012825441174209118,
      "selected_pairs": 26853.2,
      "step": 3740,
      "student_distance_mean": 33.294979858398435,
      "teacher_distance_mean": 0.8681586503982544,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.012825441174209118,
      "weighted_relation_loss": 0.012825441174209118
    },
    {
      "action_query_batch_std": 0.503338173031807,
      "action_query_norm": 44.782990264892575,
      "action_relation_loss": 0.01246678614988923,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.5079513564078577,
      "flow_matching_loss": 0.06419788561761379,
      "grad_norm": 0.17803896963596344,
      "h_stock_valid_norm": 96.77884216308594,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.847136601196446e-05,
      "loss": 0.0767,
      "positive_teacher_pairs": 27221.8,
      "query_h_norm_ratio": 0.4627410411834717,
      "relation_loss": 0.01246678614988923,
      "selected_pairs": 27221.8,
      "step": 3750,
      "student_distance_mean": 32.75820598602295,
      "teacher_distance_mean": 0.8481390655040741,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.01246678614988923,
      "weighted_relation_loss": 0.01246678614988923
    },
    {
      "action_query_batch_std": 0.5000562340021133,
      "action_query_norm": 44.77436294555664,
      "action_relation_loss": 0.012373638711869717,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.517305893358279,
      "flow_matching_loss": 0.06758014224469662,
      "grad_norm": 0.22852753102779388,
      "h_stock_valid_norm": 97.40702514648437,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.845781236092078e-05,
      "loss": 0.08,
      "positive_teacher_pairs": 26802.9,
      "query_h_norm_ratio": 0.4596639066934586,
      "relation_loss": 0.012373638711869717,
      "selected_pairs": 26802.9,
      "step": 3760,
      "student_distance_mean": 32.46127128601074,
      "teacher_distance_mean": 0.8561072766780853,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.012373638711869717,
      "weighted_relation_loss": 0.012373638711869717
    },
    {
      "action_query_batch_std": 0.5006859362125397,
      "action_query_norm": 44.77229766845703,
      "action_relation_loss": 0.012246330082416535,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.5266604303086995,
      "flow_matching_loss": 0.0680889755487442,
      "grad_norm": 0.18698503077030182,
      "h_stock_valid_norm": 97.96782531738282,
      "invalid_tail_samples": 27.1,
      "learning_rate": 9.844419982906583e-05,
      "loss": 0.0803,
      "positive_teacher_pairs": 26096.2,
      "query_h_norm_ratio": 0.45702114701271057,
      "relation_loss": 0.012246330082416535,
      "selected_pairs": 26096.2,
      "step": 3770,
      "student_distance_mean": 32.55522899627685,
      "teacher_distance_mean": 0.857233464717865,
      "valid_action_samples": 228.9,
      "weighted_auxiliary_loss": 0.012246330082416535,
      "weighted_relation_loss": 0.012246330082416535
    },
    {
      "action_query_batch_std": 0.5068958759307861,
      "action_query_norm": 44.77577514648438,
      "action_relation_loss": 0.012069155927747488,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.5360149672591206,
      "flow_matching_loss": 0.06688151136040688,
      "grad_norm": 0.18234391510486603,
      "h_stock_valid_norm": 99.69877319335937,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.843052843294008e-05,
      "loss": 0.079,
      "positive_teacher_pairs": 26648.6,
      "query_h_norm_ratio": 0.44912917613983155,
      "relation_loss": 0.012069155927747488,
      "selected_pairs": 26648.6,
      "step": 3780,
      "student_distance_mean": 32.88885555267334,
      "teacher_distance_mean": 0.8598987340927124,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.012069155927747488,
      "weighted_relation_loss": 0.012069155927747488
    },
    {
      "action_query_batch_std": 0.49789121448993684,
      "action_query_norm": 44.767595291137695,
      "action_relation_loss": 0.01255653090775013,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.5453695042095417,
      "flow_matching_loss": 0.06491978578269482,
      "grad_norm": 0.28539860248565674,
      "h_stock_valid_norm": 100.72485733032227,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.841679818915559e-05,
      "loss": 0.0775,
      "positive_teacher_pairs": 26686.9,
      "query_h_norm_ratio": 0.4444582462310791,
      "relation_loss": 0.01255653090775013,
      "selected_pairs": 26686.9,
      "step": 3790,
      "student_distance_mean": 32.473723602294925,
      "teacher_distance_mean": 0.8510994851589203,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.01255653090775013,
      "weighted_relation_loss": 0.01255653090775013
    },
    {
      "action_query_batch_std": 0.5046996504068375,
      "action_query_norm": 44.765966796875,
      "action_relation_loss": 0.012239484954625368,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.5547240411599628,
      "flow_matching_loss": 0.06492916010320186,
      "grad_norm": 0.22499065101146698,
      "h_stock_valid_norm": 100.77685775756837,
      "invalid_tail_samples": 25.8,
      "learning_rate": 9.840300911439591e-05,
      "loss": 0.0772,
      "positive_teacher_pairs": 26385.9,
      "query_h_norm_ratio": 0.44421366751194,
      "relation_loss": 0.012239484954625368,
      "selected_pairs": 26385.9,
      "step": 3800,
      "student_distance_mean": 32.864422988891604,
      "teacher_distance_mean": 0.851911437511444,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.012239484954625368,
      "weighted_relation_loss": 0.012239484954625368
    },
    {
      "action_query_batch_std": 0.5092077255249023,
      "action_query_norm": 44.772546005249026,
      "action_relation_loss": 0.013146471418440341,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.5640785781103834,
      "flow_matching_loss": 0.068349289894104,
      "grad_norm": 0.18108929693698883,
      "h_stock_valid_norm": 99.16161575317383,
      "invalid_tail_samples": 22.6,
      "learning_rate": 9.838916122541603e-05,
      "loss": 0.0815,
      "positive_teacher_pairs": 27129.3,
      "query_h_norm_ratio": 0.45156060457229613,
      "relation_loss": 0.013146471418440341,
      "selected_pairs": 27129.3,
      "step": 3810,
      "student_distance_mean": 33.107849502563475,
      "teacher_distance_mean": 0.862323659658432,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.013146471418440341,
      "weighted_relation_loss": 0.013146471418440341
    },
    {
      "action_query_batch_std": 0.4951993376016617,
      "action_query_norm": 44.75776596069336,
      "action_relation_loss": 0.012666534818708896,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.5734331150608045,
      "flow_matching_loss": 0.06549479924142361,
      "grad_norm": 0.20154550671577454,
      "h_stock_valid_norm": 96.59283981323242,
      "invalid_tail_samples": 25.0,
      "learning_rate": 9.837525453904246e-05,
      "loss": 0.0782,
      "positive_teacher_pairs": 26584.5,
      "query_h_norm_ratio": 0.4633713126182556,
      "relation_loss": 0.012666534818708896,
      "selected_pairs": 26584.5,
      "step": 3820,
      "student_distance_mean": 32.218800735473636,
      "teacher_distance_mean": 0.8469612061977386,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.012666534818708896,
      "weighted_relation_loss": 0.012666534818708896
    },
    {
      "action_query_batch_std": 0.5073511838912964,
      "action_query_norm": 44.768804931640624,
      "action_relation_loss": 0.012267236690968275,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.582787652011225,
      "flow_matching_loss": 0.06581128165125846,
      "grad_norm": 0.18928715586662292,
      "h_stock_valid_norm": 97.12501678466796,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.836128907217314e-05,
      "loss": 0.0781,
      "positive_teacher_pairs": 26873.1,
      "query_h_norm_ratio": 0.4609450399875641,
      "relation_loss": 0.012267236690968275,
      "selected_pairs": 26873.1,
      "step": 3830,
      "student_distance_mean": 32.98854522705078,
      "teacher_distance_mean": 0.8674036204814911,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.012267236690968275,
      "weighted_relation_loss": 0.012267236690968275
    },
    {
      "action_query_batch_std": 0.4962958574295044,
      "action_query_norm": 44.7613338470459,
      "action_relation_loss": 0.012436643708497286,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.5921421889616463,
      "flow_matching_loss": 0.06527865566313266,
      "grad_norm": 0.14024657011032104,
      "h_stock_valid_norm": 97.80292587280273,
      "invalid_tail_samples": 22.9,
      "learning_rate": 9.834726484177743e-05,
      "loss": 0.0777,
      "positive_teacher_pairs": 27064.6,
      "query_h_norm_ratio": 0.45768057107925414,
      "relation_loss": 0.012436643708497286,
      "selected_pairs": 27064.6,
      "step": 3840,
      "student_distance_mean": 32.3802375793457,
      "teacher_distance_mean": 0.8484031856060028,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.012436643708497286,
      "weighted_relation_loss": 0.012436643708497286
    },
    {
      "action_query_batch_std": 0.5014595210552215,
      "action_query_norm": 44.76242523193359,
      "action_relation_loss": 0.012623950839042664,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.6014967259120674,
      "flow_matching_loss": 0.06906629502773284,
      "grad_norm": 0.19247296452522278,
      "h_stock_valid_norm": 98.16304168701171,
      "invalid_tail_samples": 25.5,
      "learning_rate": 9.833318186489609e-05,
      "loss": 0.0817,
      "positive_teacher_pairs": 26455.8,
      "query_h_norm_ratio": 0.4560044378042221,
      "relation_loss": 0.012623950839042664,
      "selected_pairs": 26455.8,
      "step": 3850,
      "student_distance_mean": 32.687373352050784,
      "teacher_distance_mean": 0.8613184511661529,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.012623950839042664,
      "weighted_relation_loss": 0.012623950839042664
    },
    {
      "action_query_batch_std": 0.5107043236494064,
      "action_query_norm": 44.76896362304687,
      "action_relation_loss": 0.012162028625607491,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.6108512628624885,
      "flow_matching_loss": 0.06441294774413109,
      "grad_norm": 0.14624907076358795,
      "h_stock_valid_norm": 97.53990020751954,
      "invalid_tail_samples": 22.0,
      "learning_rate": 9.831904015864126e-05,
      "loss": 0.0766,
      "positive_teacher_pairs": 27270.3,
      "query_h_norm_ratio": 0.45899082720279694,
      "relation_loss": 0.012162028625607491,
      "selected_pairs": 27270.3,
      "step": 3860,
      "student_distance_mean": 33.232424926757815,
      "teacher_distance_mean": 0.8603314578533172,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.012162028625607491,
      "weighted_relation_loss": 0.012162028625607491
    },
    {
      "action_query_batch_std": 0.5008806645870209,
      "action_query_norm": 44.759091186523435,
      "action_relation_loss": 0.012580766901373863,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.620205799812909,
      "flow_matching_loss": 0.06415498666465283,
      "grad_norm": 0.20273198187351227,
      "h_stock_valid_norm": 98.89377975463867,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.830483974019645e-05,
      "loss": 0.0767,
      "positive_teacher_pairs": 26848.1,
      "query_h_norm_ratio": 0.4526046276092529,
      "relation_loss": 0.012580766901373863,
      "selected_pairs": 26848.1,
      "step": 3870,
      "student_distance_mean": 32.52336826324463,
      "teacher_distance_mean": 0.8484033107757568,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.012580766901373863,
      "weighted_relation_loss": 0.012580766901373863
    },
    {
      "action_query_batch_std": 0.4980372488498688,
      "action_query_norm": 44.75508003234863,
      "action_relation_loss": 0.012367087788879872,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.6295603367633302,
      "flow_matching_loss": 0.06711448356509209,
      "grad_norm": 0.24335655570030212,
      "h_stock_valid_norm": 99.26194458007812,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.82905806268165e-05,
      "loss": 0.0795,
      "positive_teacher_pairs": 27089.5,
      "query_h_norm_ratio": 0.4508819967508316,
      "relation_loss": 0.012367087788879872,
      "selected_pairs": 27089.5,
      "step": 3880,
      "student_distance_mean": 32.39482822418213,
      "teacher_distance_mean": 0.8506798863410949,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.012367087788879872,
      "weighted_relation_loss": 0.012367087788879872
    },
    {
      "action_query_batch_std": 0.4988353341817856,
      "action_query_norm": 44.75344886779785,
      "action_relation_loss": 0.012337959744036198,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.6389148737137513,
      "flow_matching_loss": 0.06695900708436966,
      "grad_norm": 0.20758822560310364,
      "h_stock_valid_norm": 99.628271484375,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.82762628358276e-05,
      "loss": 0.0793,
      "positive_teacher_pairs": 27197.4,
      "query_h_norm_ratio": 0.44920879006385805,
      "relation_loss": 0.012337959744036198,
      "selected_pairs": 27197.4,
      "step": 3890,
      "student_distance_mean": 32.40538005828857,
      "teacher_distance_mean": 0.8552921831607818,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.012337959744036198,
      "weighted_relation_loss": 0.012337959744036198
    },
    {
      "action_query_batch_std": 0.5016614466905593,
      "action_query_norm": 44.7579833984375,
      "action_relation_loss": 0.0121180459856987,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.648269410664172,
      "flow_matching_loss": 0.06738460548222065,
      "grad_norm": 0.1710558384656906,
      "h_stock_valid_norm": 98.98529052734375,
      "invalid_tail_samples": 23.1,
      "learning_rate": 9.826188638462718e-05,
      "loss": 0.0795,
      "positive_teacher_pairs": 27010.7,
      "query_h_norm_ratio": 0.4521710485219955,
      "relation_loss": 0.0121180459856987,
      "selected_pairs": 27010.7,
      "step": 3900,
      "student_distance_mean": 32.53814697265625,
      "teacher_distance_mean": 0.8569932401180267,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0121180459856987,
      "weighted_relation_loss": 0.0121180459856987
    },
    {
      "action_query_batch_std": 0.4973900705575943,
      "action_query_norm": 44.752709197998044,
      "action_relation_loss": 0.012029391340911388,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.657623947614593,
      "flow_matching_loss": 0.06679116785526276,
      "grad_norm": 0.17932623624801636,
      "h_stock_valid_norm": 97.52615051269531,
      "invalid_tail_samples": 21.6,
      "learning_rate": 9.824745129068402e-05,
      "loss": 0.0788,
      "positive_teacher_pairs": 27371.1,
      "query_h_norm_ratio": 0.4588885486125946,
      "relation_loss": 0.012029391340911388,
      "selected_pairs": 27371.1,
      "step": 3910,
      "student_distance_mean": 32.271823692321775,
      "teacher_distance_mean": 0.8559970617294311,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.012029391340911388,
      "weighted_relation_loss": 0.012029391340911388
    },
    {
      "action_query_batch_std": 0.4926137387752533,
      "action_query_norm": 44.749477005004884,
      "action_relation_loss": 0.012384537048637867,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.666978484565014,
      "flow_matching_loss": 0.0633243590593338,
      "grad_norm": 0.16539771854877472,
      "h_stock_valid_norm": 96.49797821044922,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.82329575715381e-05,
      "loss": 0.0757,
      "positive_teacher_pairs": 27181.1,
      "query_h_norm_ratio": 0.4637400388717651,
      "relation_loss": 0.012384537048637867,
      "selected_pairs": 27181.1,
      "step": 3920,
      "student_distance_mean": 32.018812561035155,
      "teacher_distance_mean": 0.852099722623825,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.012384537048637867,
      "weighted_relation_loss": 0.012384537048637867
    },
    {
      "action_query_batch_std": 0.48565883934497833,
      "action_query_norm": 44.73593292236328,
      "action_relation_loss": 0.012727680336683988,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.676333021515435,
      "flow_matching_loss": 0.06589670442044734,
      "grad_norm": 0.17991390824317932,
      "h_stock_valid_norm": 97.89198150634766,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.821840524480066e-05,
      "loss": 0.0786,
      "positive_teacher_pairs": 26801.4,
      "query_h_norm_ratio": 0.4570006012916565,
      "relation_loss": 0.012727680336683988,
      "selected_pairs": 26801.4,
      "step": 3930,
      "student_distance_mean": 31.764649391174316,
      "teacher_distance_mean": 0.8567519426345825,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.012727680336683988,
      "weighted_relation_loss": 0.012727680336683988
    },
    {
      "action_query_batch_std": 0.49551369845867155,
      "action_query_norm": 44.74156188964844,
      "action_relation_loss": 0.012398014403879642,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.685687558465856,
      "flow_matching_loss": 0.06245317943394184,
      "grad_norm": 0.1392868012189865,
      "h_stock_valid_norm": 98.83936004638672,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.820379432815414e-05,
      "loss": 0.0749,
      "positive_teacher_pairs": 26735.2,
      "query_h_norm_ratio": 0.4526721179485321,
      "relation_loss": 0.012398014403879642,
      "selected_pairs": 26735.2,
      "step": 3940,
      "student_distance_mean": 32.08341827392578,
      "teacher_distance_mean": 0.8521925926208496,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.012398014403879642,
      "weighted_relation_loss": 0.012398014403879642
    },
    {
      "action_query_batch_std": 0.5020648598670959,
      "action_query_norm": 44.74746131896973,
      "action_relation_loss": 0.012529667839407921,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.695042095416277,
      "flow_matching_loss": 0.06667602919042111,
      "grad_norm": 0.21207495033740997,
      "h_stock_valid_norm": 99.68838729858399,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.81891248393522e-05,
      "loss": 0.0792,
      "positive_teacher_pairs": 26802.2,
      "query_h_norm_ratio": 0.44887966215610503,
      "relation_loss": 0.012529667839407921,
      "selected_pairs": 26802.2,
      "step": 3950,
      "student_distance_mean": 32.66852626800537,
      "teacher_distance_mean": 0.8720831096172332,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.012529667839407921,
      "weighted_relation_loss": 0.012529667839407921
    },
    {
      "action_query_batch_std": 0.48891113996505736,
      "action_query_norm": 44.73695602416992,
      "action_relation_loss": 0.011883874703198672,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.7043966323666977,
      "flow_matching_loss": 0.06375327967107296,
      "grad_norm": 0.16446268558502197,
      "h_stock_valid_norm": 99.23433303833008,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.817439679621963e-05,
      "loss": 0.0756,
      "positive_teacher_pairs": 27225.0,
      "query_h_norm_ratio": 0.45084753036499026,
      "relation_loss": 0.011883874703198672,
      "selected_pairs": 27225.0,
      "step": 3960,
      "student_distance_mean": 31.75687255859375,
      "teacher_distance_mean": 0.8568024277687073,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.011883874703198672,
      "weighted_relation_loss": 0.011883874703198672
    },
    {
      "action_query_batch_std": 0.4865871131420135,
      "action_query_norm": 44.73028030395508,
      "action_relation_loss": 0.012367331702262164,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.713751169317119,
      "flow_matching_loss": 0.06798743642866611,
      "grad_norm": 0.1744118183851242,
      "h_stock_valid_norm": 97.17819061279297,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.815961021665243e-05,
      "loss": 0.0804,
      "positive_teacher_pairs": 26802.7,
      "query_h_norm_ratio": 0.4602989614009857,
      "relation_loss": 0.012367331702262164,
      "selected_pairs": 26802.7,
      "step": 3970,
      "student_distance_mean": 31.913373756408692,
      "teacher_distance_mean": 0.8558969020843505,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.012367331702262164,
      "weighted_relation_loss": 0.012367331702262164
    },
    {
      "action_query_batch_std": 0.4907973200082779,
      "action_query_norm": 44.73333740234375,
      "action_relation_loss": 0.011684397142380476,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.72310570626754,
      "flow_matching_loss": 0.06307297237217427,
      "grad_norm": 0.210206538438797,
      "h_stock_valid_norm": 96.47980346679688,
      "invalid_tail_samples": 25.8,
      "learning_rate": 9.814476511861763e-05,
      "loss": 0.0748,
      "positive_teacher_pairs": 26387.8,
      "query_h_norm_ratio": 0.46366012990474703,
      "relation_loss": 0.011684397142380476,
      "selected_pairs": 26387.8,
      "step": 3980,
      "student_distance_mean": 32.01730155944824,
      "teacher_distance_mean": 0.8586973011493683,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.011684397142380476,
      "weighted_relation_loss": 0.011684397142380476
    },
    {
      "action_query_batch_std": 0.4925226032733917,
      "action_query_norm": 44.73072395324707,
      "action_relation_loss": 0.011757891532033682,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.7324602432179605,
      "flow_matching_loss": 0.06320951879024506,
      "grad_norm": 0.18127548694610596,
      "h_stock_valid_norm": 96.33652267456054,
      "invalid_tail_samples": 25.3,
      "learning_rate": 9.812986152015348e-05,
      "loss": 0.075,
      "positive_teacher_pairs": 26509.3,
      "query_h_norm_ratio": 0.46432062387466433,
      "relation_loss": 0.011757891532033682,
      "selected_pairs": 26509.3,
      "step": 3990,
      "student_distance_mean": 32.05198497772217,
      "teacher_distance_mean": 0.8543035745620727,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.011757891532033682,
      "weighted_relation_loss": 0.011757891532033682
    },
    {
      "action_query_batch_std": 0.4935669392347336,
      "action_query_norm": 44.73046188354492,
      "action_relation_loss": 0.011845844238996506,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.7418147801683816,
      "flow_matching_loss": 0.06421531774103642,
      "grad_norm": 0.1943231225013733,
      "h_stock_valid_norm": 96.70641784667968,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.811489943936922e-05,
      "loss": 0.0761,
      "positive_teacher_pairs": 26943.1,
      "query_h_norm_ratio": 0.4625485002994537,
      "relation_loss": 0.011845844238996506,
      "selected_pairs": 26943.1,
      "step": 4000,
      "student_distance_mean": 32.13370761871338,
      "teacher_distance_mean": 0.8514031767845154,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.011845844238996506,
      "weighted_relation_loss": 0.011845844238996506
    },
    {
      "action_query_batch_std": 0.5011854022741318,
      "action_query_norm": 44.738748168945314,
      "action_relation_loss": 0.012656252179294825,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.7511693171188027,
      "flow_matching_loss": 0.06173973940312862,
      "grad_norm": 0.16000927984714508,
      "h_stock_valid_norm": 97.57085952758788,
      "invalid_tail_samples": 21.7,
      "learning_rate": 9.809987889444522e-05,
      "loss": 0.0744,
      "positive_teacher_pairs": 27343.5,
      "query_h_norm_ratio": 0.45853704810142515,
      "relation_loss": 0.012656252179294825,
      "selected_pairs": 27343.5,
      "step": 4010,
      "student_distance_mean": 32.66053581237793,
      "teacher_distance_mean": 0.854298448562622,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.012656252179294825,
      "weighted_relation_loss": 0.012656252179294825
    },
    {
      "action_query_batch_std": 0.48903111517429354,
      "action_query_norm": 44.72982559204102,
      "action_relation_loss": 0.012366307713091374,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.7605238540692234,
      "flow_matching_loss": 0.0641534835100174,
      "grad_norm": 0.16483472287654877,
      "h_stock_valid_norm": 98.33391647338867,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.808479990363282e-05,
      "loss": 0.0765,
      "positive_teacher_pairs": 26659.5,
      "query_h_norm_ratio": 0.4548788607120514,
      "relation_loss": 0.012366307713091374,
      "selected_pairs": 26659.5,
      "step": 4020,
      "student_distance_mean": 31.967527389526367,
      "teacher_distance_mean": 0.8535185158252716,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.012366307713091374,
      "weighted_relation_loss": 0.012366307713091374
    },
    {
      "action_query_batch_std": 0.48919505774974825,
      "action_query_norm": 44.72597961425781,
      "action_relation_loss": 0.011889457330107689,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.7698783910196445,
      "flow_matching_loss": 0.06545942090451717,
      "grad_norm": 0.22778578102588654,
      "h_stock_valid_norm": 98.20366516113282,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.806966248525445e-05,
      "loss": 0.0773,
      "positive_teacher_pairs": 26875.0,
      "query_h_norm_ratio": 0.45544224977493286,
      "relation_loss": 0.011889457330107689,
      "selected_pairs": 26875.0,
      "step": 4030,
      "student_distance_mean": 31.91651153564453,
      "teacher_distance_mean": 0.8642067015171051,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.011889457330107689,
      "weighted_relation_loss": 0.011889457330107689
    },
    {
      "action_query_batch_std": 0.4928360879421234,
      "action_query_norm": 44.725968551635745,
      "action_relation_loss": 0.011894396692514419,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.7792329279700656,
      "flow_matching_loss": 0.06244695074856281,
      "grad_norm": 0.2144189178943634,
      "h_stock_valid_norm": 96.80264053344726,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.805446665770348e-05,
      "loss": 0.0743,
      "positive_teacher_pairs": 26783.4,
      "query_h_norm_ratio": 0.46204309463500975,
      "relation_loss": 0.011894396692514419,
      "selected_pairs": 26783.4,
      "step": 4040,
      "student_distance_mean": 32.23985824584961,
      "teacher_distance_mean": 0.8530355513095855,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.011894396692514419,
      "weighted_relation_loss": 0.011894396692514419
    },
    {
      "action_query_batch_std": 0.48980326652526857,
      "action_query_norm": 44.723136520385744,
      "action_relation_loss": 0.011566008999943733,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.7885874649204867,
      "flow_matching_loss": 0.06484203301370144,
      "grad_norm": 0.2089051902294159,
      "h_stock_valid_norm": 96.15190963745117,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.803921243944429e-05,
      "loss": 0.0764,
      "positive_teacher_pairs": 27195.7,
      "query_h_norm_ratio": 0.4651321232318878,
      "relation_loss": 0.011566008999943733,
      "selected_pairs": 27195.7,
      "step": 4050,
      "student_distance_mean": 31.937775230407716,
      "teacher_distance_mean": 0.8537177801132202,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.011566008999943733,
      "weighted_relation_loss": 0.011566008999943733
    },
    {
      "action_query_batch_std": 0.4977730095386505,
      "action_query_norm": 44.72646560668945,
      "action_relation_loss": 0.01211509583517909,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.7979420018709074,
      "flow_matching_loss": 0.06573328860104084,
      "grad_norm": 0.15724951028823853,
      "h_stock_valid_norm": 97.17886199951172,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.802389984901218e-05,
      "loss": 0.0778,
      "positive_teacher_pairs": 26687.4,
      "query_h_norm_ratio": 0.46026033759117124,
      "relation_loss": 0.01211509583517909,
      "selected_pairs": 26687.4,
      "step": 4060,
      "student_distance_mean": 32.39351825714111,
      "teacher_distance_mean": 0.8590235590934754,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.01211509583517909,
      "weighted_relation_loss": 0.01211509583517909
    },
    {
      "action_query_batch_std": 0.4796783685684204,
      "action_query_norm": 44.708357620239255,
      "action_relation_loss": 0.012137691397219897,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.8072965388213285,
      "flow_matching_loss": 0.06463386379182338,
      "grad_norm": 0.20134060084819794,
      "h_stock_valid_norm": 97.2824104309082,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.80085289050134e-05,
      "loss": 0.0768,
      "positive_teacher_pairs": 26848.1,
      "query_h_norm_ratio": 0.4595796674489975,
      "relation_loss": 0.012137691397219897,
      "selected_pairs": 26848.1,
      "step": 4070,
      "student_distance_mean": 31.215640449523924,
      "teacher_distance_mean": 0.8438610851764679,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.012137691397219897,
      "weighted_relation_loss": 0.012137691397219897
    },
    {
      "action_query_batch_std": 0.476587301492691,
      "action_query_norm": 44.70201530456543,
      "action_relation_loss": 0.01251853546127677,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.816651075771749,
      "flow_matching_loss": 0.06661474406719207,
      "grad_norm": 0.20474332571029663,
      "h_stock_valid_norm": 98.05000228881836,
      "invalid_tail_samples": 22.1,
      "learning_rate": 9.799309962612508e-05,
      "loss": 0.0791,
      "positive_teacher_pairs": 27242.0,
      "query_h_norm_ratio": 0.45591698586940765,
      "relation_loss": 0.01251853546127677,
      "selected_pairs": 27242.0,
      "step": 4080,
      "student_distance_mean": 30.946748542785645,
      "teacher_distance_mean": 0.8461915612220764,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.01251853546127677,
      "weighted_relation_loss": 0.01251853546127677
    },
    {
      "action_query_batch_std": 0.49646500051021575,
      "action_query_norm": 44.72224349975586,
      "action_relation_loss": 0.012071727588772774,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.82600561272217,
      "flow_matching_loss": 0.06387505270540714,
      "grad_norm": 0.12817952036857605,
      "h_stock_valid_norm": 98.97472915649413,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.797761203109527e-05,
      "loss": 0.0759,
      "positive_teacher_pairs": 26869.9,
      "query_h_norm_ratio": 0.4518586307764053,
      "relation_loss": 0.012071727588772774,
      "selected_pairs": 26869.9,
      "step": 4090,
      "student_distance_mean": 32.41423988342285,
      "teacher_distance_mean": 0.8660725295543671,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.012071727588772774,
      "weighted_relation_loss": 0.012071727588772774
    },
    {
      "action_query_batch_std": 0.49525006413459777,
      "action_query_norm": 44.71897087097168,
      "action_relation_loss": 0.011929974704980851,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.8353601496725913,
      "flow_matching_loss": 0.0627901803702116,
      "grad_norm": 0.14107996225357056,
      "h_stock_valid_norm": 99.22732772827149,
      "invalid_tail_samples": 22.0,
      "learning_rate": 9.796206613874283e-05,
      "loss": 0.0747,
      "positive_teacher_pairs": 27269.9,
      "query_h_norm_ratio": 0.4506823092699051,
      "relation_loss": 0.011929974704980851,
      "selected_pairs": 27269.9,
      "step": 4100,
      "student_distance_mean": 32.32077198028564,
      "teacher_distance_mean": 0.8584900915622711,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.011929974704980851,
      "weighted_relation_loss": 0.011929974704980851
    },
    {
      "action_query_batch_std": 0.4931157737970352,
      "action_query_norm": 44.71880111694336,
      "action_relation_loss": 0.012348782457411289,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.8447146866230124,
      "flow_matching_loss": 0.06276600584387779,
      "grad_norm": 0.1523396223783493,
      "h_stock_valid_norm": 97.42144241333008,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.794646196795754e-05,
      "loss": 0.0751,
      "positive_teacher_pairs": 27082.6,
      "query_h_norm_ratio": 0.45903458893299104,
      "relation_loss": 0.012348782457411289,
      "selected_pairs": 27082.6,
      "step": 4110,
      "student_distance_mean": 32.079802703857425,
      "teacher_distance_mean": 0.8581604957580566,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.012348782457411289,
      "weighted_relation_loss": 0.012348782457411289
    },
    {
      "action_query_batch_std": 0.4892886370420456,
      "action_query_norm": 44.709743118286134,
      "action_relation_loss": 0.011934395786374808,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.854069223573433,
      "flow_matching_loss": 0.06360470689833164,
      "grad_norm": 0.17791353166103363,
      "h_stock_valid_norm": 97.14489059448242,
      "invalid_tail_samples": 26.0,
      "learning_rate": 9.793079953769987e-05,
      "loss": 0.0755,
      "positive_teacher_pairs": 26350.1,
      "query_h_norm_ratio": 0.4602422326803207,
      "relation_loss": 0.011934395786374808,
      "selected_pairs": 26350.1,
      "step": 4120,
      "student_distance_mean": 31.90547275543213,
      "teacher_distance_mean": 0.8556612491607666,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.011934395786374808,
      "weighted_relation_loss": 0.011934395786374808
    },
    {
      "action_query_batch_std": 0.4843148797750473,
      "action_query_norm": 44.705759811401364,
      "action_relation_loss": 0.012219623941928149,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.863423760523854,
      "flow_matching_loss": 0.06304322592914105,
      "grad_norm": 0.13786223530769348,
      "h_stock_valid_norm": 98.48421478271484,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.79150788670012e-05,
      "loss": 0.0753,
      "positive_teacher_pairs": 26756.2,
      "query_h_norm_ratio": 0.4539639979600906,
      "relation_loss": 0.012219623941928149,
      "selected_pairs": 26756.2,
      "step": 4130,
      "student_distance_mean": 31.59982509613037,
      "teacher_distance_mean": 0.8467204988002777,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.012219623941928149,
      "weighted_relation_loss": 0.012219623941928149
    },
    {
      "action_query_batch_std": 0.5021383345127106,
      "action_query_norm": 44.71746139526367,
      "action_relation_loss": 0.011798292957246303,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.872778297474275,
      "flow_matching_loss": 0.06211251094937324,
      "grad_norm": 0.1456925868988037,
      "h_stock_valid_norm": 98.65587005615234,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.78992999749636e-05,
      "loss": 0.0739,
      "positive_teacher_pairs": 27103.1,
      "query_h_norm_ratio": 0.45326857566833495,
      "relation_loss": 0.011798292957246303,
      "selected_pairs": 27103.1,
      "step": 4140,
      "student_distance_mean": 32.5983283996582,
      "teacher_distance_mean": 0.8560358762741089,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.011798292957246303,
      "weighted_relation_loss": 0.011798292957246303
    },
    {
      "action_query_batch_std": 0.4957401782274246,
      "action_query_norm": 44.70914001464844,
      "action_relation_loss": 0.011511268652975559,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.882132834424696,
      "flow_matching_loss": 0.05925083942711353,
      "grad_norm": 0.14211450517177582,
      "h_stock_valid_norm": 98.44928359985352,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.788346288075994e-05,
      "loss": 0.0708,
      "positive_teacher_pairs": 26664.0,
      "query_h_norm_ratio": 0.454136723279953,
      "relation_loss": 0.011511268652975559,
      "selected_pairs": 26664.0,
      "step": 4150,
      "student_distance_mean": 32.20151214599609,
      "teacher_distance_mean": 0.8560909986495971,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.011511268652975559,
      "weighted_relation_loss": 0.011511268652975559
    },
    {
      "action_query_batch_std": 0.5020980685949326,
      "action_query_norm": 44.70951538085937,
      "action_relation_loss": 0.012686720583587885,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.891487371375117,
      "flow_matching_loss": 0.06484077535569668,
      "grad_norm": 0.16020163893699646,
      "h_stock_valid_norm": 97.36874008178711,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.786756760363373e-05,
      "loss": 0.0775,
      "positive_teacher_pairs": 26664.3,
      "query_h_norm_ratio": 0.459189835190773,
      "relation_loss": 0.012686720583587885,
      "selected_pairs": 26664.3,
      "step": 4160,
      "student_distance_mean": 32.75017471313477,
      "teacher_distance_mean": 0.8549234867095947,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.012686720583587885,
      "weighted_relation_loss": 0.012686720583587885
    },
    {
      "action_query_batch_std": 0.48687762916088106,
      "action_query_norm": 44.692475128173825,
      "action_relation_loss": 0.011550105549395085,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.900841908325538,
      "flow_matching_loss": 0.06322018317878246,
      "grad_norm": 0.19621816277503967,
      "h_stock_valid_norm": 97.45462112426758,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.78516141628993e-05,
      "loss": 0.0748,
      "positive_teacher_pairs": 26811.7,
      "query_h_norm_ratio": 0.4586184620857239,
      "relation_loss": 0.011550105549395085,
      "selected_pairs": 26811.7,
      "step": 4170,
      "student_distance_mean": 31.658778953552247,
      "teacher_distance_mean": 0.8499408423900604,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.011550105549395085,
      "weighted_relation_loss": 0.011550105549395085
    },
    {
      "action_query_batch_std": 0.4834982603788376,
      "action_query_norm": 44.68993873596192,
      "action_relation_loss": 0.01161039350554347,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.9101964452759588,
      "flow_matching_loss": 0.0634170725941658,
      "grad_norm": 0.1514216810464859,
      "h_stock_valid_norm": 98.11413116455078,
      "invalid_tail_samples": 27.2,
      "learning_rate": 9.783560257794154e-05,
      "loss": 0.075,
      "positive_teacher_pairs": 26062.4,
      "query_h_norm_ratio": 0.455493700504303,
      "relation_loss": 0.01161039350554347,
      "selected_pairs": 26062.4,
      "step": 4180,
      "student_distance_mean": 31.71172389984131,
      "teacher_distance_mean": 0.860990172624588,
      "valid_action_samples": 228.8,
      "weighted_auxiliary_loss": 0.01161039350554347,
      "weighted_relation_loss": 0.01161039350554347
    },
    {
      "action_query_batch_std": 0.4911423534154892,
      "action_query_norm": 44.69410667419434,
      "action_relation_loss": 0.012151943519711494,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.91955098222638,
      "flow_matching_loss": 0.06475902013480664,
      "grad_norm": 0.2146235853433609,
      "h_stock_valid_norm": 97.68422775268554,
      "invalid_tail_samples": 27.4,
      "learning_rate": 9.781953286821603e-05,
      "loss": 0.0769,
      "positive_teacher_pairs": 26019.4,
      "query_h_norm_ratio": 0.45755078792572024,
      "relation_loss": 0.012151943519711494,
      "selected_pairs": 26019.4,
      "step": 4190,
      "student_distance_mean": 32.12355117797851,
      "teacher_distance_mean": 0.8541475296020508,
      "valid_action_samples": 228.6,
      "weighted_auxiliary_loss": 0.012151943519711494,
      "weighted_relation_loss": 0.012151943519711494
    },
    {
      "action_query_batch_std": 0.4886058121919632,
      "action_query_norm": 44.696729278564455,
      "action_relation_loss": 0.011705342587083578,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.9289055191768005,
      "flow_matching_loss": 0.06346636079251766,
      "grad_norm": 0.18153846263885498,
      "h_stock_valid_norm": 96.54879760742188,
      "invalid_tail_samples": 22.6,
      "learning_rate": 9.780340505324901e-05,
      "loss": 0.0752,
      "positive_teacher_pairs": 27131.4,
      "query_h_norm_ratio": 0.4629499793052673,
      "relation_loss": 0.011705342587083578,
      "selected_pairs": 27131.4,
      "step": 4200,
      "student_distance_mean": 31.793663024902344,
      "teacher_distance_mean": 0.8545210659503937,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.011705342587083578,
      "weighted_relation_loss": 0.011705342587083578
    },
    {
      "action_query_batch_std": 0.48631568551063536,
      "action_query_norm": 44.69388122558594,
      "action_relation_loss": 0.012468181364238263,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.9382600561272216,
      "flow_matching_loss": 0.06512399278581142,
      "grad_norm": 0.1259641796350479,
      "h_stock_valid_norm": 97.86862640380859,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.778721915263727e-05,
      "loss": 0.0776,
      "positive_teacher_pairs": 26735.3,
      "query_h_norm_ratio": 0.45668198764324186,
      "relation_loss": 0.012468181364238263,
      "selected_pairs": 26735.3,
      "step": 4210,
      "student_distance_mean": 31.803791427612303,
      "teacher_distance_mean": 0.849684226512909,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.012468181364238263,
      "weighted_relation_loss": 0.012468181364238263
    },
    {
      "action_query_batch_std": 0.48850697576999663,
      "action_query_norm": 44.69288444519043,
      "action_relation_loss": 0.011577840335667134,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.9476145930776427,
      "flow_matching_loss": 0.06156722269952297,
      "grad_norm": 0.1836107075214386,
      "h_stock_valid_norm": 98.32676315307617,
      "invalid_tail_samples": 26.0,
      "learning_rate": 9.777097518604824e-05,
      "loss": 0.0731,
      "positive_teacher_pairs": 26348.0,
      "query_h_norm_ratio": 0.454537969827652,
      "relation_loss": 0.011577840335667134,
      "selected_pairs": 26348.0,
      "step": 4220,
      "student_distance_mean": 31.957689094543458,
      "teacher_distance_mean": 0.859817099571228,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.011577840335667134,
      "weighted_relation_loss": 0.011577840335667134
    },
    {
      "action_query_batch_std": 0.4939830362796783,
      "action_query_norm": 44.69771156311035,
      "action_relation_loss": 0.011711669247597456,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.956969130028064,
      "flow_matching_loss": 0.061564536020159724,
      "grad_norm": 0.20274661481380463,
      "h_stock_valid_norm": 97.2482177734375,
      "invalid_tail_samples": 26.2,
      "learning_rate": 9.775467317321984e-05,
      "loss": 0.0733,
      "positive_teacher_pairs": 26295.2,
      "query_h_norm_ratio": 0.4596293121576309,
      "relation_loss": 0.011711669247597456,
      "selected_pairs": 26295.2,
      "step": 4230,
      "student_distance_mean": 32.30258560180664,
      "teacher_distance_mean": 0.8584621667861938,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.011711669247597456,
      "weighted_relation_loss": 0.011711669247597456
    },
    {
      "action_query_batch_std": 0.4864064782857895,
      "action_query_norm": 44.687329864501955,
      "action_relation_loss": 0.011956537701189518,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.9663236669784845,
      "flow_matching_loss": 0.060270878672599795,
      "grad_norm": 0.15266776084899902,
      "h_stock_valid_norm": 97.91023864746094,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.773831313396055e-05,
      "loss": 0.0722,
      "positive_teacher_pairs": 26737.3,
      "query_h_norm_ratio": 0.4564189314842224,
      "relation_loss": 0.011956537701189518,
      "selected_pairs": 26737.3,
      "step": 4240,
      "student_distance_mean": 31.709935188293457,
      "teacher_distance_mean": 0.8500476002693176,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.011956537701189518,
      "weighted_relation_loss": 0.011956537701189518
    },
    {
      "action_query_batch_std": 0.48514932096004487,
      "action_query_norm": 44.67906951904297,
      "action_relation_loss": 0.011764228995889425,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.9756782039289056,
      "flow_matching_loss": 0.0629463616758585,
      "grad_norm": 0.1957734078168869,
      "h_stock_valid_norm": 98.65422897338867,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.77218950881494e-05,
      "loss": 0.0747,
      "positive_teacher_pairs": 26757.6,
      "query_h_norm_ratio": 0.45289023220539093,
      "relation_loss": 0.011764228995889425,
      "selected_pairs": 26757.6,
      "step": 4250,
      "student_distance_mean": 31.64683418273926,
      "teacher_distance_mean": 0.8525583028793335,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.011764228995889425,
      "weighted_relation_loss": 0.011764228995889425
    },
    {
      "action_query_batch_std": 0.4786090701818466,
      "action_query_norm": 44.676635360717775,
      "action_relation_loss": 0.01237577125430107,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.9850327408793262,
      "flow_matching_loss": 0.062339679896831514,
      "grad_norm": 0.14075328409671783,
      "h_stock_valid_norm": 98.47944412231445,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.770541905573583e-05,
      "loss": 0.0747,
      "positive_teacher_pairs": 27221.4,
      "query_h_norm_ratio": 0.4536678224802017,
      "relation_loss": 0.01237577125430107,
      "selected_pairs": 27221.4,
      "step": 4260,
      "student_distance_mean": 31.154992294311523,
      "teacher_distance_mean": 0.8456102311611176,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.01237577125430107,
      "weighted_relation_loss": 0.01237577125430107
    },
    {
      "action_query_batch_std": 0.4906246572732925,
      "action_query_norm": 44.67830696105957,
      "action_relation_loss": 0.011472592037171126,
      "auxiliary_loss_weight": 1.0,
      "epoch": 3.9943872778297473,
      "flow_matching_loss": 0.063986511901021,
      "grad_norm": 0.21975094079971313,
      "h_stock_valid_norm": 97.65817413330078,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.768888505673976e-05,
      "loss": 0.0755,
      "positive_teacher_pairs": 26926.1,
      "query_h_norm_ratio": 0.45750282406806947,
      "relation_loss": 0.011472592037171126,
      "selected_pairs": 26926.1,
      "step": 4270,
      "student_distance_mean": 31.95136375427246,
      "teacher_distance_mean": 0.8535879731178284,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.011472592037171126,
      "weighted_relation_loss": 0.011472592037171126
    },
    {
      "action_query_batch_std": 0.49029124081134795,
      "action_query_norm": 44.684488677978514,
      "action_relation_loss": 0.011864703055471182,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.003741814780168,
      "flow_matching_loss": 0.062429235130548474,
      "grad_norm": 0.18626056611537933,
      "h_stock_valid_norm": 97.80342178344726,
      "invalid_tail_samples": 22.1,
      "learning_rate": 9.767229311125162e-05,
      "loss": 0.0743,
      "positive_teacher_pairs": 24219.5,
      "query_h_norm_ratio": 0.4568846493959427,
      "relation_loss": 0.011864703055471182,
      "selected_pairs": 24219.5,
      "step": 4280,
      "student_distance_mean": 31.959187507629395,
      "teacher_distance_mean": 0.8598978579044342,
      "valid_action_samples": 214.0,
      "weighted_auxiliary_loss": 0.011864703055471182,
      "weighted_relation_loss": 0.011864703055471182
    },
    {
      "action_query_batch_std": 0.48132733404636385,
      "action_query_norm": 44.67593116760254,
      "action_relation_loss": 0.01155067915096879,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.0130963517305895,
      "flow_matching_loss": 0.05924685075879097,
      "grad_norm": 0.2325945496559143,
      "h_stock_valid_norm": 97.62627792358398,
      "invalid_tail_samples": 23.0,
      "learning_rate": 9.765564323943211e-05,
      "loss": 0.0708,
      "positive_teacher_pairs": 27033.7,
      "query_h_norm_ratio": 0.45762604773044585,
      "relation_loss": 0.01155067915096879,
      "selected_pairs": 27033.7,
      "step": 4290,
      "student_distance_mean": 31.44678726196289,
      "teacher_distance_mean": 0.8497637033462524,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.01155067915096879,
      "weighted_relation_loss": 0.01155067915096879
    },
    {
      "action_query_batch_std": 0.4839516252279282,
      "action_query_norm": 44.67446784973144,
      "action_relation_loss": 0.010956278629601001,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.022450888681011,
      "flow_matching_loss": 0.06086861714720726,
      "grad_norm": 0.24077951908111572,
      "h_stock_valid_norm": 98.57005004882812,
      "invalid_tail_samples": 22.7,
      "learning_rate": 9.763893546151244e-05,
      "loss": 0.0718,
      "positive_teacher_pairs": 27106.7,
      "query_h_norm_ratio": 0.4532294899225235,
      "relation_loss": 0.010956278629601001,
      "selected_pairs": 27106.7,
      "step": 4300,
      "student_distance_mean": 31.61969566345215,
      "teacher_distance_mean": 0.8561823725700378,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.010956278629601001,
      "weighted_relation_loss": 0.010956278629601001
    },
    {
      "action_query_batch_std": 0.49132702350616453,
      "action_query_norm": 44.68361396789551,
      "action_relation_loss": 0.011174877174198627,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.031805425631431,
      "flow_matching_loss": 0.06387092508375644,
      "grad_norm": 0.2245328277349472,
      "h_stock_valid_norm": 99.06171875,
      "invalid_tail_samples": 25.2,
      "learning_rate": 9.762216979779412e-05,
      "loss": 0.075,
      "positive_teacher_pairs": 26526.0,
      "query_h_norm_ratio": 0.45107578337192533,
      "relation_loss": 0.011174877174198627,
      "selected_pairs": 26526.0,
      "step": 4310,
      "student_distance_mean": 32.04370059967041,
      "teacher_distance_mean": 0.8548923313617707,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.011174877174198627,
      "weighted_relation_loss": 0.011174877174198627
    },
    {
      "action_query_batch_std": 0.49131631255149844,
      "action_query_norm": 44.680717849731444,
      "action_relation_loss": 0.011678330041468144,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.041159962581852,
      "flow_matching_loss": 0.06245320998132229,
      "grad_norm": 0.20551294088363647,
      "h_stock_valid_norm": 98.31324615478516,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.760534626864902e-05,
      "loss": 0.0741,
      "positive_teacher_pairs": 26904.8,
      "query_h_norm_ratio": 0.4544881969690323,
      "relation_loss": 0.011678330041468144,
      "selected_pairs": 26904.8,
      "step": 4320,
      "student_distance_mean": 32.102206993103025,
      "teacher_distance_mean": 0.8691844344139099,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.011678330041468144,
      "weighted_relation_loss": 0.011678330041468144
    },
    {
      "action_query_batch_std": 0.48587165474891664,
      "action_query_norm": 44.67300033569336,
      "action_relation_loss": 0.01121129048988223,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.050514499532273,
      "flow_matching_loss": 0.06315078362822532,
      "grad_norm": 0.17519234120845795,
      "h_stock_valid_norm": 97.45741958618164,
      "invalid_tail_samples": 22.6,
      "learning_rate": 9.758846489451931e-05,
      "loss": 0.0744,
      "positive_teacher_pairs": 27129.6,
      "query_h_norm_ratio": 0.45838998854160307,
      "relation_loss": 0.01121129048988223,
      "selected_pairs": 27129.6,
      "step": 4330,
      "student_distance_mean": 31.71350841522217,
      "teacher_distance_mean": 0.854158878326416,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.01121129048988223,
      "weighted_relation_loss": 0.01121129048988223
    },
    {
      "action_query_batch_std": 0.4892267733812332,
      "action_query_norm": 44.67500686645508,
      "action_relation_loss": 0.01163405403494835,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.059869036482694,
      "flow_matching_loss": 0.05856317691504955,
      "grad_norm": 0.13761676847934723,
      "h_stock_valid_norm": 97.73229217529297,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.757152569591748e-05,
      "loss": 0.0702,
      "positive_teacher_pairs": 26892.2,
      "query_h_norm_ratio": 0.4571189105510712,
      "relation_loss": 0.01163405403494835,
      "selected_pairs": 26892.2,
      "step": 4340,
      "student_distance_mean": 32.05109405517578,
      "teacher_distance_mean": 0.8538895905017853,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.01163405403494835,
      "weighted_relation_loss": 0.01163405403494835
    },
    {
      "action_query_batch_std": 0.4859079569578171,
      "action_query_norm": 44.67094268798828,
      "action_relation_loss": 0.011473249830305576,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.069223573433115,
      "flow_matching_loss": 0.061414001509547234,
      "grad_norm": 0.18441987037658691,
      "h_stock_valid_norm": 97.80487518310547,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.75545286934262e-05,
      "loss": 0.0729,
      "positive_teacher_pairs": 26662.3,
      "query_h_norm_ratio": 0.4567373186349869,
      "relation_loss": 0.011473249830305576,
      "selected_pairs": 26662.3,
      "step": 4350,
      "student_distance_mean": 31.576799392700195,
      "teacher_distance_mean": 0.8594427227973938,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.011473249830305576,
      "weighted_relation_loss": 0.011473249830305576
    },
    {
      "action_query_batch_std": 0.4842667281627655,
      "action_query_norm": 44.665875625610354,
      "action_relation_loss": 0.011399702914059163,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.078578110383536,
      "flow_matching_loss": 0.06104168593883515,
      "grad_norm": 0.1651957482099533,
      "h_stock_valid_norm": 97.95064315795898,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.753747390769847e-05,
      "loss": 0.0724,
      "positive_teacher_pairs": 26919.9,
      "query_h_norm_ratio": 0.4560112118721008,
      "relation_loss": 0.011399702914059163,
      "selected_pairs": 26919.9,
      "step": 4360,
      "student_distance_mean": 31.677694892883302,
      "teacher_distance_mean": 0.8522559344768524,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.011399702914059163,
      "weighted_relation_loss": 0.011399702914059163
    },
    {
      "action_query_batch_std": 0.4920110672712326,
      "action_query_norm": 44.671090316772464,
      "action_relation_loss": 0.011609708424657583,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.087932647333957,
      "flow_matching_loss": 0.060176413506269455,
      "grad_norm": 0.19707530736923218,
      "h_stock_valid_norm": 98.1568817138672,
      "invalid_tail_samples": 25.2,
      "learning_rate": 9.752036135945744e-05,
      "loss": 0.0718,
      "positive_teacher_pairs": 26532.4,
      "query_h_norm_ratio": 0.4551044166088104,
      "relation_loss": 0.011609708424657583,
      "selected_pairs": 26532.4,
      "step": 4370,
      "student_distance_mean": 31.99028263092041,
      "teacher_distance_mean": 0.8547831535339355,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.011609708424657583,
      "weighted_relation_loss": 0.011609708424657583
    },
    {
      "action_query_batch_std": 0.49369449019432066,
      "action_query_norm": 44.672459030151366,
      "action_relation_loss": 0.011393618118017913,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.097287184284378,
      "flow_matching_loss": 0.06009755730628967,
      "grad_norm": 0.21613436937332153,
      "h_stock_valid_norm": 97.11608428955078,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.750319106949649e-05,
      "loss": 0.0715,
      "positive_teacher_pairs": 26993.3,
      "query_h_norm_ratio": 0.46000429689884187,
      "relation_loss": 0.011393618118017913,
      "selected_pairs": 26993.3,
      "step": 4380,
      "student_distance_mean": 31.97266902923584,
      "teacher_distance_mean": 0.8566392600536347,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.011393618118017913,
      "weighted_relation_loss": 0.011393618118017913
    },
    {
      "action_query_batch_std": 0.4895231038331985,
      "action_query_norm": 44.66709251403809,
      "action_relation_loss": 0.011286083888262511,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.106641721234799,
      "flow_matching_loss": 0.06438359543681145,
      "grad_norm": 0.2154623419046402,
      "h_stock_valid_norm": 96.63888778686524,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.748596305867913e-05,
      "loss": 0.0757,
      "positive_teacher_pairs": 27151.0,
      "query_h_norm_ratio": 0.4622103750705719,
      "relation_loss": 0.011286083888262511,
      "selected_pairs": 27151.0,
      "step": 4390,
      "student_distance_mean": 31.823552513122557,
      "teacher_distance_mean": 0.8620568454265595,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.011286083888262511,
      "weighted_relation_loss": 0.011286083888262511
    },
    {
      "action_query_batch_std": 0.4856409221887589,
      "action_query_norm": 44.66225128173828,
      "action_relation_loss": 0.010999523475766182,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.11599625818522,
      "flow_matching_loss": 0.062426619976758954,
      "grad_norm": 0.16652366518974304,
      "h_stock_valid_norm": 97.11091079711915,
      "invalid_tail_samples": 26.2,
      "learning_rate": 9.746867734793903e-05,
      "loss": 0.0734,
      "positive_teacher_pairs": 26302.5,
      "query_h_norm_ratio": 0.459913381934166,
      "relation_loss": 0.010999523475766182,
      "selected_pairs": 26302.5,
      "step": 4400,
      "student_distance_mean": 31.93087215423584,
      "teacher_distance_mean": 0.8513237237930298,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.010999523475766182,
      "weighted_relation_loss": 0.010999523475766182
    },
    {
      "action_query_batch_std": 0.48037177622318267,
      "action_query_norm": 44.6569278717041,
      "action_relation_loss": 0.01149112805724144,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.125350795135641,
      "flow_matching_loss": 0.059937655180692676,
      "grad_norm": 0.19937966763973236,
      "h_stock_valid_norm": 97.31497573852539,
      "invalid_tail_samples": 22.7,
      "learning_rate": 9.745133395827993e-05,
      "loss": 0.0714,
      "positive_teacher_pairs": 27102.3,
      "query_h_norm_ratio": 0.4588987648487091,
      "relation_loss": 0.01149112805724144,
      "selected_pairs": 27102.3,
      "step": 4410,
      "student_distance_mean": 31.305478668212892,
      "teacher_distance_mean": 0.8472659826278687,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.01149112805724144,
      "weighted_relation_loss": 0.01149112805724144
    },
    {
      "action_query_batch_std": 0.4852568119764328,
      "action_query_norm": 44.65861015319824,
      "action_relation_loss": 0.011739479284733534,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.134705332086062,
      "flow_matching_loss": 0.06057310551404953,
      "grad_norm": 0.1585836112499237,
      "h_stock_valid_norm": 96.86598434448243,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.743393291077572e-05,
      "loss": 0.0723,
      "positive_teacher_pairs": 26930.6,
      "query_h_norm_ratio": 0.46103730201721194,
      "relation_loss": 0.011739479284733534,
      "selected_pairs": 26930.6,
      "step": 4420,
      "student_distance_mean": 31.780502128601075,
      "teacher_distance_mean": 0.8570042788982392,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.011739479284733534,
      "weighted_relation_loss": 0.011739479284733534
    },
    {
      "action_query_batch_std": 0.4903068631887436,
      "action_query_norm": 44.66474571228027,
      "action_relation_loss": 0.011205479968339205,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.144059869036482,
      "flow_matching_loss": 0.06057635992765427,
      "grad_norm": 0.1590687483549118,
      "h_stock_valid_norm": 98.12592010498047,
      "invalid_tail_samples": 25.2,
      "learning_rate": 9.741647422657028e-05,
      "loss": 0.0718,
      "positive_teacher_pairs": 26530.8,
      "query_h_norm_ratio": 0.4552041858434677,
      "relation_loss": 0.011205479968339205,
      "selected_pairs": 26530.8,
      "step": 4430,
      "student_distance_mean": 32.07075862884521,
      "teacher_distance_mean": 0.8581833899021148,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.011205479968339205,
      "weighted_relation_loss": 0.011205479968339205
    },
    {
      "action_query_batch_std": 0.4867987990379333,
      "action_query_norm": 44.6582332611084,
      "action_relation_loss": 0.011650414485484362,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.153414405986903,
      "flow_matching_loss": 0.05809453204274177,
      "grad_norm": 0.16929128766059875,
      "h_stock_valid_norm": 98.97723617553712,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.739895792687758e-05,
      "loss": 0.0697,
      "positive_teacher_pairs": 26760.8,
      "query_h_norm_ratio": 0.45119922757148745,
      "relation_loss": 0.011650414485484362,
      "selected_pairs": 26760.8,
      "step": 4440,
      "student_distance_mean": 31.66482620239258,
      "teacher_distance_mean": 0.848652821779251,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.011650414485484362,
      "weighted_relation_loss": 0.011650414485484362
    },
    {
      "action_query_batch_std": 0.4836955636739731,
      "action_query_norm": 44.65111236572265,
      "action_relation_loss": 0.011221872549504042,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.1627689429373245,
      "flow_matching_loss": 0.05858741775155067,
      "grad_norm": 0.1579994559288025,
      "h_stock_valid_norm": 98.44500961303712,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.738138403298157e-05,
      "loss": 0.0698,
      "positive_teacher_pairs": 26949.0,
      "query_h_norm_ratio": 0.45357124209403993,
      "relation_loss": 0.011221872549504042,
      "selected_pairs": 26949.0,
      "step": 4450,
      "student_distance_mean": 31.55134048461914,
      "teacher_distance_mean": 0.8586035370826721,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.011221872549504042,
      "weighted_relation_loss": 0.011221872549504042
    },
    {
      "action_query_batch_std": 0.480848103761673,
      "action_query_norm": 44.64903182983399,
      "action_relation_loss": 0.01100112646818161,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.172123479887746,
      "flow_matching_loss": 0.06281192265450955,
      "grad_norm": 0.11683952063322067,
      "h_stock_valid_norm": 98.03258819580078,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.736375256623619e-05,
      "loss": 0.0738,
      "positive_teacher_pairs": 27174.7,
      "query_h_norm_ratio": 0.45545746982097624,
      "relation_loss": 0.01100112646818161,
      "selected_pairs": 27174.7,
      "step": 4460,
      "student_distance_mean": 31.389215850830077,
      "teacher_distance_mean": 0.8532196283340454,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.01100112646818161,
      "weighted_relation_loss": 0.01100112646818161
    },
    {
      "action_query_batch_std": 0.47904442250728607,
      "action_query_norm": 44.645391845703124,
      "action_relation_loss": 0.011569245345890522,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.181478016838167,
      "flow_matching_loss": 0.06273704618215561,
      "grad_norm": 0.18029938638210297,
      "h_stock_valid_norm": 98.59153900146484,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.734606354806533e-05,
      "loss": 0.0743,
      "positive_teacher_pairs": 26616.6,
      "query_h_norm_ratio": 0.4528328269720078,
      "relation_loss": 0.011569245345890522,
      "selected_pairs": 26616.6,
      "step": 4470,
      "student_distance_mean": 31.507681846618652,
      "teacher_distance_mean": 0.8553594470024108,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.011569245345890522,
      "weighted_relation_loss": 0.011569245345890522
    },
    {
      "action_query_batch_std": 0.48095901906490324,
      "action_query_norm": 44.64445838928223,
      "action_relation_loss": 0.011203652061522008,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.190832553788588,
      "flow_matching_loss": 0.059112031012773514,
      "grad_norm": 0.16249191761016846,
      "h_stock_valid_norm": 98.00116882324218,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.73283169999628e-05,
      "loss": 0.0703,
      "positive_teacher_pairs": 26854.5,
      "query_h_norm_ratio": 0.45555371046066284,
      "relation_loss": 0.011203652061522008,
      "selected_pairs": 26854.5,
      "step": 4480,
      "student_distance_mean": 31.373843574523924,
      "teacher_distance_mean": 0.8572259604930877,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.011203652061522008,
      "weighted_relation_loss": 0.011203652061522008
    },
    {
      "action_query_batch_std": 0.4741243481636047,
      "action_query_norm": 44.640362930297854,
      "action_relation_loss": 0.010691721830517054,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.200187090739009,
      "flow_matching_loss": 0.05920554362237453,
      "grad_norm": 0.16440874338150024,
      "h_stock_valid_norm": 97.74185943603516,
      "invalid_tail_samples": 24.4,
      "learning_rate": 9.731051294349238e-05,
      "loss": 0.0699,
      "positive_teacher_pairs": 26718.2,
      "query_h_norm_ratio": 0.45672535300254824,
      "relation_loss": 0.010691721830517054,
      "selected_pairs": 26718.2,
      "step": 4490,
      "student_distance_mean": 31.04232997894287,
      "teacher_distance_mean": 0.8595609426498413,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.010691721830517054,
      "weighted_relation_loss": 0.010691721830517054
    },
    {
      "action_query_batch_std": 0.4848698228597641,
      "action_query_norm": 44.65585632324219,
      "action_relation_loss": 0.011342097725719213,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.209541627689429,
      "flow_matching_loss": 0.061369079351425174,
      "grad_norm": 0.14659997820854187,
      "h_stock_valid_norm": 96.99055480957031,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.729265140028762e-05,
      "loss": 0.0727,
      "positive_teacher_pairs": 26734.2,
      "query_h_norm_ratio": 0.4604169547557831,
      "relation_loss": 0.011342097725719213,
      "selected_pairs": 26734.2,
      "step": 4500,
      "student_distance_mean": 31.66017017364502,
      "teacher_distance_mean": 0.8598315060138703,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.011342097725719213,
      "weighted_relation_loss": 0.011342097725719213
    },
    {
      "action_query_batch_std": 0.48925178647041323,
      "action_query_norm": 44.657542037963864,
      "action_relation_loss": 0.010468462668359279,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.21889616463985,
      "flow_matching_loss": 0.05865804851055145,
      "grad_norm": 0.15548422932624817,
      "h_stock_valid_norm": 98.09261169433594,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.727473239205201e-05,
      "loss": 0.0691,
      "positive_teacher_pairs": 27220.2,
      "query_h_norm_ratio": 0.4552746504545212,
      "relation_loss": 0.010468462668359279,
      "selected_pairs": 27220.2,
      "step": 4510,
      "student_distance_mean": 31.90668601989746,
      "teacher_distance_mean": 0.868209844827652,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.010468462668359279,
      "weighted_relation_loss": 0.010468462668359279
    },
    {
      "action_query_batch_std": 0.4824057310819626,
      "action_query_norm": 44.643218994140625,
      "action_relation_loss": 0.010889562405645847,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.228250701590271,
      "flow_matching_loss": 0.06317678764462471,
      "grad_norm": 0.1366969794034958,
      "h_stock_valid_norm": 98.46768646240234,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.725675594055883e-05,
      "loss": 0.0741,
      "positive_teacher_pairs": 26849.1,
      "query_h_norm_ratio": 0.45338983833789825,
      "relation_loss": 0.010889562405645847,
      "selected_pairs": 26849.1,
      "step": 4520,
      "student_distance_mean": 31.61036911010742,
      "teacher_distance_mean": 0.8600769460201263,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.010889562405645847,
      "weighted_relation_loss": 0.010889562405645847
    },
    {
      "action_query_batch_std": 0.4878004491329193,
      "action_query_norm": 44.643828582763675,
      "action_relation_loss": 0.010994256287813187,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.237605238540692,
      "flow_matching_loss": 0.06207930222153664,
      "grad_norm": 0.18794040381908417,
      "h_stock_valid_norm": 97.3303955078125,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.723872206765116e-05,
      "loss": 0.0731,
      "positive_teacher_pairs": 26843.8,
      "query_h_norm_ratio": 0.4586914837360382,
      "relation_loss": 0.010994256287813187,
      "selected_pairs": 26843.8,
      "step": 4530,
      "student_distance_mean": 31.89982452392578,
      "teacher_distance_mean": 0.8624886691570282,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.010994256287813187,
      "weighted_relation_loss": 0.010994256287813187
    },
    {
      "action_query_batch_std": 0.4759587287902832,
      "action_query_norm": 44.63103713989258,
      "action_relation_loss": 0.010897161439061165,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.2469597754911135,
      "flow_matching_loss": 0.061255042999982835,
      "grad_norm": 0.17383834719657898,
      "h_stock_valid_norm": 96.59940948486329,
      "invalid_tail_samples": 25.0,
      "learning_rate": 9.722063079524185e-05,
      "loss": 0.0722,
      "positive_teacher_pairs": 26571.7,
      "query_h_norm_ratio": 0.46202893257141114,
      "relation_loss": 0.010897161439061165,
      "selected_pairs": 26571.7,
      "step": 4540,
      "student_distance_mean": 31.046372985839845,
      "teacher_distance_mean": 0.854794156551361,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.010897161439061165,
      "weighted_relation_loss": 0.010897161439061165
    },
    {
      "action_query_batch_std": 0.48387300968170166,
      "action_query_norm": 44.64029884338379,
      "action_relation_loss": 0.01241676053032279,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.256314312441534,
      "flow_matching_loss": 0.06366722919046879,
      "grad_norm": 0.1564827561378479,
      "h_stock_valid_norm": 97.34317626953126,
      "invalid_tail_samples": 22.1,
      "learning_rate": 9.720248214531351e-05,
      "loss": 0.0761,
      "positive_teacher_pairs": 27244.9,
      "query_h_norm_ratio": 0.45859664082527163,
      "relation_loss": 0.01241676053032279,
      "selected_pairs": 27244.9,
      "step": 4550,
      "student_distance_mean": 31.680897903442382,
      "teacher_distance_mean": 0.8614430844783783,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.01241676053032279,
      "weighted_relation_loss": 0.01241676053032279
    },
    {
      "action_query_batch_std": 0.4853141099214554,
      "action_query_norm": 44.636871337890625,
      "action_relation_loss": 0.011568349599838258,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.265668849391955,
      "flow_matching_loss": 0.060392207279801366,
      "grad_norm": 0.1880442202091217,
      "h_stock_valid_norm": 98.66981887817383,
      "invalid_tail_samples": 25.5,
      "learning_rate": 9.718427613991848e-05,
      "loss": 0.072,
      "positive_teacher_pairs": 26466.5,
      "query_h_norm_ratio": 0.45239087343215945,
      "relation_loss": 0.011568349599838258,
      "selected_pairs": 26466.5,
      "step": 4560,
      "student_distance_mean": 31.691842079162598,
      "teacher_distance_mean": 0.8586371839046478,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.011568349599838258,
      "weighted_relation_loss": 0.011568349599838258
    },
    {
      "action_query_batch_std": 0.4789921432733536,
      "action_query_norm": 44.6303409576416,
      "action_relation_loss": 0.01113472692668438,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.275023386342376,
      "flow_matching_loss": 0.06215410307049751,
      "grad_norm": 0.1575939655303955,
      "h_stock_valid_norm": 99.17158660888671,
      "invalid_tail_samples": 25.8,
      "learning_rate": 9.716601280117873e-05,
      "loss": 0.0733,
      "positive_teacher_pairs": 26393.0,
      "query_h_norm_ratio": 0.4500330686569214,
      "relation_loss": 0.01113472692668438,
      "selected_pairs": 26393.0,
      "step": 4570,
      "student_distance_mean": 31.233748626708984,
      "teacher_distance_mean": 0.858333170413971,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.01113472692668438,
      "weighted_relation_loss": 0.01113472692668438
    },
    {
      "action_query_batch_std": 0.4720147579908371,
      "action_query_norm": 44.623372650146486,
      "action_relation_loss": 0.011372232530266046,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.284377923292797,
      "flow_matching_loss": 0.06087645255029202,
      "grad_norm": 0.14358626306056976,
      "h_stock_valid_norm": 98.70921020507812,
      "invalid_tail_samples": 21.9,
      "learning_rate": 9.714769215128596e-05,
      "loss": 0.0722,
      "positive_teacher_pairs": 27292.4,
      "query_h_norm_ratio": 0.45207626819610597,
      "relation_loss": 0.011372232530266046,
      "selected_pairs": 27292.4,
      "step": 4580,
      "student_distance_mean": 30.893610191345214,
      "teacher_distance_mean": 0.8537873804569245,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.011372232530266046,
      "weighted_relation_loss": 0.011372232530266046
    },
    {
      "action_query_batch_std": 0.4774122923612595,
      "action_query_norm": 44.619384384155275,
      "action_relation_loss": 0.011609192565083503,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.293732460243218,
      "flow_matching_loss": 0.06207299679517746,
      "grad_norm": 0.15702596306800842,
      "h_stock_valid_norm": 97.41052169799805,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.712931421250152e-05,
      "loss": 0.0737,
      "positive_teacher_pairs": 26826.6,
      "query_h_norm_ratio": 0.45806140303611753,
      "relation_loss": 0.011609192565083503,
      "selected_pairs": 26826.6,
      "step": 4590,
      "student_distance_mean": 31.173062324523926,
      "teacher_distance_mean": 0.8575679302215576,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.011609192565083503,
      "weighted_relation_loss": 0.011609192565083503
    },
    {
      "action_query_batch_std": 0.4809464693069458,
      "action_query_norm": 44.622750473022464,
      "action_relation_loss": 0.01132254432886839,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.303086997193639,
      "flow_matching_loss": 0.06152411177754402,
      "grad_norm": 0.14098532497882843,
      "h_stock_valid_norm": 96.93267135620117,
      "invalid_tail_samples": 23.0,
      "learning_rate": 9.711087900715627e-05,
      "loss": 0.0728,
      "positive_teacher_pairs": 27035.3,
      "query_h_norm_ratio": 0.46035189032554624,
      "relation_loss": 0.01132254432886839,
      "selected_pairs": 27035.3,
      "step": 4600,
      "student_distance_mean": 31.515067481994627,
      "teacher_distance_mean": 0.8551149368286133,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.01132254432886839,
      "weighted_relation_loss": 0.01132254432886839
    },
    {
      "action_query_batch_std": 0.48771590292453765,
      "action_query_norm": 44.631241989135745,
      "action_relation_loss": 0.011634179204702378,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.31244153414406,
      "flow_matching_loss": 0.061808598786592485,
      "grad_norm": 0.16439922153949738,
      "h_stock_valid_norm": 97.14056854248047,
      "invalid_tail_samples": 22.6,
      "learning_rate": 9.709238655765078e-05,
      "loss": 0.0734,
      "positive_teacher_pairs": 27133.0,
      "query_h_norm_ratio": 0.45945313572883606,
      "relation_loss": 0.011634179204702378,
      "selected_pairs": 27133.0,
      "step": 4610,
      "student_distance_mean": 31.628091430664064,
      "teacher_distance_mean": 0.8517165899276733,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.011634179204702378,
      "weighted_relation_loss": 0.011634179204702378
    },
    {
      "action_query_batch_std": 0.47286940515041354,
      "action_query_norm": 44.616310119628906,
      "action_relation_loss": 0.011336973961442709,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.3217960710944805,
      "flow_matching_loss": 0.05857089795172214,
      "grad_norm": 0.20241835713386536,
      "h_stock_valid_norm": 96.9460075378418,
      "invalid_tail_samples": 21.1,
      "learning_rate": 9.707383688645511e-05,
      "loss": 0.0699,
      "positive_teacher_pairs": 27479.4,
      "query_h_norm_ratio": 0.46022018492221833,
      "relation_loss": 0.011336973961442709,
      "selected_pairs": 27479.4,
      "step": 4620,
      "student_distance_mean": 30.79219913482666,
      "teacher_distance_mean": 0.8474745929241181,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.011336973961442709,
      "weighted_relation_loss": 0.011336973961442709
    },
    {
      "action_query_batch_std": 0.4888583213090897,
      "action_query_norm": 44.62730369567871,
      "action_relation_loss": 0.011107947770506143,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.331150608044902,
      "flow_matching_loss": 0.060914424061775205,
      "grad_norm": 0.1555032879114151,
      "h_stock_valid_norm": 97.59872207641601,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.705523001610883e-05,
      "loss": 0.072,
      "positive_teacher_pairs": 27172.1,
      "query_h_norm_ratio": 0.4572596251964569,
      "relation_loss": 0.011107947770506143,
      "selected_pairs": 27172.1,
      "step": 4630,
      "student_distance_mean": 31.735394096374513,
      "teacher_distance_mean": 0.8648318529129029,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.011107947770506143,
      "weighted_relation_loss": 0.011107947770506143
    },
    {
      "action_query_batch_std": 0.4833998531103134,
      "action_query_norm": 44.61777191162109,
      "action_relation_loss": 0.011443063803017139,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.340505144995323,
      "flow_matching_loss": 0.061836658790707585,
      "grad_norm": 0.17931468784809113,
      "h_stock_valid_norm": 98.5478904724121,
      "invalid_tail_samples": 25.8,
      "learning_rate": 9.703656596922107e-05,
      "loss": 0.0733,
      "positive_teacher_pairs": 26385.8,
      "query_h_norm_ratio": 0.45275644958019257,
      "relation_loss": 0.011443063803017139,
      "selected_pairs": 26385.8,
      "step": 4640,
      "student_distance_mean": 31.51028480529785,
      "teacher_distance_mean": 0.846158766746521,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.011443063803017139,
      "weighted_relation_loss": 0.011443063803017139
    },
    {
      "action_query_batch_std": 0.4836196959018707,
      "action_query_norm": 44.61889533996582,
      "action_relation_loss": 0.011531687062233686,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.349859681945744,
      "flow_matching_loss": 0.06246611624956131,
      "grad_norm": 0.17619940638542175,
      "h_stock_valid_norm": 98.21700134277344,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.70178447684704e-05,
      "loss": 0.074,
      "positive_teacher_pairs": 26946.1,
      "query_h_norm_ratio": 0.4542907863855362,
      "relation_loss": 0.011531687062233686,
      "selected_pairs": 26946.1,
      "step": 4650,
      "student_distance_mean": 31.624494743347167,
      "teacher_distance_mean": 0.8579635083675384,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.011531687062233686,
      "weighted_relation_loss": 0.011531687062233686
    },
    {
      "action_query_batch_std": 0.4850062996149063,
      "action_query_norm": 44.61730270385742,
      "action_relation_loss": 0.011505767144262791,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.359214218896165,
      "flow_matching_loss": 0.06081499233841896,
      "grad_norm": 0.13855530321598053,
      "h_stock_valid_norm": 98.37376022338867,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.699906643660483e-05,
      "loss": 0.0723,
      "positive_teacher_pairs": 26876.2,
      "query_h_norm_ratio": 0.45355442464351653,
      "relation_loss": 0.011505767144262791,
      "selected_pairs": 26876.2,
      "step": 4660,
      "student_distance_mean": 31.809961318969727,
      "teacher_distance_mean": 0.842037433385849,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.011505767144262791,
      "weighted_relation_loss": 0.011505767144262791
    },
    {
      "action_query_batch_std": 0.472590371966362,
      "action_query_norm": 44.60501403808594,
      "action_relation_loss": 0.011090197693556547,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.368568755846586,
      "flow_matching_loss": 0.05772481523454189,
      "grad_norm": 0.15913350880146027,
      "h_stock_valid_norm": 99.07179718017578,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.698023099644185e-05,
      "loss": 0.0688,
      "positive_teacher_pairs": 27174.4,
      "query_h_norm_ratio": 0.4502326428890228,
      "relation_loss": 0.011090197693556547,
      "selected_pairs": 27174.4,
      "step": 4670,
      "student_distance_mean": 31.0340669631958,
      "teacher_distance_mean": 0.8416971802711487,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.011090197693556547,
      "weighted_relation_loss": 0.011090197693556547
    },
    {
      "action_query_batch_std": 0.47626790404319763,
      "action_query_norm": 44.607723236083984,
      "action_relation_loss": 0.010870990995317697,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.377923292797006,
      "flow_matching_loss": 0.06090589538216591,
      "grad_norm": 0.24240738153457642,
      "h_stock_valid_norm": 97.86138153076172,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.696133847086823e-05,
      "loss": 0.0718,
      "positive_teacher_pairs": 26758.8,
      "query_h_norm_ratio": 0.4558517158031464,
      "relation_loss": 0.010870990995317697,
      "selected_pairs": 26758.8,
      "step": 4680,
      "student_distance_mean": 31.231603813171386,
      "teacher_distance_mean": 0.8511555433273316,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.010870990995317697,
      "weighted_relation_loss": 0.010870990995317697
    },
    {
      "action_query_batch_std": 0.47204284369945526,
      "action_query_norm": 44.604292678833005,
      "action_relation_loss": 0.011253321636468172,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.387277829747427,
      "flow_matching_loss": 0.062100795656442644,
      "grad_norm": 0.1298438459634781,
      "h_stock_valid_norm": 97.15622329711914,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.694238888284022e-05,
      "loss": 0.0734,
      "positive_teacher_pairs": 27203.0,
      "query_h_norm_ratio": 0.45910407304763795,
      "relation_loss": 0.011253321636468172,
      "selected_pairs": 27203.0,
      "step": 4690,
      "student_distance_mean": 30.77375144958496,
      "teacher_distance_mean": 0.8515803694725037,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.011253321636468172,
      "weighted_relation_loss": 0.011253321636468172
    },
    {
      "action_query_batch_std": 0.4788667321205139,
      "action_query_norm": 44.611009216308595,
      "action_relation_loss": 0.01092166369780898,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.396632366697848,
      "flow_matching_loss": 0.058808876946568486,
      "grad_norm": 0.11534646153450012,
      "h_stock_valid_norm": 98.25363540649414,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.692338225538333e-05,
      "loss": 0.0697,
      "positive_teacher_pairs": 26989.5,
      "query_h_norm_ratio": 0.45404491722583773,
      "relation_loss": 0.01092166369780898,
      "selected_pairs": 26989.5,
      "step": 4700,
      "student_distance_mean": 31.231554222106933,
      "teacher_distance_mean": 0.859032267332077,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.01092166369780898,
      "weighted_relation_loss": 0.01092166369780898
    },
    {
      "action_query_batch_std": 0.4737900346517563,
      "action_query_norm": 44.59915008544922,
      "action_relation_loss": 0.011571981851011515,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.4059869036482695,
      "flow_matching_loss": 0.05972997359931469,
      "grad_norm": 0.21930193901062012,
      "h_stock_valid_norm": 97.71176605224609,
      "invalid_tail_samples": 25.8,
      "learning_rate": 9.690431861159241e-05,
      "loss": 0.0713,
      "positive_teacher_pairs": 26391.4,
      "query_h_norm_ratio": 0.4564462095499039,
      "relation_loss": 0.011571981851011515,
      "selected_pairs": 26391.4,
      "step": 4710,
      "student_distance_mean": 30.91868362426758,
      "teacher_distance_mean": 0.8513824045658112,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.011571981851011515,
      "weighted_relation_loss": 0.011571981851011515
    },
    {
      "action_query_batch_std": 0.4830204606056213,
      "action_query_norm": 44.607756805419925,
      "action_relation_loss": 0.011456313822418452,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.415341440598691,
      "flow_matching_loss": 0.06176236718893051,
      "grad_norm": 0.1615096628665924,
      "h_stock_valid_norm": 97.41975173950195,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.688519797463161e-05,
      "loss": 0.0732,
      "positive_teacher_pairs": 26615.1,
      "query_h_norm_ratio": 0.45789695084095,
      "relation_loss": 0.011456313822418452,
      "selected_pairs": 26615.1,
      "step": 4720,
      "student_distance_mean": 31.410317993164064,
      "teacher_distance_mean": 0.8550209999084473,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.011456313822418452,
      "weighted_relation_loss": 0.011456313822418452
    },
    {
      "action_query_batch_std": 0.47762841284275054,
      "action_query_norm": 44.6041316986084,
      "action_relation_loss": 0.011337349005043507,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.424695977549112,
      "flow_matching_loss": 0.05662374719977379,
      "grad_norm": 0.15343689918518066,
      "h_stock_valid_norm": 97.49034729003907,
      "invalid_tail_samples": 25.1,
      "learning_rate": 9.686602036773426e-05,
      "loss": 0.068,
      "positive_teacher_pairs": 26551.4,
      "query_h_norm_ratio": 0.45752844512462615,
      "relation_loss": 0.011337349005043507,
      "selected_pairs": 26551.4,
      "step": 4730,
      "student_distance_mean": 31.38118705749512,
      "teacher_distance_mean": 0.8547833263874054,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.011337349005043507,
      "weighted_relation_loss": 0.011337349005043507
    },
    {
      "action_query_batch_std": 0.47172988653182985,
      "action_query_norm": 44.59167976379395,
      "action_relation_loss": 0.010897765588015317,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.434050514499532,
      "flow_matching_loss": 0.059386293217539785,
      "grad_norm": 0.17499171197414398,
      "h_stock_valid_norm": 98.05049667358398,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.684678581420302e-05,
      "loss": 0.0703,
      "positive_teacher_pairs": 26685.8,
      "query_h_norm_ratio": 0.45478755235671997,
      "relation_loss": 0.010897765588015317,
      "selected_pairs": 26685.8,
      "step": 4740,
      "student_distance_mean": 30.892574501037597,
      "teacher_distance_mean": 0.8494265258312226,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.010897765588015317,
      "weighted_relation_loss": 0.010897765588015317
    },
    {
      "action_query_batch_std": 0.4790138274431229,
      "action_query_norm": 44.60157241821289,
      "action_relation_loss": 0.011213570553809404,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.443405051449953,
      "flow_matching_loss": 0.059408992528915405,
      "grad_norm": 0.18973404169082642,
      "h_stock_valid_norm": 98.31545562744141,
      "invalid_tail_samples": 25.4,
      "learning_rate": 9.682749433740962e-05,
      "loss": 0.0706,
      "positive_teacher_pairs": 26479.2,
      "query_h_norm_ratio": 0.4536680281162262,
      "relation_loss": 0.011213570553809404,
      "selected_pairs": 26479.2,
      "step": 4750,
      "student_distance_mean": 31.345954132080077,
      "teacher_distance_mean": 0.8615129232406616,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.011213570553809404,
      "weighted_relation_loss": 0.011213570553809404
    },
    {
      "action_query_batch_std": 0.47546874582767484,
      "action_query_norm": 44.597883987426755,
      "action_relation_loss": 0.011607900261878967,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.452759588400374,
      "flow_matching_loss": 0.06174094527959824,
      "grad_norm": 0.14014892280101776,
      "h_stock_valid_norm": 97.24396667480468,
      "invalid_tail_samples": 26.1,
      "learning_rate": 9.680814596079507e-05,
      "loss": 0.0733,
      "positive_teacher_pairs": 26325.6,
      "query_h_norm_ratio": 0.4586283802986145,
      "relation_loss": 0.011607900261878967,
      "selected_pairs": 26325.6,
      "step": 4760,
      "student_distance_mean": 31.159216690063477,
      "teacher_distance_mean": 0.8525031805038452,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.011607900261878967,
      "weighted_relation_loss": 0.011607900261878967
    },
    {
      "action_query_batch_std": 0.48545712828636167,
      "action_query_norm": 44.606280136108396,
      "action_relation_loss": 0.010745825339108705,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.462114125350795,
      "flow_matching_loss": 0.06272085905075073,
      "grad_norm": 0.12498600035905838,
      "h_stock_valid_norm": 97.02926940917969,
      "invalid_tail_samples": 25.2,
      "learning_rate": 9.678874070786945e-05,
      "loss": 0.0735,
      "positive_teacher_pairs": 26525.3,
      "query_h_norm_ratio": 0.45972239077091215,
      "relation_loss": 0.010745825339108705,
      "selected_pairs": 26525.3,
      "step": 4770,
      "student_distance_mean": 31.541488265991212,
      "teacher_distance_mean": 0.8562284111976624,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.010745825339108705,
      "weighted_relation_loss": 0.010745825339108705
    },
    {
      "action_query_batch_std": 0.48604093194007875,
      "action_query_norm": 44.6054931640625,
      "action_relation_loss": 0.011157059669494629,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.471468662301216,
      "flow_matching_loss": 0.0625683069229126,
      "grad_norm": 0.21537937223911285,
      "h_stock_valid_norm": 97.57547225952149,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.676927860221199e-05,
      "loss": 0.0737,
      "positive_teacher_pairs": 26920.2,
      "query_h_norm_ratio": 0.45714166164398196,
      "relation_loss": 0.011157059669494629,
      "selected_pairs": 26920.2,
      "step": 4780,
      "student_distance_mean": 31.87130641937256,
      "teacher_distance_mean": 0.8626834034919739,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.011157059669494629,
      "weighted_relation_loss": 0.011157059669494629
    },
    {
      "action_query_batch_std": 0.47917007505893705,
      "action_query_norm": 44.59697380065918,
      "action_relation_loss": 0.010909816436469554,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.480823199251637,
      "flow_matching_loss": 0.06227344758808613,
      "grad_norm": 0.23051388561725616,
      "h_stock_valid_norm": 98.6093978881836,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.674975966747097e-05,
      "loss": 0.0732,
      "positive_teacher_pairs": 26915.8,
      "query_h_norm_ratio": 0.452274951338768,
      "relation_loss": 0.010909816436469554,
      "selected_pairs": 26915.8,
      "step": 4790,
      "student_distance_mean": 31.357781600952148,
      "teacher_distance_mean": 0.8509296298027038,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.010909816436469554,
      "weighted_relation_loss": 0.010909816436469554
    },
    {
      "action_query_batch_std": 0.4785579741001129,
      "action_query_norm": 44.59437103271485,
      "action_relation_loss": 0.010820436570793391,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.490177736202058,
      "flow_matching_loss": 0.06207271218299866,
      "grad_norm": 0.16860006749629974,
      "h_stock_valid_norm": 97.98242263793945,
      "invalid_tail_samples": 25.6,
      "learning_rate": 9.673018392736374e-05,
      "loss": 0.0729,
      "positive_teacher_pairs": 26436.9,
      "query_h_norm_ratio": 0.45513266921043394,
      "relation_loss": 0.010820436570793391,
      "selected_pairs": 26436.9,
      "step": 4800,
      "student_distance_mean": 31.44944820404053,
      "teacher_distance_mean": 0.8617861151695252,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.010820436570793391,
      "weighted_relation_loss": 0.010820436570793391
    },
    {
      "action_query_batch_std": 0.4734389305114746,
      "action_query_norm": 44.581753158569335,
      "action_relation_loss": 0.010976470913738013,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.499532273152479,
      "flow_matching_loss": 0.06252138763666153,
      "grad_norm": 0.19652220606803894,
      "h_stock_valid_norm": 97.42757949829101,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.671055140567667e-05,
      "loss": 0.0735,
      "positive_teacher_pairs": 27087.2,
      "query_h_norm_ratio": 0.45759263038635256,
      "relation_loss": 0.010976470913738013,
      "selected_pairs": 27087.2,
      "step": 4810,
      "student_distance_mean": 31.036231422424315,
      "teacher_distance_mean": 0.8558677971363068,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.010976470913738013,
      "weighted_relation_loss": 0.010976470913738013
    },
    {
      "action_query_batch_std": 0.473494416475296,
      "action_query_norm": 44.57926292419434,
      "action_relation_loss": 0.011285113915801049,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.5088868101029,
      "flow_matching_loss": 0.059890955686569214,
      "grad_norm": 0.13768263161182404,
      "h_stock_valid_norm": 97.10989837646484,
      "invalid_tail_samples": 21.7,
      "learning_rate": 9.669086212626511e-05,
      "loss": 0.0712,
      "positive_teacher_pairs": 27337.9,
      "query_h_norm_ratio": 0.4590637177228928,
      "relation_loss": 0.011285113915801049,
      "selected_pairs": 27337.9,
      "step": 4820,
      "student_distance_mean": 30.8895357131958,
      "teacher_distance_mean": 0.8533408939838409,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.011285113915801049,
      "weighted_relation_loss": 0.011285113915801049
    },
    {
      "action_query_batch_std": 0.4667089760303497,
      "action_query_norm": 44.568471145629886,
      "action_relation_loss": 0.011150000616908073,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.518241347053321,
      "flow_matching_loss": 0.06328401453793049,
      "grad_norm": 0.17451463639736176,
      "h_stock_valid_norm": 96.58174591064453,
      "invalid_tail_samples": 23.1,
      "learning_rate": 9.667111611305341e-05,
      "loss": 0.0744,
      "positive_teacher_pairs": 27014.6,
      "query_h_norm_ratio": 0.46146859526634215,
      "relation_loss": 0.011150000616908073,
      "selected_pairs": 27014.6,
      "step": 4830,
      "student_distance_mean": 30.583532333374023,
      "teacher_distance_mean": 0.8485257387161255,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.011150000616908073,
      "weighted_relation_loss": 0.011150000616908073
    },
    {
      "action_query_batch_std": 0.477264815568924,
      "action_query_norm": 44.58040428161621,
      "action_relation_loss": 0.010620412975549698,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.527595884003742,
      "flow_matching_loss": 0.06302038691937924,
      "grad_norm": 0.17436060309410095,
      "h_stock_valid_norm": 96.37134170532227,
      "invalid_tail_samples": 21.8,
      "learning_rate": 9.665131339003486e-05,
      "loss": 0.0736,
      "positive_teacher_pairs": 27315.2,
      "query_h_norm_ratio": 0.4625918120145798,
      "relation_loss": 0.010620412975549698,
      "selected_pairs": 27315.2,
      "step": 4840,
      "student_distance_mean": 31.186006927490233,
      "teacher_distance_mean": 0.8546050429344177,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.010620412975549698,
      "weighted_relation_loss": 0.010620412975549698
    },
    {
      "action_query_batch_std": 0.47579711079597475,
      "action_query_norm": 44.581181335449216,
      "action_relation_loss": 0.01120602861046791,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.536950420954163,
      "flow_matching_loss": 0.06020042151212692,
      "grad_norm": 0.20322643220424652,
      "h_stock_valid_norm": 96.9911376953125,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.663145398127158e-05,
      "loss": 0.0714,
      "positive_teacher_pairs": 26943.8,
      "query_h_norm_ratio": 0.4596570491790771,
      "relation_loss": 0.01120602861046791,
      "selected_pairs": 26943.8,
      "step": 4850,
      "student_distance_mean": 31.169797706604005,
      "teacher_distance_mean": 0.8466946005821228,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.01120602861046791,
      "weighted_relation_loss": 0.01120602861046791
    },
    {
      "action_query_batch_std": 0.47179285883903505,
      "action_query_norm": 44.5773365020752,
      "action_relation_loss": 0.010997678339481353,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.546304957904583,
      "flow_matching_loss": 0.059587153047323226,
      "grad_norm": 0.13359685242176056,
      "h_stock_valid_norm": 99.0075180053711,
      "invalid_tail_samples": 25.9,
      "learning_rate": 9.661153791089467e-05,
      "loss": 0.0706,
      "positive_teacher_pairs": 26363.4,
      "query_h_norm_ratio": 0.4502559930086136,
      "relation_loss": 0.010997678339481353,
      "selected_pairs": 26363.4,
      "step": 4860,
      "student_distance_mean": 30.899089050292968,
      "teacher_distance_mean": 0.8519186794757843,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.010997678339481353,
      "weighted_relation_loss": 0.010997678339481353
    },
    {
      "action_query_batch_std": 0.4780888020992279,
      "action_query_norm": 44.581966400146484,
      "action_relation_loss": 0.01091439351439476,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.555659494855004,
      "flow_matching_loss": 0.058934358134865764,
      "grad_norm": 0.13457541167736053,
      "h_stock_valid_norm": 99.60351028442383,
      "invalid_tail_samples": 26.7,
      "learning_rate": 9.659156520310402e-05,
      "loss": 0.0698,
      "positive_teacher_pairs": 26178.7,
      "query_h_norm_ratio": 0.44759646356105803,
      "relation_loss": 0.01091439351439476,
      "selected_pairs": 26178.7,
      "step": 4870,
      "student_distance_mean": 31.407452392578126,
      "teacher_distance_mean": 0.855468612909317,
      "valid_action_samples": 229.3,
      "weighted_auxiliary_loss": 0.01091439351439476,
      "weighted_relation_loss": 0.01091439351439476
    },
    {
      "action_query_batch_std": 0.48306448459625245,
      "action_query_norm": 44.58366394042969,
      "action_relation_loss": 0.01092251567170024,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.5650140318054255,
      "flow_matching_loss": 0.06076401583850384,
      "grad_norm": 0.16106362640857697,
      "h_stock_valid_norm": 99.50317077636718,
      "invalid_tail_samples": 25.6,
      "learning_rate": 9.657153588216835e-05,
      "loss": 0.0717,
      "positive_teacher_pairs": 26435.9,
      "query_h_norm_ratio": 0.44806512296199796,
      "relation_loss": 0.01092251567170024,
      "selected_pairs": 26435.9,
      "step": 4880,
      "student_distance_mean": 31.67693977355957,
      "teacher_distance_mean": 0.851275771856308,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.01092251567170024,
      "weighted_relation_loss": 0.01092251567170024
    },
    {
      "action_query_batch_std": 0.4675379544496536,
      "action_query_norm": 44.567996978759766,
      "action_relation_loss": 0.010932593699544668,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.574368568755847,
      "flow_matching_loss": 0.06152478605508804,
      "grad_norm": 0.164012610912323,
      "h_stock_valid_norm": 99.333154296875,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.655144997242516e-05,
      "loss": 0.0725,
      "positive_teacher_pairs": 27193.1,
      "query_h_norm_ratio": 0.4486780405044556,
      "relation_loss": 0.010932593699544668,
      "selected_pairs": 27193.1,
      "step": 4890,
      "student_distance_mean": 30.56433219909668,
      "teacher_distance_mean": 0.8548653542995452,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.010932593699544668,
      "weighted_relation_loss": 0.010932593699544668
    },
    {
      "action_query_batch_std": 0.4669442862272263,
      "action_query_norm": 44.56623764038086,
      "action_relation_loss": 0.010680625960230828,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.583723105706268,
      "flow_matching_loss": 0.05885823704302311,
      "grad_norm": 0.13858655095100403,
      "h_stock_valid_norm": 98.76271209716796,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.653130749828075e-05,
      "loss": 0.0695,
      "positive_teacher_pairs": 27151.5,
      "query_h_norm_ratio": 0.4512540251016617,
      "relation_loss": 0.010680625960230828,
      "selected_pairs": 27151.5,
      "step": 4900,
      "student_distance_mean": 30.532863426208497,
      "teacher_distance_mean": 0.8450495004653931,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.010680625960230828,
      "weighted_relation_loss": 0.010680625960230828
    },
    {
      "action_query_batch_std": 0.4716722756624222,
      "action_query_norm": 44.56760482788086,
      "action_relation_loss": 0.011405212245881558,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.593077642656689,
      "flow_matching_loss": 0.060553425922989845,
      "grad_norm": 0.14502468705177307,
      "h_stock_valid_norm": 97.85074920654297,
      "invalid_tail_samples": 25.3,
      "learning_rate": 9.65111084842101e-05,
      "loss": 0.072,
      "positive_teacher_pairs": 26505.1,
      "query_h_norm_ratio": 0.4554706603288651,
      "relation_loss": 0.011405212245881558,
      "selected_pairs": 26505.1,
      "step": 4910,
      "student_distance_mean": 30.769984817504884,
      "teacher_distance_mean": 0.8430170536041259,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.011405212245881558,
      "weighted_relation_loss": 0.011405212245881558
    },
    {
      "action_query_batch_std": 0.47341051399707795,
      "action_query_norm": 44.570486068725586,
      "action_relation_loss": 0.010581251420080661,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.60243217960711,
      "flow_matching_loss": 0.060425407439470294,
      "grad_norm": 0.2024688422679901,
      "h_stock_valid_norm": 97.05233764648438,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.649085295475695e-05,
      "loss": 0.071,
      "positive_teacher_pairs": 26832.7,
      "query_h_norm_ratio": 0.45924695730209353,
      "relation_loss": 0.010581251420080661,
      "selected_pairs": 26832.7,
      "step": 4920,
      "student_distance_mean": 30.929676055908203,
      "teacher_distance_mean": 0.8564192950725555,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.010581251420080661,
      "weighted_relation_loss": 0.010581251420080661
    },
    {
      "action_query_batch_std": 0.48204918801784513,
      "action_query_norm": 44.579124450683594,
      "action_relation_loss": 0.011003592610359192,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.61178671655753,
      "flow_matching_loss": 0.05673377923667431,
      "grad_norm": 0.17540369927883148,
      "h_stock_valid_norm": 96.69012603759765,
      "invalid_tail_samples": 23.0,
      "learning_rate": 9.647054093453365e-05,
      "loss": 0.0677,
      "positive_teacher_pairs": 27038.6,
      "query_h_norm_ratio": 0.461056587100029,
      "relation_loss": 0.011003592610359192,
      "selected_pairs": 27038.6,
      "step": 4930,
      "student_distance_mean": 31.69102535247803,
      "teacher_distance_mean": 0.8650088012218475,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.011003592610359192,
      "weighted_relation_loss": 0.011003592610359192
    },
    {
      "action_query_batch_std": 0.4706333577632904,
      "action_query_norm": 44.56258430480957,
      "action_relation_loss": 0.011111374851316214,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.621141253507951,
      "flow_matching_loss": 0.0603554654866457,
      "grad_norm": 0.12269844859838486,
      "h_stock_valid_norm": 97.73267669677735,
      "invalid_tail_samples": 23.0,
      "learning_rate": 9.645017244822123e-05,
      "loss": 0.0715,
      "positive_teacher_pairs": 27036.7,
      "query_h_norm_ratio": 0.45597007870674133,
      "relation_loss": 0.011111374851316214,
      "selected_pairs": 27036.7,
      "step": 4940,
      "student_distance_mean": 30.74917678833008,
      "teacher_distance_mean": 0.8473906695842743,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.011111374851316214,
      "weighted_relation_loss": 0.011111374851316214
    },
    {
      "action_query_batch_std": 0.4794232338666916,
      "action_query_norm": 44.57789497375488,
      "action_relation_loss": 0.01098253708332777,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.630495790458372,
      "flow_matching_loss": 0.05968321114778519,
      "grad_norm": 0.1616075485944748,
      "h_stock_valid_norm": 98.94126892089844,
      "invalid_tail_samples": 24.4,
      "learning_rate": 9.642974752056931e-05,
      "loss": 0.0707,
      "positive_teacher_pairs": 26715.9,
      "query_h_norm_ratio": 0.45055278241634367,
      "relation_loss": 0.01098253708332777,
      "selected_pairs": 26715.9,
      "step": 4950,
      "student_distance_mean": 31.157347869873046,
      "teacher_distance_mean": 0.8540888667106629,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.01098253708332777,
      "weighted_relation_loss": 0.01098253708332777
    },
    {
      "action_query_batch_std": 0.47334825098514555,
      "action_query_norm": 44.564932632446286,
      "action_relation_loss": 0.011318635195493698,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.639850327408793,
      "flow_matching_loss": 0.06151516996324062,
      "grad_norm": 0.16433756053447723,
      "h_stock_valid_norm": 98.92442398071289,
      "invalid_tail_samples": 26.3,
      "learning_rate": 9.640926617639613e-05,
      "loss": 0.0728,
      "positive_teacher_pairs": 26277.7,
      "query_h_norm_ratio": 0.450498902797699,
      "relation_loss": 0.011318635195493698,
      "selected_pairs": 26277.7,
      "step": 4960,
      "student_distance_mean": 30.9157470703125,
      "teacher_distance_mean": 0.8560550689697266,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.011318635195493698,
      "weighted_relation_loss": 0.011318635195493698
    },
    {
      "action_query_batch_std": 0.47006484866142273,
      "action_query_norm": 44.55738830566406,
      "action_relation_loss": 0.010539456643164158,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.6492048643592145,
      "flow_matching_loss": 0.06044447757303715,
      "grad_norm": 0.18317262828350067,
      "h_stock_valid_norm": 98.04531936645508,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.638872844058843e-05,
      "loss": 0.071,
      "positive_teacher_pairs": 26803.0,
      "query_h_norm_ratio": 0.45446870625019076,
      "relation_loss": 0.010539456643164158,
      "selected_pairs": 26803.0,
      "step": 4970,
      "student_distance_mean": 30.768224906921386,
      "teacher_distance_mean": 0.850550377368927,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.010539456643164158,
      "weighted_relation_loss": 0.010539456643164158
    },
    {
      "action_query_batch_std": 0.4706401973962784,
      "action_query_norm": 44.558993911743165,
      "action_relation_loss": 0.010516241379082202,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.658559401309635,
      "flow_matching_loss": 0.05969969630241394,
      "grad_norm": 0.15148480236530304,
      "h_stock_valid_norm": 97.92623291015624,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.63681343381015e-05,
      "loss": 0.0702,
      "positive_teacher_pairs": 26901.1,
      "query_h_norm_ratio": 0.4550293505191803,
      "relation_loss": 0.010516241379082202,
      "selected_pairs": 26901.1,
      "step": 4980,
      "student_distance_mean": 30.76738872528076,
      "teacher_distance_mean": 0.8551260471343994,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.010516241379082202,
      "weighted_relation_loss": 0.010516241379082202
    },
    {
      "action_query_batch_std": 0.47348553836345675,
      "action_query_norm": 44.561174392700195,
      "action_relation_loss": 0.010439570248126983,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.667913938260056,
      "flow_matching_loss": 0.06257754489779473,
      "grad_norm": 0.22943086922168732,
      "h_stock_valid_norm": 97.20879516601562,
      "invalid_tail_samples": 25.3,
      "learning_rate": 9.634748389395914e-05,
      "loss": 0.073,
      "positive_teacher_pairs": 26502.4,
      "query_h_norm_ratio": 0.4584125131368637,
      "relation_loss": 0.010439570248126983,
      "selected_pairs": 26502.4,
      "step": 4990,
      "student_distance_mean": 31.047990798950195,
      "teacher_distance_mean": 0.857477217912674,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.010439570248126983,
      "weighted_relation_loss": 0.010439570248126983
    },
    {
      "action_query_batch_std": 0.4742398202419281,
      "action_query_norm": 44.56442070007324,
      "action_relation_loss": 0.01062879040837288,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.677268475210477,
      "flow_matching_loss": 0.05997963957488537,
      "grad_norm": 0.1883011758327484,
      "h_stock_valid_norm": 98.10619506835937,
      "invalid_tail_samples": 23.0,
      "learning_rate": 9.632677713325353e-05,
      "loss": 0.0706,
      "positive_teacher_pairs": 27039.4,
      "query_h_norm_ratio": 0.45425851345062257,
      "relation_loss": 0.01062879040837288,
      "selected_pairs": 27039.4,
      "step": 5000,
      "student_distance_mean": 31.0267333984375,
      "teacher_distance_mean": 0.8576675653457642,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.01062879040837288,
      "weighted_relation_loss": 0.01062879040837288
    },
    {
      "action_query_batch_std": 0.4775197505950928,
      "action_query_norm": 44.565985107421874,
      "action_relation_loss": 0.010852222610265017,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.686623012160898,
      "flow_matching_loss": 0.058492694050073624,
      "grad_norm": 0.23020419478416443,
      "h_stock_valid_norm": 98.01765289306641,
      "invalid_tail_samples": 20.9,
      "learning_rate": 9.63060140811454e-05,
      "loss": 0.0693,
      "positive_teacher_pairs": 27536.3,
      "query_h_norm_ratio": 0.454680797457695,
      "relation_loss": 0.010852222610265017,
      "selected_pairs": 27536.3,
      "step": 5010,
      "student_distance_mean": 31.21657524108887,
      "teacher_distance_mean": 0.8532325923442841,
      "valid_action_samples": 235.1,
      "weighted_auxiliary_loss": 0.010852222610265017,
      "weighted_relation_loss": 0.010852222610265017
    },
    {
      "action_query_batch_std": 0.47528857588768003,
      "action_query_norm": 44.56001739501953,
      "action_relation_loss": 0.01072999257594347,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.695977549111319,
      "flow_matching_loss": 0.05870809406042099,
      "grad_norm": 0.20936009287834167,
      "h_stock_valid_norm": 97.09817733764649,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.628519476286379e-05,
      "loss": 0.0694,
      "positive_teacher_pairs": 26875.0,
      "query_h_norm_ratio": 0.45892215371131895,
      "relation_loss": 0.01072999257594347,
      "selected_pairs": 26875.0,
      "step": 5020,
      "student_distance_mean": 31.063265991210937,
      "teacher_distance_mean": 0.850562858581543,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.01072999257594347,
      "weighted_relation_loss": 0.01072999257594347
    },
    {
      "action_query_batch_std": 0.47343148589134215,
      "action_query_norm": 44.55351600646973,
      "action_relation_loss": 0.010524470079690217,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.70533208606174,
      "flow_matching_loss": 0.05844341702759266,
      "grad_norm": 0.15647576749324799,
      "h_stock_valid_norm": 97.55539321899414,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.626431920370612e-05,
      "loss": 0.069,
      "positive_teacher_pairs": 27154.3,
      "query_h_norm_ratio": 0.45670463144779205,
      "relation_loss": 0.010524470079690217,
      "selected_pairs": 27154.3,
      "step": 5030,
      "student_distance_mean": 31.018959617614748,
      "teacher_distance_mean": 0.855182284116745,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.010524470079690217,
      "weighted_relation_loss": 0.010524470079690217
    },
    {
      "action_query_batch_std": 0.4740124523639679,
      "action_query_norm": 44.55236663818359,
      "action_relation_loss": 0.010681039467453957,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.714686623012161,
      "flow_matching_loss": 0.060596184059977534,
      "grad_norm": 0.21190790832042694,
      "h_stock_valid_norm": 97.01664886474609,
      "invalid_tail_samples": 22.1,
      "learning_rate": 9.624338742903819e-05,
      "loss": 0.0713,
      "positive_teacher_pairs": 27257.5,
      "query_h_norm_ratio": 0.45924893021583557,
      "relation_loss": 0.010681039467453957,
      "selected_pairs": 27257.5,
      "step": 5040,
      "student_distance_mean": 31.026520347595216,
      "teacher_distance_mean": 0.8637486219406127,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.010681039467453957,
      "weighted_relation_loss": 0.010681039467453957
    },
    {
      "action_query_batch_std": 0.4673721641302109,
      "action_query_norm": 44.54244689941406,
      "action_relation_loss": 0.010744010377675296,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.724041159962582,
      "flow_matching_loss": 0.060037712007761,
      "grad_norm": 0.19431184232234955,
      "h_stock_valid_norm": 95.67474899291992,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.622239946429406e-05,
      "loss": 0.0708,
      "positive_teacher_pairs": 26686.5,
      "query_h_norm_ratio": 0.4655632257461548,
      "relation_loss": 0.010744010377675296,
      "selected_pairs": 26686.5,
      "step": 5050,
      "student_distance_mean": 30.594982528686522,
      "teacher_distance_mean": 0.849944430589676,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.010744010377675296,
      "weighted_relation_loss": 0.010744010377675296
    },
    {
      "action_query_batch_std": 0.4690026789903641,
      "action_query_norm": 44.54506149291992,
      "action_relation_loss": 0.011632287502288818,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.733395696913003,
      "flow_matching_loss": 0.061757927387952806,
      "grad_norm": 0.24666526913642883,
      "h_stock_valid_norm": 96.20722274780273,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.620135533497609e-05,
      "loss": 0.0734,
      "positive_teacher_pairs": 26639.2,
      "query_h_norm_ratio": 0.46301400661468506,
      "relation_loss": 0.011632287502288818,
      "selected_pairs": 26639.2,
      "step": 5060,
      "student_distance_mean": 30.53177261352539,
      "teacher_distance_mean": 0.8419934511184692,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.011632287502288818,
      "weighted_relation_loss": 0.011632287502288818
    },
    {
      "action_query_batch_std": 0.46315441727638246,
      "action_query_norm": 44.53357849121094,
      "action_relation_loss": 0.011171984858810902,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.742750233863424,
      "flow_matching_loss": 0.05745612755417824,
      "grad_norm": 0.15902678668498993,
      "h_stock_valid_norm": 96.93706588745117,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.61802550666549e-05,
      "loss": 0.0686,
      "positive_teacher_pairs": 26802.0,
      "query_h_norm_ratio": 0.4594123899936676,
      "relation_loss": 0.011171984858810902,
      "selected_pairs": 26802.0,
      "step": 5070,
      "student_distance_mean": 30.493192863464355,
      "teacher_distance_mean": 0.8519473433494568,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.011171984858810902,
      "weighted_relation_loss": 0.011171984858810902
    },
    {
      "action_query_batch_std": 0.47375534772872924,
      "action_query_norm": 44.54473457336426,
      "action_relation_loss": 0.01051085814833641,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.752104770813845,
      "flow_matching_loss": 0.05852865166962147,
      "grad_norm": 0.20195744931697845,
      "h_stock_valid_norm": 97.41410446166992,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.615909868496928e-05,
      "loss": 0.069,
      "positive_teacher_pairs": 27221.0,
      "query_h_norm_ratio": 0.4572752982378006,
      "relation_loss": 0.01051085814833641,
      "selected_pairs": 27221.0,
      "step": 5080,
      "student_distance_mean": 30.934306144714355,
      "teacher_distance_mean": 0.8542192816734314,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.01051085814833641,
      "weighted_relation_loss": 0.01051085814833641
    },
    {
      "action_query_batch_std": 0.4683733105659485,
      "action_query_norm": 44.53494110107422,
      "action_relation_loss": 0.010814825911074877,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.761459307764266,
      "flow_matching_loss": 0.059667378291487695,
      "grad_norm": 0.1550617516040802,
      "h_stock_valid_norm": 98.30290222167969,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.613788621562622e-05,
      "loss": 0.0705,
      "positive_teacher_pairs": 26737.6,
      "query_h_norm_ratio": 0.45305168330669404,
      "relation_loss": 0.010814825911074877,
      "selected_pairs": 26737.6,
      "step": 5090,
      "student_distance_mean": 30.63443088531494,
      "teacher_distance_mean": 0.8538103401660919,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.010814825911074877,
      "weighted_relation_loss": 0.010814825911074877
    },
    {
      "action_query_batch_std": 0.46091423332691195,
      "action_query_norm": 44.52429618835449,
      "action_relation_loss": 0.010964601952582597,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.770813844714686,
      "flow_matching_loss": 0.06179225854575634,
      "grad_norm": 0.16290627419948578,
      "h_stock_valid_norm": 99.40829315185547,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.611661768440093e-05,
      "loss": 0.0728,
      "positive_teacher_pairs": 26997.2,
      "query_h_norm_ratio": 0.44789781868457795,
      "relation_loss": 0.010964601952582597,
      "selected_pairs": 26997.2,
      "step": 5100,
      "student_distance_mean": 30.249217987060547,
      "teacher_distance_mean": 0.842427134513855,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.010964601952582597,
      "weighted_relation_loss": 0.010964601952582597
    },
    {
      "action_query_batch_std": 0.46705633103847505,
      "action_query_norm": 44.535217666625975,
      "action_relation_loss": 0.01076023355126381,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.780168381665107,
      "flow_matching_loss": 0.057597062364220616,
      "grad_norm": 0.17867691814899445,
      "h_stock_valid_norm": 99.00219345092773,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.609529311713661e-05,
      "loss": 0.0684,
      "positive_teacher_pairs": 27201.1,
      "query_h_norm_ratio": 0.4498463302850723,
      "relation_loss": 0.01076023355126381,
      "selected_pairs": 27201.1,
      "step": 5110,
      "student_distance_mean": 30.650707054138184,
      "teacher_distance_mean": 0.8490824282169342,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.01076023355126381,
      "weighted_relation_loss": 0.01076023355126381
    },
    {
      "action_query_batch_std": 0.46507246494293214,
      "action_query_norm": 44.527614212036134,
      "action_relation_loss": 0.010697720386087895,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.789522918615528,
      "flow_matching_loss": 0.05934034064412117,
      "grad_norm": 0.15436503291130066,
      "h_stock_valid_norm": 98.42008972167969,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.607391253974466e-05,
      "loss": 0.07,
      "positive_teacher_pairs": 27084.9,
      "query_h_norm_ratio": 0.45243552029132844,
      "relation_loss": 0.010697720386087895,
      "selected_pairs": 27084.9,
      "step": 5120,
      "student_distance_mean": 30.494623374938964,
      "teacher_distance_mean": 0.8553180038928986,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.010697720386087895,
      "weighted_relation_loss": 0.010697720386087895
    },
    {
      "action_query_batch_std": 0.4739867776632309,
      "action_query_norm": 44.536935806274414,
      "action_relation_loss": 0.01055995626375079,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.7988774555659495,
      "flow_matching_loss": 0.05802274867892265,
      "grad_norm": 0.14702925086021423,
      "h_stock_valid_norm": 97.55060958862305,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.605247597820448e-05,
      "loss": 0.0686,
      "positive_teacher_pairs": 26688.3,
      "query_h_norm_ratio": 0.45656523704528806,
      "relation_loss": 0.01055995626375079,
      "selected_pairs": 26688.3,
      "step": 5130,
      "student_distance_mean": 31.20287952423096,
      "teacher_distance_mean": 0.8544264376163483,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.01055995626375079,
      "weighted_relation_loss": 0.01055995626375079
    },
    {
      "action_query_batch_std": 0.469257453083992,
      "action_query_norm": 44.53114700317383,
      "action_relation_loss": 0.010762082412838936,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.808231992516371,
      "flow_matching_loss": 0.059296340122818945,
      "grad_norm": 0.13697196543216705,
      "h_stock_valid_norm": 96.24519195556641,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.603098345856354e-05,
      "loss": 0.0701,
      "positive_teacher_pairs": 26894.6,
      "query_h_norm_ratio": 0.46268978118896487,
      "relation_loss": 0.010762082412838936,
      "selected_pairs": 26894.6,
      "step": 5140,
      "student_distance_mean": 30.745322036743165,
      "teacher_distance_mean": 0.849615889787674,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.010762082412838936,
      "weighted_relation_loss": 0.010762082412838936
    },
    {
      "action_query_batch_std": 0.47177145183086394,
      "action_query_norm": 44.54129600524902,
      "action_relation_loss": 0.010296674538403749,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.817586529466792,
      "flow_matching_loss": 0.05805052220821381,
      "grad_norm": 0.1679845154285431,
      "h_stock_valid_norm": 96.69689178466797,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.600943500693725e-05,
      "loss": 0.0683,
      "positive_teacher_pairs": 26767.9,
      "query_h_norm_ratio": 0.46063134670257566,
      "relation_loss": 0.010296674538403749,
      "selected_pairs": 26767.9,
      "step": 5150,
      "student_distance_mean": 31.088643074035645,
      "teacher_distance_mean": 0.8628664433956146,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.010296674538403749,
      "weighted_relation_loss": 0.010296674538403749
    },
    {
      "action_query_batch_std": 0.4777324467897415,
      "action_query_norm": 44.542981719970705,
      "action_relation_loss": 0.010374738648533822,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.826941066417213,
      "flow_matching_loss": 0.058956443890929223,
      "grad_norm": 0.17407619953155518,
      "h_stock_valid_norm": 98.08067398071289,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.598783064950902e-05,
      "loss": 0.0693,
      "positive_teacher_pairs": 26736.7,
      "query_h_norm_ratio": 0.4541607141494751,
      "relation_loss": 0.010374738648533822,
      "selected_pairs": 26736.7,
      "step": 5160,
      "student_distance_mean": 31.379065132141115,
      "teacher_distance_mean": 0.866655296087265,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.010374738648533822,
      "weighted_relation_loss": 0.010374738648533822
    },
    {
      "action_query_batch_std": 0.46468949019908906,
      "action_query_norm": 44.52590065002441,
      "action_relation_loss": 0.010427276510745287,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.836295603367633,
      "flow_matching_loss": 0.057298829406499864,
      "grad_norm": 0.1650007665157318,
      "h_stock_valid_norm": 97.96951751708984,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.596617041253018e-05,
      "loss": 0.0677,
      "positive_teacher_pairs": 26747.3,
      "query_h_norm_ratio": 0.45449121594429015,
      "relation_loss": 0.010427276510745287,
      "selected_pairs": 26747.3,
      "step": 5170,
      "student_distance_mean": 30.493934631347656,
      "teacher_distance_mean": 0.8484009385108948,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.010427276510745287,
      "weighted_relation_loss": 0.010427276510745287
    },
    {
      "action_query_batch_std": 0.4687958717346191,
      "action_query_norm": 44.52809600830078,
      "action_relation_loss": 0.010655425861477852,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.845650140318054,
      "flow_matching_loss": 0.05816437415778637,
      "grad_norm": 0.13978244364261627,
      "h_stock_valid_norm": 97.35988922119141,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.594445432231996e-05,
      "loss": 0.0688,
      "positive_teacher_pairs": 26881.2,
      "query_h_norm_ratio": 0.4573604494333267,
      "relation_loss": 0.010655425861477852,
      "selected_pairs": 26881.2,
      "step": 5180,
      "student_distance_mean": 30.702563667297362,
      "teacher_distance_mean": 0.8520939767360687,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.010655425861477852,
      "weighted_relation_loss": 0.010655425861477852
    },
    {
      "action_query_batch_std": 0.46883051097393036,
      "action_query_norm": 44.524614334106445,
      "action_relation_loss": 0.010816155653446912,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.855004677268475,
      "flow_matching_loss": 0.059589442983269694,
      "grad_norm": 0.1765194535255432,
      "h_stock_valid_norm": 98.00332489013672,
      "invalid_tail_samples": 21.0,
      "learning_rate": 9.592268240526547e-05,
      "loss": 0.0704,
      "positive_teacher_pairs": 27504.2,
      "query_h_norm_ratio": 0.45433332920074465,
      "relation_loss": 0.010816155653446912,
      "selected_pairs": 27504.2,
      "step": 5190,
      "student_distance_mean": 30.669753646850587,
      "teacher_distance_mean": 0.8591581821441651,
      "valid_action_samples": 235.0,
      "weighted_auxiliary_loss": 0.010816155653446912,
      "weighted_relation_loss": 0.010816155653446912
    },
    {
      "action_query_batch_std": 0.47344416975975034,
      "action_query_norm": 44.533667373657224,
      "action_relation_loss": 0.010670022387057543,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.864359214218896,
      "flow_matching_loss": 0.058266342431306836,
      "grad_norm": 0.13722383975982666,
      "h_stock_valid_norm": 98.36186752319335,
      "invalid_tail_samples": 26.2,
      "learning_rate": 9.590085468782162e-05,
      "loss": 0.0689,
      "positive_teacher_pairs": 26300.9,
      "query_h_norm_ratio": 0.452759838104248,
      "relation_loss": 0.010670022387057543,
      "selected_pairs": 26300.9,
      "step": 5200,
      "student_distance_mean": 31.188782501220704,
      "teacher_distance_mean": 0.8556196630001068,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.010670022387057543,
      "weighted_relation_loss": 0.010670022387057543
    },
    {
      "action_query_batch_std": 0.4778501927852631,
      "action_query_norm": 44.54062347412109,
      "action_relation_loss": 0.010777425300329924,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.873713751169317,
      "flow_matching_loss": 0.058803100138902664,
      "grad_norm": 0.12316308915615082,
      "h_stock_valid_norm": 97.10140838623047,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.587897119651116e-05,
      "loss": 0.0696,
      "positive_teacher_pairs": 27092.1,
      "query_h_norm_ratio": 0.45871747434139254,
      "relation_loss": 0.010777425300329924,
      "selected_pairs": 27092.1,
      "step": 5210,
      "student_distance_mean": 31.418117141723634,
      "teacher_distance_mean": 0.8626244604587555,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.010777425300329924,
      "weighted_relation_loss": 0.010777425300329924
    },
    {
      "action_query_batch_std": 0.47014871835708616,
      "action_query_norm": 44.53233642578125,
      "action_relation_loss": 0.010819801967591048,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.883068288119738,
      "flow_matching_loss": 0.05806494392454624,
      "grad_norm": 0.14445506036281586,
      "h_stock_valid_norm": 96.18160858154297,
      "invalid_tail_samples": 25.0,
      "learning_rate": 9.585703195792459e-05,
      "loss": 0.0689,
      "positive_teacher_pairs": 26572.1,
      "query_h_norm_ratio": 0.4630118548870087,
      "relation_loss": 0.010819801967591048,
      "selected_pairs": 26572.1,
      "step": 5220,
      "student_distance_mean": 30.85983810424805,
      "teacher_distance_mean": 0.8452275097370148,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.010819801967591048,
      "weighted_relation_loss": 0.010819801967591048
    },
    {
      "action_query_batch_std": 0.468090632557869,
      "action_query_norm": 44.52007713317871,
      "action_relation_loss": 0.01047374764457345,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.892422825070159,
      "flow_matching_loss": 0.05793922357261181,
      "grad_norm": 0.17665323615074158,
      "h_stock_valid_norm": 96.51542587280274,
      "invalid_tail_samples": 25.5,
      "learning_rate": 9.583503699872016e-05,
      "loss": 0.0684,
      "positive_teacher_pairs": 26461.9,
      "query_h_norm_ratio": 0.4612860858440399,
      "relation_loss": 0.01047374764457345,
      "selected_pairs": 26461.9,
      "step": 5230,
      "student_distance_mean": 30.76550750732422,
      "teacher_distance_mean": 0.8630922675132752,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.01047374764457345,
      "weighted_relation_loss": 0.01047374764457345
    },
    {
      "action_query_batch_std": 0.4663187265396118,
      "action_query_norm": 44.51611328125,
      "action_relation_loss": 0.010687540750950575,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.90177736202058,
      "flow_matching_loss": 0.0594515573233366,
      "grad_norm": 0.18697144091129303,
      "h_stock_valid_norm": 97.09613571166992,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.581298634562381e-05,
      "loss": 0.0701,
      "positive_teacher_pairs": 26825.0,
      "query_h_norm_ratio": 0.4584773153066635,
      "relation_loss": 0.010687540750950575,
      "selected_pairs": 26825.0,
      "step": 5240,
      "student_distance_mean": 30.69414710998535,
      "teacher_distance_mean": 0.8586979329586029,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.010687540750950575,
      "weighted_relation_loss": 0.010687540750950575
    },
    {
      "action_query_batch_std": 0.462951123714447,
      "action_query_norm": 44.512169647216794,
      "action_relation_loss": 0.010569591261446475,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.911131898971001,
      "flow_matching_loss": 0.05867917835712433,
      "grad_norm": 0.15067555010318756,
      "h_stock_valid_norm": 97.10260238647462,
      "invalid_tail_samples": 22.1,
      "learning_rate": 9.579088002542917e-05,
      "loss": 0.0692,
      "positive_teacher_pairs": 27242.4,
      "query_h_norm_ratio": 0.45840536057949066,
      "relation_loss": 0.010569591261446475,
      "selected_pairs": 27242.4,
      "step": 5250,
      "student_distance_mean": 30.469448852539063,
      "teacher_distance_mean": 0.8503828704357147,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.010569591261446475,
      "weighted_relation_loss": 0.010569591261446475
    },
    {
      "action_query_batch_std": 0.46916369795799256,
      "action_query_norm": 44.51854782104492,
      "action_relation_loss": 0.01064548445865512,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.920486435921422,
      "flow_matching_loss": 0.06018410734832287,
      "grad_norm": 0.12317422032356262,
      "h_stock_valid_norm": 97.67572250366212,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.57687180649975e-05,
      "loss": 0.0708,
      "positive_teacher_pairs": 26881.6,
      "query_h_norm_ratio": 0.4557800978422165,
      "relation_loss": 0.01064548445865512,
      "selected_pairs": 26881.6,
      "step": 5260,
      "student_distance_mean": 30.791304206848146,
      "teacher_distance_mean": 0.8583162486553192,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.01064548445865512,
      "weighted_relation_loss": 0.01064548445865512
    },
    {
      "action_query_batch_std": 0.47094165086746215,
      "action_query_norm": 44.51808319091797,
      "action_relation_loss": 0.010875755455344915,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.929840972871843,
      "flow_matching_loss": 0.05639662407338619,
      "grad_norm": 0.1637793481349945,
      "h_stock_valid_norm": 97.30905456542969,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.574650049125768e-05,
      "loss": 0.0673,
      "positive_teacher_pairs": 26687.0,
      "query_h_norm_ratio": 0.4574990004301071,
      "relation_loss": 0.010875755455344915,
      "selected_pairs": 26687.0,
      "step": 5270,
      "student_distance_mean": 30.843579483032226,
      "teacher_distance_mean": 0.8575884401798248,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.010875755455344915,
      "weighted_relation_loss": 0.010875755455344915
    },
    {
      "action_query_batch_std": 0.468611142039299,
      "action_query_norm": 44.51769485473633,
      "action_relation_loss": 0.010585291590541602,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.939195509822264,
      "flow_matching_loss": 0.0577006209641695,
      "grad_norm": 0.15810175240039825,
      "h_stock_valid_norm": 96.1660400390625,
      "invalid_tail_samples": 21.7,
      "learning_rate": 9.572422733120614e-05,
      "loss": 0.0683,
      "positive_teacher_pairs": 27336.7,
      "query_h_norm_ratio": 0.46294593513011933,
      "relation_loss": 0.010585291590541602,
      "selected_pairs": 27336.7,
      "step": 5280,
      "student_distance_mean": 30.78942928314209,
      "teacher_distance_mean": 0.8600590169429779,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.010585291590541602,
      "weighted_relation_loss": 0.010585291590541602
    },
    {
      "action_query_batch_std": 0.4707574248313904,
      "action_query_norm": 44.512294387817384,
      "action_relation_loss": 0.010733773838728667,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.948550046772684,
      "flow_matching_loss": 0.06194098591804505,
      "grad_norm": 0.14786523580551147,
      "h_stock_valid_norm": 96.99586715698243,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.570189861190689e-05,
      "loss": 0.0727,
      "positive_teacher_pairs": 26801.0,
      "query_h_norm_ratio": 0.4589333862066269,
      "relation_loss": 0.010733773838728667,
      "selected_pairs": 26801.0,
      "step": 5290,
      "student_distance_mean": 30.845726585388185,
      "teacher_distance_mean": 0.8572879016399384,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.010733773838728667,
      "weighted_relation_loss": 0.010733773838728667
    },
    {
      "action_query_batch_std": 0.46948912739753723,
      "action_query_norm": 44.51028594970703,
      "action_relation_loss": 0.010231702402234078,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.9579045837231055,
      "flow_matching_loss": 0.056610897183418274,
      "grad_norm": 0.10976507514715195,
      "h_stock_valid_norm": 99.25062866210938,
      "invalid_tail_samples": 25.8,
      "learning_rate": 9.56795143604914e-05,
      "loss": 0.0668,
      "positive_teacher_pairs": 26387.4,
      "query_h_norm_ratio": 0.4484895884990692,
      "relation_loss": 0.010231702402234078,
      "selected_pairs": 26387.4,
      "step": 5300,
      "student_distance_mean": 30.857221603393555,
      "teacher_distance_mean": 0.8592241823673248,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.010231702402234078,
      "weighted_relation_loss": 0.010231702402234078
    },
    {
      "action_query_batch_std": 0.46559666395187377,
      "action_query_norm": 44.507342147827146,
      "action_relation_loss": 0.010597234964370728,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.967259120673527,
      "flow_matching_loss": 0.05862019434571266,
      "grad_norm": 0.13214638829231262,
      "h_stock_valid_norm": 100.4261489868164,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.565707460415869e-05,
      "loss": 0.0692,
      "positive_teacher_pairs": 26897.7,
      "query_h_norm_ratio": 0.4431956589221954,
      "relation_loss": 0.010597234964370728,
      "selected_pairs": 26897.7,
      "step": 5310,
      "student_distance_mean": 30.421637535095215,
      "teacher_distance_mean": 0.8477039992809295,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.010597234964370728,
      "weighted_relation_loss": 0.010597234964370728
    },
    {
      "action_query_batch_std": 0.4609119057655334,
      "action_query_norm": 44.49583358764649,
      "action_relation_loss": 0.01023478815332055,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.976613657623948,
      "flow_matching_loss": 0.05627688057720661,
      "grad_norm": 0.15345819294452667,
      "h_stock_valid_norm": 98.71137390136718,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.563457937017515e-05,
      "loss": 0.0665,
      "positive_teacher_pairs": 26850.6,
      "query_h_norm_ratio": 0.4507847219705582,
      "relation_loss": 0.01023478815332055,
      "selected_pairs": 26850.6,
      "step": 5320,
      "student_distance_mean": 30.23900909423828,
      "teacher_distance_mean": 0.8510671377182006,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.01023478815332055,
      "weighted_relation_loss": 0.01023478815332055
    },
    {
      "action_query_batch_std": 0.46247434616088867,
      "action_query_norm": 44.499371719360354,
      "action_relation_loss": 0.010416782833635807,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.985968194574369,
      "flow_matching_loss": 0.05793861486017704,
      "grad_norm": 0.15064896643161774,
      "h_stock_valid_norm": 98.15128936767579,
      "invalid_tail_samples": 20.6,
      "learning_rate": 9.56120286858746e-05,
      "loss": 0.0684,
      "positive_teacher_pairs": 27596.7,
      "query_h_norm_ratio": 0.45338033735752103,
      "relation_loss": 0.010416782833635807,
      "selected_pairs": 27596.7,
      "step": 5330,
      "student_distance_mean": 30.291440391540526,
      "teacher_distance_mean": 0.863540095090866,
      "valid_action_samples": 235.4,
      "weighted_auxiliary_loss": 0.010416782833635807,
      "weighted_relation_loss": 0.010416782833635807
    },
    {
      "action_query_batch_std": 0.4603968232870102,
      "action_query_norm": 44.49518737792969,
      "action_relation_loss": 0.010812332574278116,
      "auxiliary_loss_weight": 1.0,
      "epoch": 4.99532273152479,
      "flow_matching_loss": 0.06053634658455849,
      "grad_norm": 0.16499066352844238,
      "h_stock_valid_norm": 97.9950798034668,
      "invalid_tail_samples": 22.9,
      "learning_rate": 9.558942257865829e-05,
      "loss": 0.0713,
      "positive_teacher_pairs": 27059.2,
      "query_h_norm_ratio": 0.4540602207183838,
      "relation_loss": 0.010812332574278116,
      "selected_pairs": 27059.2,
      "step": 5340,
      "student_distance_mean": 30.29606685638428,
      "teacher_distance_mean": 0.8540811717510224,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.010812332574278116,
      "weighted_relation_loss": 0.010812332574278116
    },
    {
      "action_query_batch_std": 0.4624186486005783,
      "action_query_norm": 44.49411582946777,
      "action_relation_loss": 0.010178569890558719,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.00467726847521,
      "flow_matching_loss": 0.05705821253359318,
      "grad_norm": 0.1787833422422409,
      "h_stock_valid_norm": 97.31760635375977,
      "invalid_tail_samples": 26.0,
      "learning_rate": 9.556676107599472e-05,
      "loss": 0.0672,
      "positive_teacher_pairs": 23362.0,
      "query_h_norm_ratio": 0.45720998346805575,
      "relation_loss": 0.010178569890558719,
      "selected_pairs": 23362.0,
      "step": 5350,
      "student_distance_mean": 30.506744956970216,
      "teacher_distance_mean": 0.8513019919395447,
      "valid_action_samples": 210.1,
      "weighted_auxiliary_loss": 0.010178569890558719,
      "weighted_relation_loss": 0.010178569890558719
    },
    {
      "action_query_batch_std": 0.466109561920166,
      "action_query_norm": 44.49307403564453,
      "action_relation_loss": 0.010010905843228102,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.014031805425631,
      "flow_matching_loss": 0.05881690829992294,
      "grad_norm": 0.1797187626361847,
      "h_stock_valid_norm": 98.02097625732422,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.554404420541978e-05,
      "loss": 0.0688,
      "positive_teacher_pairs": 26740.2,
      "query_h_norm_ratio": 0.4539165377616882,
      "relation_loss": 0.010010905843228102,
      "selected_pairs": 26740.2,
      "step": 5360,
      "student_distance_mean": 30.55073184967041,
      "teacher_distance_mean": 0.8578934133052826,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.010010905843228102,
      "weighted_relation_loss": 0.010010905843228102
    },
    {
      "action_query_batch_std": 0.4655102550983429,
      "action_query_norm": 44.49757766723633,
      "action_relation_loss": 0.01018579863011837,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.023386342376052,
      "flow_matching_loss": 0.056698080897331235,
      "grad_norm": 0.18831880390644073,
      "h_stock_valid_norm": 98.3624641418457,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.55212719945366e-05,
      "loss": 0.0669,
      "positive_teacher_pairs": 26661.4,
      "query_h_norm_ratio": 0.45238859951496124,
      "relation_loss": 0.01018579863011837,
      "selected_pairs": 26661.4,
      "step": 5370,
      "student_distance_mean": 30.61063289642334,
      "teacher_distance_mean": 0.856038236618042,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.01018579863011837,
      "weighted_relation_loss": 0.01018579863011837
    },
    {
      "action_query_batch_std": 0.46971548199653623,
      "action_query_norm": 44.499623489379886,
      "action_relation_loss": 0.010296604223549366,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.032740879326473,
      "flow_matching_loss": 0.05732854530215263,
      "grad_norm": 0.2268390655517578,
      "h_stock_valid_norm": 98.10992126464843,
      "invalid_tail_samples": 22.7,
      "learning_rate": 9.549844447101559e-05,
      "loss": 0.0676,
      "positive_teacher_pairs": 27112.7,
      "query_h_norm_ratio": 0.4535733938217163,
      "relation_loss": 0.010296604223549366,
      "selected_pairs": 27112.7,
      "step": 5380,
      "student_distance_mean": 30.95980739593506,
      "teacher_distance_mean": 0.8519276976585388,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.010296604223549366,
      "weighted_relation_loss": 0.010296604223549366
    },
    {
      "action_query_batch_std": 0.45987104177474974,
      "action_query_norm": 44.487037658691406,
      "action_relation_loss": 0.010015705600380898,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.0420954162768945,
      "flow_matching_loss": 0.05804821774363518,
      "grad_norm": 0.1721368432044983,
      "h_stock_valid_norm": 97.48740768432617,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.547556166259433e-05,
      "loss": 0.0681,
      "positive_teacher_pairs": 27224.7,
      "query_h_norm_ratio": 0.4563397467136383,
      "relation_loss": 0.010015705600380898,
      "selected_pairs": 27224.7,
      "step": 5390,
      "student_distance_mean": 30.267913818359375,
      "teacher_distance_mean": 0.8577889621257782,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.010015705600380898,
      "weighted_relation_loss": 0.010015705600380898
    },
    {
      "action_query_batch_std": 0.46920148730278016,
      "action_query_norm": 44.49984931945801,
      "action_relation_loss": 0.01035102317109704,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.051449953227316,
      "flow_matching_loss": 0.05641200989484787,
      "grad_norm": 0.20308993756771088,
      "h_stock_valid_norm": 97.89458999633788,
      "invalid_tail_samples": 25.3,
      "learning_rate": 9.545262359707756e-05,
      "loss": 0.0668,
      "positive_teacher_pairs": 26507.0,
      "query_h_norm_ratio": 0.4545732825994492,
      "relation_loss": 0.01035102317109704,
      "selected_pairs": 26507.0,
      "step": 5400,
      "student_distance_mean": 30.843988990783693,
      "teacher_distance_mean": 0.8597228944301605,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.01035102317109704,
      "weighted_relation_loss": 0.01035102317109704
    },
    {
      "action_query_batch_std": 0.47202526926994326,
      "action_query_norm": 44.496514892578126,
      "action_relation_loss": 0.010051693860441446,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.060804490177736,
      "flow_matching_loss": 0.05617568828165531,
      "grad_norm": 0.1907641589641571,
      "h_stock_valid_norm": 97.72307891845703,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.542963030233724e-05,
      "loss": 0.0662,
      "positive_teacher_pairs": 26779.8,
      "query_h_norm_ratio": 0.45533818006515503,
      "relation_loss": 0.010051693860441446,
      "selected_pairs": 26779.8,
      "step": 5410,
      "student_distance_mean": 30.891250801086425,
      "teacher_distance_mean": 0.8540614604949951,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.010051693860441446,
      "weighted_relation_loss": 0.010051693860441446
    },
    {
      "action_query_batch_std": 0.4607544541358948,
      "action_query_norm": 44.48362693786621,
      "action_relation_loss": 0.010367305018007755,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.070159027128157,
      "flow_matching_loss": 0.05648953542113304,
      "grad_norm": 0.22620831429958344,
      "h_stock_valid_norm": 96.80540161132812,
      "invalid_tail_samples": 25.0,
      "learning_rate": 9.540658180631237e-05,
      "loss": 0.0669,
      "positive_teacher_pairs": 26569.3,
      "query_h_norm_ratio": 0.4595257639884949,
      "relation_loss": 0.010367305018007755,
      "selected_pairs": 26569.3,
      "step": 5420,
      "student_distance_mean": 30.37175941467285,
      "teacher_distance_mean": 0.8501707792282105,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.010367305018007755,
      "weighted_relation_loss": 0.010367305018007755
    },
    {
      "action_query_batch_std": 0.4586956262588501,
      "action_query_norm": 44.48326950073242,
      "action_relation_loss": 0.009494176879525185,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.079513564078578,
      "flow_matching_loss": 0.05611078664660454,
      "grad_norm": 0.2758028209209442,
      "h_stock_valid_norm": 96.38125915527344,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.538347813700904e-05,
      "loss": 0.0656,
      "positive_teacher_pairs": 26971.9,
      "query_h_norm_ratio": 0.4615385830402374,
      "relation_loss": 0.009494176879525185,
      "selected_pairs": 26971.9,
      "step": 5430,
      "student_distance_mean": 30.07367534637451,
      "teacher_distance_mean": 0.8568255782127381,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.009494176879525185,
      "weighted_relation_loss": 0.009494176879525185
    },
    {
      "action_query_batch_std": 0.45493516623973845,
      "action_query_norm": 44.47109184265137,
      "action_relation_loss": 0.010777524020522834,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.088868101028999,
      "flow_matching_loss": 0.057774904742836954,
      "grad_norm": 0.23359715938568115,
      "h_stock_valid_norm": 96.375,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.536031932250036e-05,
      "loss": 0.0686,
      "positive_teacher_pairs": 26619.8,
      "query_h_norm_ratio": 0.4614449828863144,
      "relation_loss": 0.010777524020522834,
      "selected_pairs": 26619.8,
      "step": 5440,
      "student_distance_mean": 29.999864768981933,
      "teacher_distance_mean": 0.8475438714027405,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.010777524020522834,
      "weighted_relation_loss": 0.010777524020522834
    },
    {
      "action_query_batch_std": 0.4725250333547592,
      "action_query_norm": 44.48736381530762,
      "action_relation_loss": 0.01025816798210144,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.09822263797942,
      "flow_matching_loss": 0.056328079849481585,
      "grad_norm": 0.2647758722305298,
      "h_stock_valid_norm": 96.7108772277832,
      "invalid_tail_samples": 26.6,
      "learning_rate": 9.533710539092653e-05,
      "loss": 0.0666,
      "positive_teacher_pairs": 26205.6,
      "query_h_norm_ratio": 0.4600121796131134,
      "relation_loss": 0.01025816798210144,
      "selected_pairs": 26205.6,
      "step": 5450,
      "student_distance_mean": 31.15473041534424,
      "teacher_distance_mean": 0.8568397521972656,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.01025816798210144,
      "weighted_relation_loss": 0.01025816798210144
    },
    {
      "action_query_batch_std": 0.45880578458309174,
      "action_query_norm": 44.47494239807129,
      "action_relation_loss": 0.010205160826444626,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.107577174929841,
      "flow_matching_loss": 0.056325193122029306,
      "grad_norm": 0.1516932100057602,
      "h_stock_valid_norm": 97.68159561157226,
      "invalid_tail_samples": 25.3,
      "learning_rate": 9.531383637049464e-05,
      "loss": 0.0665,
      "positive_teacher_pairs": 26509.4,
      "query_h_norm_ratio": 0.4553090661764145,
      "relation_loss": 0.010205160826444626,
      "selected_pairs": 26509.4,
      "step": 5460,
      "student_distance_mean": 30.19518165588379,
      "teacher_distance_mean": 0.857812124490738,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.010205160826444626,
      "weighted_relation_loss": 0.010205160826444626
    },
    {
      "action_query_batch_std": 0.4630785048007965,
      "action_query_norm": 44.47742614746094,
      "action_relation_loss": 0.009616729151457549,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.1169317118802615,
      "flow_matching_loss": 0.058741701394319536,
      "grad_norm": 0.15232516825199127,
      "h_stock_valid_norm": 97.73568344116211,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.529051228947875e-05,
      "loss": 0.0684,
      "positive_teacher_pairs": 26646.4,
      "query_h_norm_ratio": 0.45508241951465606,
      "relation_loss": 0.009616729151457549,
      "selected_pairs": 26646.4,
      "step": 5470,
      "student_distance_mean": 30.216324234008788,
      "teacher_distance_mean": 0.8524903059005737,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.009616729151457549,
      "weighted_relation_loss": 0.009616729151457549
    },
    {
      "action_query_batch_std": 0.4666549891233444,
      "action_query_norm": 44.48037757873535,
      "action_relation_loss": 0.009932750836014748,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.126286248830683,
      "flow_matching_loss": 0.05457545518875122,
      "grad_norm": 0.13202421367168427,
      "h_stock_valid_norm": 97.53422698974609,
      "invalid_tail_samples": 21.6,
      "learning_rate": 9.52671331762198e-05,
      "loss": 0.0645,
      "positive_teacher_pairs": 27361.6,
      "query_h_norm_ratio": 0.4560543805360794,
      "relation_loss": 0.009932750836014748,
      "selected_pairs": 27361.6,
      "step": 5480,
      "student_distance_mean": 30.593003273010254,
      "teacher_distance_mean": 0.8556136310100555,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.009932750836014748,
      "weighted_relation_loss": 0.009932750836014748
    },
    {
      "action_query_batch_std": 0.45772537887096404,
      "action_query_norm": 44.46580085754395,
      "action_relation_loss": 0.010217720828950404,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.135640785781104,
      "flow_matching_loss": 0.0572394959628582,
      "grad_norm": 0.15679073333740234,
      "h_stock_valid_norm": 97.92560272216797,
      "invalid_tail_samples": 26.1,
      "learning_rate": 9.524369905912565e-05,
      "loss": 0.0675,
      "positive_teacher_pairs": 26319.3,
      "query_h_norm_ratio": 0.45408074259757997,
      "relation_loss": 0.010217720828950404,
      "selected_pairs": 26319.3,
      "step": 5490,
      "student_distance_mean": 30.115767097473146,
      "teacher_distance_mean": 0.8472604572772979,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.010217720828950404,
      "weighted_relation_loss": 0.010217720828950404
    },
    {
      "action_query_batch_std": 0.4623848468065262,
      "action_query_norm": 44.47999114990235,
      "action_relation_loss": 0.01034839153289795,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.144995322731525,
      "flow_matching_loss": 0.05465189516544342,
      "grad_norm": 0.13499678671360016,
      "h_stock_valid_norm": 97.64279098510742,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.522020996667092e-05,
      "loss": 0.065,
      "positive_teacher_pairs": 26922.7,
      "query_h_norm_ratio": 0.45554250180721284,
      "relation_loss": 0.01034839153289795,
      "selected_pairs": 26922.7,
      "step": 5500,
      "student_distance_mean": 30.51542224884033,
      "teacher_distance_mean": 0.8516034066677094,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.01034839153289795,
      "weighted_relation_loss": 0.01034839153289795
    },
    {
      "action_query_batch_std": 0.46003003120422364,
      "action_query_norm": 44.468010330200194,
      "action_relation_loss": 0.009918551798909903,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.154349859681946,
      "flow_matching_loss": 0.05585098564624787,
      "grad_norm": 0.11677186191082001,
      "h_stock_valid_norm": 97.5019302368164,
      "invalid_tail_samples": 21.8,
      "learning_rate": 9.519666592739709e-05,
      "loss": 0.0658,
      "positive_teacher_pairs": 27312.2,
      "query_h_norm_ratio": 0.45607649981975557,
      "relation_loss": 0.009918551798909903,
      "selected_pairs": 27312.2,
      "step": 5510,
      "student_distance_mean": 30.17954921722412,
      "teacher_distance_mean": 0.8508839786052704,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.009918551798909903,
      "weighted_relation_loss": 0.009918551798909903
    },
    {
      "action_query_batch_std": 0.4637344002723694,
      "action_query_norm": 44.47708854675293,
      "action_relation_loss": 0.009801881946623325,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.163704396632367,
      "flow_matching_loss": 0.05556888170540333,
      "grad_norm": 0.15205782651901245,
      "h_stock_valid_norm": 98.63709869384766,
      "invalid_tail_samples": 25.7,
      "learning_rate": 9.517306696991241e-05,
      "loss": 0.0654,
      "positive_teacher_pairs": 26408.9,
      "query_h_norm_ratio": 0.45093099772930145,
      "relation_loss": 0.009801881946623325,
      "selected_pairs": 26408.9,
      "step": 5520,
      "student_distance_mean": 30.39806785583496,
      "teacher_distance_mean": 0.8579248428344727,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.009801881946623325,
      "weighted_relation_loss": 0.009801881946623325
    },
    {
      "action_query_batch_std": 0.4601523786783218,
      "action_query_norm": 44.47241287231445,
      "action_relation_loss": 0.00968975517898798,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.173058933582787,
      "flow_matching_loss": 0.05748551115393639,
      "grad_norm": 0.1863618940114975,
      "h_stock_valid_norm": 99.0117805480957,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.51494131228918e-05,
      "loss": 0.0672,
      "positive_teacher_pairs": 26663.8,
      "query_h_norm_ratio": 0.44916502833366395,
      "relation_loss": 0.00968975517898798,
      "selected_pairs": 26663.8,
      "step": 5530,
      "student_distance_mean": 30.348663902282716,
      "teacher_distance_mean": 0.855718195438385,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.00968975517898798,
      "weighted_relation_loss": 0.00968975517898798
    },
    {
      "action_query_batch_std": 0.45831796526908875,
      "action_query_norm": 44.464629364013675,
      "action_relation_loss": 0.009933676198124885,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.182413470533208,
      "flow_matching_loss": 0.05704486072063446,
      "grad_norm": 0.13702456653118134,
      "h_stock_valid_norm": 98.0314208984375,
      "invalid_tail_samples": 22.6,
      "learning_rate": 9.512570441507695e-05,
      "loss": 0.067,
      "positive_teacher_pairs": 27129.1,
      "query_h_norm_ratio": 0.4535798400640488,
      "relation_loss": 0.009933676198124885,
      "selected_pairs": 27129.1,
      "step": 5540,
      "student_distance_mean": 30.04604263305664,
      "teacher_distance_mean": 0.8497236490249633,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.009933676198124885,
      "weighted_relation_loss": 0.009933676198124885
    },
    {
      "action_query_batch_std": 0.46388944387435915,
      "action_query_norm": 44.474311447143556,
      "action_relation_loss": 0.010135497152805328,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.191768007483629,
      "flow_matching_loss": 0.057414236664772036,
      "grad_norm": 0.17846597731113434,
      "h_stock_valid_norm": 97.42723236083984,
      "invalid_tail_samples": 25.1,
      "learning_rate": 9.510194087527615e-05,
      "loss": 0.0675,
      "positive_teacher_pairs": 26548.2,
      "query_h_norm_ratio": 0.4564937949180603,
      "relation_loss": 0.010135497152805328,
      "selected_pairs": 26548.2,
      "step": 5550,
      "student_distance_mean": 30.592297554016113,
      "teacher_distance_mean": 0.8557537257671356,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.010135497152805328,
      "weighted_relation_loss": 0.010135497152805328
    },
    {
      "action_query_batch_std": 0.46662903428077696,
      "action_query_norm": 44.47632522583008,
      "action_relation_loss": 0.010112702008336782,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.2011225444340505,
      "flow_matching_loss": 0.057743947952985764,
      "grad_norm": 0.16036531329154968,
      "h_stock_valid_norm": 96.90269241333007,
      "invalid_tail_samples": 20.7,
      "learning_rate": 9.507812253236435e-05,
      "loss": 0.0679,
      "positive_teacher_pairs": 27571.6,
      "query_h_norm_ratio": 0.45898381173610686,
      "relation_loss": 0.010112702008336782,
      "selected_pairs": 27571.6,
      "step": 5560,
      "student_distance_mean": 30.565406227111815,
      "teacher_distance_mean": 0.8554112195968628,
      "valid_action_samples": 235.3,
      "weighted_auxiliary_loss": 0.010112702008336782,
      "weighted_relation_loss": 0.010112702008336782
    },
    {
      "action_query_batch_std": 0.4667451918125153,
      "action_query_norm": 44.467427825927736,
      "action_relation_loss": 0.010349617619067431,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.210477081384472,
      "flow_matching_loss": 0.05744264051318169,
      "grad_norm": 0.17077533900737762,
      "h_stock_valid_norm": 97.40260696411133,
      "invalid_tail_samples": 25.2,
      "learning_rate": 9.505424941528309e-05,
      "loss": 0.0678,
      "positive_teacher_pairs": 26526.3,
      "query_h_norm_ratio": 0.45653477907180784,
      "relation_loss": 0.010349617619067431,
      "selected_pairs": 26526.3,
      "step": 5570,
      "student_distance_mean": 30.610702896118163,
      "teacher_distance_mean": 0.8516004681587219,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.010349617619067431,
      "weighted_relation_loss": 0.010349617619067431
    },
    {
      "action_query_batch_std": 0.4559825509786606,
      "action_query_norm": 44.45131645202637,
      "action_relation_loss": 0.010663763340562581,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.219831618334893,
      "flow_matching_loss": 0.05658529959619045,
      "grad_norm": 0.17997197806835175,
      "h_stock_valid_norm": 97.4094108581543,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.503032155304046e-05,
      "loss": 0.0672,
      "positive_teacher_pairs": 26946.0,
      "query_h_norm_ratio": 0.4563403159379959,
      "relation_loss": 0.010663763340562581,
      "selected_pairs": 26946.0,
      "step": 5580,
      "student_distance_mean": 29.99890193939209,
      "teacher_distance_mean": 0.8465550601482391,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.010663763340562581,
      "weighted_relation_loss": 0.010663763340562581
    },
    {
      "action_query_batch_std": 0.4614201545715332,
      "action_query_norm": 44.451694869995116,
      "action_relation_loss": 0.01028045816347003,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.229186155285314,
      "flow_matching_loss": 0.05852229334414005,
      "grad_norm": 0.2121015340089798,
      "h_stock_valid_norm": 97.81442489624024,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.500633897471106e-05,
      "loss": 0.0688,
      "positive_teacher_pairs": 27174.9,
      "query_h_norm_ratio": 0.4544524222612381,
      "relation_loss": 0.01028045816347003,
      "selected_pairs": 27174.9,
      "step": 5590,
      "student_distance_mean": 30.24936237335205,
      "teacher_distance_mean": 0.8582954406738281,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.01028045816347003,
      "weighted_relation_loss": 0.01028045816347003
    },
    {
      "action_query_batch_std": 0.46422176659107206,
      "action_query_norm": 44.46242561340332,
      "action_relation_loss": 0.01050627389922738,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.238540692235734,
      "flow_matching_loss": 0.05849032290279865,
      "grad_norm": 0.19582752883434296,
      "h_stock_valid_norm": 98.09600677490235,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.498230170943596e-05,
      "loss": 0.069,
      "positive_teacher_pairs": 26855.0,
      "query_h_norm_ratio": 0.45326103568077086,
      "relation_loss": 0.01050627389922738,
      "selected_pairs": 26855.0,
      "step": 5600,
      "student_distance_mean": 30.645285987854002,
      "teacher_distance_mean": 0.8573739290237427,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.01050627389922738,
      "weighted_relation_loss": 0.01050627389922738
    },
    {
      "action_query_batch_std": 0.4617850840091705,
      "action_query_norm": 44.45829277038574,
      "action_relation_loss": 0.009888693969696761,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.247895229186155,
      "flow_matching_loss": 0.05889240466058254,
      "grad_norm": 0.1980690360069275,
      "h_stock_valid_norm": 98.26953506469727,
      "invalid_tail_samples": 22.7,
      "learning_rate": 9.495820978642275e-05,
      "loss": 0.0688,
      "positive_teacher_pairs": 27106.6,
      "query_h_norm_ratio": 0.45242084860801696,
      "relation_loss": 0.009888693969696761,
      "selected_pairs": 27106.6,
      "step": 5610,
      "student_distance_mean": 30.41705780029297,
      "teacher_distance_mean": 0.8524515271186829,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.009888693969696761,
      "weighted_relation_loss": 0.009888693969696761
    },
    {
      "action_query_batch_std": 0.4674449384212494,
      "action_query_norm": 44.462676239013675,
      "action_relation_loss": 0.009547174908220768,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.257249766136576,
      "flow_matching_loss": 0.05532098487019539,
      "grad_norm": 0.20321601629257202,
      "h_stock_valid_norm": 98.16927108764648,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.493406323494535e-05,
      "loss": 0.0649,
      "positive_teacher_pairs": 26876.7,
      "query_h_norm_ratio": 0.45292515158653257,
      "relation_loss": 0.009547174908220768,
      "selected_pairs": 26876.7,
      "step": 5620,
      "student_distance_mean": 30.727289962768555,
      "teacher_distance_mean": 0.8634278893470764,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.009547174908220768,
      "weighted_relation_loss": 0.009547174908220768
    },
    {
      "action_query_batch_std": 0.46321935057640073,
      "action_query_norm": 44.453612899780275,
      "action_relation_loss": 0.010539350286126136,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.266604303086997,
      "flow_matching_loss": 0.05568384751677513,
      "grad_norm": 0.20078308880329132,
      "h_stock_valid_norm": 97.29437942504883,
      "invalid_tail_samples": 26.4,
      "learning_rate": 9.490986208434413e-05,
      "loss": 0.0662,
      "positive_teacher_pairs": 26261.9,
      "query_h_norm_ratio": 0.4569007635116577,
      "relation_loss": 0.010539350286126136,
      "selected_pairs": 26261.9,
      "step": 5630,
      "student_distance_mean": 30.654185104370118,
      "teacher_distance_mean": 0.8597780525684356,
      "valid_action_samples": 229.6,
      "weighted_auxiliary_loss": 0.010539350286126136,
      "weighted_relation_loss": 0.010539350286126136
    },
    {
      "action_query_batch_std": 0.46961563229560854,
      "action_query_norm": 44.46015739440918,
      "action_relation_loss": 0.009664115775376559,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.275958840037418,
      "flow_matching_loss": 0.05684331543743611,
      "grad_norm": 0.19611386954784393,
      "h_stock_valid_norm": 96.91186904907227,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.488560636402577e-05,
      "loss": 0.0665,
      "positive_teacher_pairs": 27202.2,
      "query_h_norm_ratio": 0.4587726265192032,
      "relation_loss": 0.009664115775376559,
      "selected_pairs": 27202.2,
      "step": 5640,
      "student_distance_mean": 30.828265762329103,
      "teacher_distance_mean": 0.8587603509426117,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.009664115775376559,
      "weighted_relation_loss": 0.009664115775376559
    },
    {
      "action_query_batch_std": 0.45870558321475985,
      "action_query_norm": 44.44692764282227,
      "action_relation_loss": 0.010662468057125807,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.285313376987839,
      "flow_matching_loss": 0.055469457432627677,
      "grad_norm": 0.1589549481868744,
      "h_stock_valid_norm": 96.83389053344726,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.486129610346321e-05,
      "loss": 0.0661,
      "positive_teacher_pairs": 27172.4,
      "query_h_norm_ratio": 0.45900379717350004,
      "relation_loss": 0.010662468057125807,
      "selected_pairs": 27172.4,
      "step": 5650,
      "student_distance_mean": 30.080289840698242,
      "teacher_distance_mean": 0.8472028136253357,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.010662468057125807,
      "weighted_relation_loss": 0.010662468057125807
    },
    {
      "action_query_batch_std": 0.46071145832538607,
      "action_query_norm": 44.4477783203125,
      "action_relation_loss": 0.010131543409079314,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.29466791393826,
      "flow_matching_loss": 0.05807409957051277,
      "grad_norm": 0.16800439357757568,
      "h_stock_valid_norm": 97.0482292175293,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.483693133219576e-05,
      "loss": 0.0682,
      "positive_teacher_pairs": 26824.9,
      "query_h_norm_ratio": 0.45801106095314026,
      "relation_loss": 0.010131543409079314,
      "selected_pairs": 26824.9,
      "step": 5660,
      "student_distance_mean": 30.207889556884766,
      "teacher_distance_mean": 0.8477981448173523,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.010131543409079314,
      "weighted_relation_loss": 0.010131543409079314
    },
    {
      "action_query_batch_std": 0.4695704519748688,
      "action_query_norm": 44.458303451538086,
      "action_relation_loss": 0.010258884262293577,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.304022450888681,
      "flow_matching_loss": 0.05873115137219429,
      "grad_norm": 0.16547974944114685,
      "h_stock_valid_norm": 98.67412185668945,
      "invalid_tail_samples": 25.5,
      "learning_rate": 9.481251207982888e-05,
      "loss": 0.069,
      "positive_teacher_pairs": 26458.3,
      "query_h_norm_ratio": 0.45056706070899966,
      "relation_loss": 0.010258884262293577,
      "selected_pairs": 26458.3,
      "step": 5670,
      "student_distance_mean": 30.905543899536134,
      "teacher_distance_mean": 0.854397988319397,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.010258884262293577,
      "weighted_relation_loss": 0.010258884262293577
    },
    {
      "action_query_batch_std": 0.4614900439977646,
      "action_query_norm": 44.44602966308594,
      "action_relation_loss": 0.009865005500614643,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.313376987839102,
      "flow_matching_loss": 0.056024659797549246,
      "grad_norm": 0.14838598668575287,
      "h_stock_valid_norm": 98.85812454223633,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.47880383760343e-05,
      "loss": 0.0659,
      "positive_teacher_pairs": 26971.7,
      "query_h_norm_ratio": 0.44959723949432373,
      "relation_loss": 0.009865005500614643,
      "selected_pairs": 26971.7,
      "step": 5680,
      "student_distance_mean": 30.379248046875,
      "teacher_distance_mean": 0.854418933391571,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.009865005500614643,
      "weighted_relation_loss": 0.009865005500614643
    },
    {
      "action_query_batch_std": 0.45787869691848754,
      "action_query_norm": 44.44284286499023,
      "action_relation_loss": 0.010103879403322935,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.322731524789523,
      "flow_matching_loss": 0.055026764050126076,
      "grad_norm": 0.14275801181793213,
      "h_stock_valid_norm": 98.22311935424804,
      "invalid_tail_samples": 23.1,
      "learning_rate": 9.476351025054983e-05,
      "loss": 0.0651,
      "positive_teacher_pairs": 27018.2,
      "query_h_norm_ratio": 0.4524759978055954,
      "relation_loss": 0.010103879403322935,
      "selected_pairs": 27018.2,
      "step": 5690,
      "student_distance_mean": 30.06444149017334,
      "teacher_distance_mean": 0.8476301014423371,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.010103879403322935,
      "weighted_relation_loss": 0.010103879403322935
    },
    {
      "action_query_batch_std": 0.47446260452270506,
      "action_query_norm": 44.46695518493652,
      "action_relation_loss": 0.010197317507117987,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.332086061739944,
      "flow_matching_loss": 0.05758887454867363,
      "grad_norm": 0.14005981385707855,
      "h_stock_valid_norm": 97.81685256958008,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.473892773317952e-05,
      "loss": 0.0678,
      "positive_teacher_pairs": 26873.0,
      "query_h_norm_ratio": 0.45460157096385956,
      "relation_loss": 0.010197317507117987,
      "selected_pairs": 26873.0,
      "step": 5700,
      "student_distance_mean": 31.150943946838378,
      "teacher_distance_mean": 0.8639989137649536,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.010197317507117987,
      "weighted_relation_loss": 0.010197317507117987
    },
    {
      "action_query_batch_std": 0.45746939480304716,
      "action_query_norm": 44.43068580627441,
      "action_relation_loss": 0.010447035171091556,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.341440598690365,
      "flow_matching_loss": 0.05494793392717838,
      "grad_norm": 0.16450324654579163,
      "h_stock_valid_norm": 96.70695266723632,
      "invalid_tail_samples": 26.3,
      "learning_rate": 9.471429085379338e-05,
      "loss": 0.0654,
      "positive_teacher_pairs": 26276.1,
      "query_h_norm_ratio": 0.45944463908672334,
      "relation_loss": 0.010447035171091556,
      "selected_pairs": 26276.1,
      "step": 5710,
      "student_distance_mean": 30.124466323852538,
      "teacher_distance_mean": 0.8409736752510071,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.010447035171091556,
      "weighted_relation_loss": 0.010447035171091556
    },
    {
      "action_query_batch_std": 0.46417878568172455,
      "action_query_norm": 44.44221725463867,
      "action_relation_loss": 0.00970436930656433,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.3507951356407855,
      "flow_matching_loss": 0.05474642850458622,
      "grad_norm": 0.12987326085567474,
      "h_stock_valid_norm": 97.45935363769532,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.468959964232757e-05,
      "loss": 0.0645,
      "positive_teacher_pairs": 27154.9,
      "query_h_norm_ratio": 0.456025430560112,
      "relation_loss": 0.00970436930656433,
      "selected_pairs": 27154.9,
      "step": 5720,
      "student_distance_mean": 30.56943588256836,
      "teacher_distance_mean": 0.86240856051445,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.00970436930656433,
      "weighted_relation_loss": 0.00970436930656433
    },
    {
      "action_query_batch_std": 0.46095643043518064,
      "action_query_norm": 44.439216232299806,
      "action_relation_loss": 0.010550551302731036,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.360149672591207,
      "flow_matching_loss": 0.057208411395549774,
      "grad_norm": 0.16043561697006226,
      "h_stock_valid_norm": 98.30114364624023,
      "invalid_tail_samples": 25.1,
      "learning_rate": 9.466485412878425e-05,
      "loss": 0.0678,
      "positive_teacher_pairs": 26553.9,
      "query_h_norm_ratio": 0.4520761251449585,
      "relation_loss": 0.010550551302731036,
      "selected_pairs": 26553.9,
      "step": 5730,
      "student_distance_mean": 30.391420364379883,
      "teacher_distance_mean": 0.85621936917305,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.010550551302731036,
      "weighted_relation_loss": 0.010550551302731036
    },
    {
      "action_query_batch_std": 0.4659139782190323,
      "action_query_norm": 44.443824768066406,
      "action_relation_loss": 0.010489293374121189,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.369504209541628,
      "flow_matching_loss": 0.055446216464042665,
      "grad_norm": 0.1806381195783615,
      "h_stock_valid_norm": 98.24057006835938,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.464005434323154e-05,
      "loss": 0.0659,
      "positive_teacher_pairs": 26803.3,
      "query_h_norm_ratio": 0.452401003241539,
      "relation_loss": 0.010489293374121189,
      "selected_pairs": 26803.3,
      "step": 5740,
      "student_distance_mean": 30.457016372680663,
      "teacher_distance_mean": 0.8550743043422699,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.010489293374121189,
      "weighted_relation_loss": 0.010489293374121189
    },
    {
      "action_query_batch_std": 0.47198974192142484,
      "action_query_norm": 44.45279808044434,
      "action_relation_loss": 0.010851148702204227,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.378858746492049,
      "flow_matching_loss": 0.0572961263358593,
      "grad_norm": 0.15322798490524292,
      "h_stock_valid_norm": 98.16355209350586,
      "invalid_tail_samples": 22.9,
      "learning_rate": 9.461520031580352e-05,
      "loss": 0.0681,
      "positive_teacher_pairs": 27059.6,
      "query_h_norm_ratio": 0.4528502941131592,
      "relation_loss": 0.010851148702204227,
      "selected_pairs": 27059.6,
      "step": 5750,
      "student_distance_mean": 31.049225997924804,
      "teacher_distance_mean": 0.8533148527145386,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.010851148702204227,
      "weighted_relation_loss": 0.010851148702204227
    },
    {
      "action_query_batch_std": 0.47300895154476164,
      "action_query_norm": 44.45415077209473,
      "action_relation_loss": 0.009969782549887896,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.38821328344247,
      "flow_matching_loss": 0.05691568776965141,
      "grad_norm": 0.16113042831420898,
      "h_stock_valid_norm": 97.64050216674805,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.459029207670019e-05,
      "loss": 0.0669,
      "positive_teacher_pairs": 27181.1,
      "query_h_norm_ratio": 0.455286905169487,
      "relation_loss": 0.009969782549887896,
      "selected_pairs": 27181.1,
      "step": 5760,
      "student_distance_mean": 31.06012420654297,
      "teacher_distance_mean": 0.8608278632164001,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.009969782549887896,
      "weighted_relation_loss": 0.009969782549887896
    },
    {
      "action_query_batch_std": 0.4602161645889282,
      "action_query_norm": 44.432304763793944,
      "action_relation_loss": 0.009685267694294453,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.397567820392891,
      "flow_matching_loss": 0.057624992728233335,
      "grad_norm": 0.20415723323822021,
      "h_stock_valid_norm": 97.2189323425293,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.456532965618737e-05,
      "loss": 0.0673,
      "positive_teacher_pairs": 26613.9,
      "query_h_norm_ratio": 0.4570468544960022,
      "relation_loss": 0.009685267694294453,
      "selected_pairs": 26613.9,
      "step": 5770,
      "student_distance_mean": 30.284081649780273,
      "teacher_distance_mean": 0.8584268271923066,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.009685267694294453,
      "weighted_relation_loss": 0.009685267694294453
    },
    {
      "action_query_batch_std": 0.45905067026615143,
      "action_query_norm": 44.420373153686526,
      "action_relation_loss": 0.010402035992592573,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.406922357343311,
      "flow_matching_loss": 0.059287427365779875,
      "grad_norm": 0.17394444346427917,
      "h_stock_valid_norm": 96.95918045043945,
      "invalid_tail_samples": 22.0,
      "learning_rate": 9.454031308459681e-05,
      "loss": 0.0697,
      "positive_teacher_pairs": 27269.6,
      "query_h_norm_ratio": 0.45814719796180725,
      "relation_loss": 0.010402035992592573,
      "selected_pairs": 27269.6,
      "step": 5780,
      "student_distance_mean": 30.085239791870116,
      "teacher_distance_mean": 0.8550061762332917,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.010402035992592573,
      "weighted_relation_loss": 0.010402035992592573
    },
    {
      "action_query_batch_std": 0.4651230573654175,
      "action_query_norm": 44.4313247680664,
      "action_relation_loss": 0.010210509970784187,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.416276894293732,
      "flow_matching_loss": 0.05614817179739475,
      "grad_norm": 0.15981845557689667,
      "h_stock_valid_norm": 97.91874465942382,
      "invalid_tail_samples": 25.7,
      "learning_rate": 9.451524239232595e-05,
      "loss": 0.0664,
      "positive_teacher_pairs": 26415.6,
      "query_h_norm_ratio": 0.4537590891122818,
      "relation_loss": 0.010210509970784187,
      "selected_pairs": 26415.6,
      "step": 5790,
      "student_distance_mean": 30.595170783996583,
      "teacher_distance_mean": 0.857056725025177,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.010210509970784187,
      "weighted_relation_loss": 0.010210509970784187
    },
    {
      "action_query_batch_std": 0.45523581802845003,
      "action_query_norm": 44.409922409057614,
      "action_relation_loss": 0.00971279013901949,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.425631431244153,
      "flow_matching_loss": 0.05711888372898102,
      "grad_norm": 0.16253171861171722,
      "h_stock_valid_norm": 97.98268966674804,
      "invalid_tail_samples": 25.4,
      "learning_rate": 9.449011760983809e-05,
      "loss": 0.0668,
      "positive_teacher_pairs": 26490.2,
      "query_h_norm_ratio": 0.45324391722679136,
      "relation_loss": 0.00971279013901949,
      "selected_pairs": 26490.2,
      "step": 5800,
      "student_distance_mean": 30.042434692382812,
      "teacher_distance_mean": 0.8560605823993683,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.00971279013901949,
      "weighted_relation_loss": 0.00971279013901949
    },
    {
      "action_query_batch_std": 0.46223153471946715,
      "action_query_norm": 44.42335090637207,
      "action_relation_loss": 0.00989930909126997,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.4349859681945745,
      "flow_matching_loss": 0.05642121359705925,
      "grad_norm": 0.16943354904651642,
      "h_stock_valid_norm": 98.16489028930664,
      "invalid_tail_samples": 21.9,
      "learning_rate": 9.446493876766218e-05,
      "loss": 0.0663,
      "positive_teacher_pairs": 27287.8,
      "query_h_norm_ratio": 0.4525415688753128,
      "relation_loss": 0.00989930909126997,
      "selected_pairs": 27287.8,
      "step": 5810,
      "student_distance_mean": 30.309312629699708,
      "teacher_distance_mean": 0.86078240275383,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.00989930909126997,
      "weighted_relation_loss": 0.00989930909126997
    },
    {
      "action_query_batch_std": 0.46590551137924197,
      "action_query_norm": 44.42822074890137,
      "action_relation_loss": 0.009857076313346624,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.444340505144996,
      "flow_matching_loss": 0.05988284945487976,
      "grad_norm": 0.17071965336799622,
      "h_stock_valid_norm": 97.84404678344727,
      "invalid_tail_samples": 22.9,
      "learning_rate": 9.44397058963929e-05,
      "loss": 0.0697,
      "positive_teacher_pairs": 27059.5,
      "query_h_norm_ratio": 0.45407589077949523,
      "relation_loss": 0.009857076313346624,
      "selected_pairs": 27059.5,
      "step": 5820,
      "student_distance_mean": 30.603860282897948,
      "teacher_distance_mean": 0.8592698216438294,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.009857076313346624,
      "weighted_relation_loss": 0.009857076313346624
    },
    {
      "action_query_batch_std": 0.4658143579959869,
      "action_query_norm": 44.42831230163574,
      "action_relation_loss": 0.010277347918599845,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.453695042095417,
      "flow_matching_loss": 0.05551888905465603,
      "grad_norm": 0.17104381322860718,
      "h_stock_valid_norm": 97.3867088317871,
      "invalid_tail_samples": 24.9,
      "learning_rate": 9.441441902669056e-05,
      "loss": 0.0658,
      "positive_teacher_pairs": 26595.7,
      "query_h_norm_ratio": 0.4562119603157043,
      "relation_loss": 0.010277347918599845,
      "selected_pairs": 26595.7,
      "step": 5830,
      "student_distance_mean": 30.63106002807617,
      "teacher_distance_mean": 0.852252197265625,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.010277347918599845,
      "weighted_relation_loss": 0.010277347918599845
    },
    {
      "action_query_batch_std": 0.46761485636234285,
      "action_query_norm": 44.42727851867676,
      "action_relation_loss": 0.010146623663604259,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.463049579045837,
      "flow_matching_loss": 0.057515525817871095,
      "grad_norm": 0.11148476600646973,
      "h_stock_valid_norm": 96.78280563354492,
      "invalid_tail_samples": 23.1,
      "learning_rate": 9.43890781892811e-05,
      "loss": 0.0677,
      "positive_teacher_pairs": 27014.4,
      "query_h_norm_ratio": 0.45905018150806426,
      "relation_loss": 0.010146623663604259,
      "selected_pairs": 27014.4,
      "step": 5840,
      "student_distance_mean": 30.710929107666015,
      "teacher_distance_mean": 0.8576492011547089,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.010146623663604259,
      "weighted_relation_loss": 0.010146623663604259
    },
    {
      "action_query_batch_std": 0.4621096134185791,
      "action_query_norm": 44.41819610595703,
      "action_relation_loss": 0.010024525038897992,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.472404115996258,
      "flow_matching_loss": 0.05572235658764839,
      "grad_norm": 0.18359176814556122,
      "h_stock_valid_norm": 96.94707260131835,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.436368341495603e-05,
      "loss": 0.0657,
      "positive_teacher_pairs": 26949.8,
      "query_h_norm_ratio": 0.45818165242671965,
      "relation_loss": 0.010024525038897992,
      "selected_pairs": 26949.8,
      "step": 5850,
      "student_distance_mean": 30.42134666442871,
      "teacher_distance_mean": 0.8625922679901123,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.010024525038897992,
      "weighted_relation_loss": 0.010024525038897992
    },
    {
      "action_query_batch_std": 0.46028893887996675,
      "action_query_norm": 44.41684532165527,
      "action_relation_loss": 0.010288200806826353,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.481758652946679,
      "flow_matching_loss": 0.05661280006170273,
      "grad_norm": 0.16207651793956757,
      "h_stock_valid_norm": 97.71236190795898,
      "invalid_tail_samples": 21.7,
      "learning_rate": 9.43382347345724e-05,
      "loss": 0.0669,
      "positive_teacher_pairs": 27335.9,
      "query_h_norm_ratio": 0.4545745700597763,
      "relation_loss": 0.010288200806826353,
      "selected_pairs": 27335.9,
      "step": 5860,
      "student_distance_mean": 30.250422477722168,
      "teacher_distance_mean": 0.84439337849617,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.010288200806826353,
      "weighted_relation_loss": 0.010288200806826353
    },
    {
      "action_query_batch_std": 0.4501174807548523,
      "action_query_norm": 44.402897644042966,
      "action_relation_loss": 0.01025913804769516,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.4911131898971,
      "flow_matching_loss": 0.055977766215801236,
      "grad_norm": 0.10780028253793716,
      "h_stock_valid_norm": 97.14859542846679,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.431273217905273e-05,
      "loss": 0.0662,
      "positive_teacher_pairs": 26948.9,
      "query_h_norm_ratio": 0.4570683866739273,
      "relation_loss": 0.01025913804769516,
      "selected_pairs": 26948.9,
      "step": 5870,
      "student_distance_mean": 29.730620193481446,
      "teacher_distance_mean": 0.8482034087181092,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.01025913804769516,
      "weighted_relation_loss": 0.01025913804769516
    },
    {
      "action_query_batch_std": 0.46259030401706697,
      "action_query_norm": 44.41642951965332,
      "action_relation_loss": 0.009681705757975579,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.500467726847521,
      "flow_matching_loss": 0.05528463907539845,
      "grad_norm": 0.14801591634750366,
      "h_stock_valid_norm": 97.46987762451172,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.428717577938504e-05,
      "loss": 0.065,
      "positive_teacher_pairs": 26688.4,
      "query_h_norm_ratio": 0.4557032436132431,
      "relation_loss": 0.009681705757975579,
      "selected_pairs": 26688.4,
      "step": 5880,
      "student_distance_mean": 30.4281099319458,
      "teacher_distance_mean": 0.8569596529006958,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.009681705757975579,
      "weighted_relation_loss": 0.009681705757975579
    },
    {
      "action_query_batch_std": 0.4725989431142807,
      "action_query_norm": 44.42492256164551,
      "action_relation_loss": 0.009510473441332578,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.509822263797942,
      "flow_matching_loss": 0.05678310878574848,
      "grad_norm": 0.14036735892295837,
      "h_stock_valid_norm": 98.80630722045899,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.426156556662276e-05,
      "loss": 0.0663,
      "positive_teacher_pairs": 26641.6,
      "query_h_norm_ratio": 0.4496217042207718,
      "relation_loss": 0.009510473441332578,
      "selected_pairs": 26641.6,
      "step": 5890,
      "student_distance_mean": 31.064429092407227,
      "teacher_distance_mean": 0.8621391654014587,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.009510473441332578,
      "weighted_relation_loss": 0.009510473441332578
    },
    {
      "action_query_batch_std": 0.45610350370407104,
      "action_query_norm": 44.40373649597168,
      "action_relation_loss": 0.00928083686158061,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.519176800748363,
      "flow_matching_loss": 0.05652114674448967,
      "grad_norm": 0.14504118263721466,
      "h_stock_valid_norm": 98.30258407592774,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.423590157188474e-05,
      "loss": 0.0658,
      "positive_teacher_pairs": 26851.1,
      "query_h_norm_ratio": 0.4517204940319061,
      "relation_loss": 0.00928083686158061,
      "selected_pairs": 26851.1,
      "step": 5900,
      "student_distance_mean": 30.01916847229004,
      "teacher_distance_mean": 0.8449424505233765,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.00928083686158061,
      "weighted_relation_loss": 0.00928083686158061
    },
    {
      "action_query_batch_std": 0.4580643057823181,
      "action_query_norm": 44.40094413757324,
      "action_relation_loss": 0.009931937139481307,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.528531337698784,
      "flow_matching_loss": 0.05799199864268303,
      "grad_norm": 0.21096394956111908,
      "h_stock_valid_norm": 97.13454742431641,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.421018382635513e-05,
      "loss": 0.0679,
      "positive_teacher_pairs": 26630.4,
      "query_h_norm_ratio": 0.4571189761161804,
      "relation_loss": 0.009931937139481307,
      "selected_pairs": 26630.4,
      "step": 5910,
      "student_distance_mean": 30.211460304260253,
      "teacher_distance_mean": 0.8508593380451203,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.009931937139481307,
      "weighted_relation_loss": 0.009931937139481307
    },
    {
      "action_query_batch_std": 0.4682838559150696,
      "action_query_norm": 44.416053009033206,
      "action_relation_loss": 0.00926386248320341,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.537885874649205,
      "flow_matching_loss": 0.05699743926525116,
      "grad_norm": 0.1435352861881256,
      "h_stock_valid_norm": 97.40003051757813,
      "invalid_tail_samples": 24.4,
      "learning_rate": 9.418441236128343e-05,
      "loss": 0.0663,
      "positive_teacher_pairs": 26711.6,
      "query_h_norm_ratio": 0.4560258716344833,
      "relation_loss": 0.00926386248320341,
      "selected_pairs": 26711.6,
      "step": 5920,
      "student_distance_mean": 30.863034439086913,
      "teacher_distance_mean": 0.8722420156002044,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.00926386248320341,
      "weighted_relation_loss": 0.00926386248320341
    },
    {
      "action_query_batch_std": 0.4587642639875412,
      "action_query_norm": 44.40054626464844,
      "action_relation_loss": 0.009795720223337412,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.547240411599626,
      "flow_matching_loss": 0.054876609891653064,
      "grad_norm": 0.164826437830925,
      "h_stock_valid_norm": 97.47001571655274,
      "invalid_tail_samples": 20.0,
      "learning_rate": 9.41585872079844e-05,
      "loss": 0.0647,
      "positive_teacher_pairs": 27744.1,
      "query_h_norm_ratio": 0.45553444921970365,
      "relation_loss": 0.009795720223337412,
      "selected_pairs": 27744.1,
      "step": 5930,
      "student_distance_mean": 30.208484840393066,
      "teacher_distance_mean": 0.8573518812656402,
      "valid_action_samples": 236.0,
      "weighted_auxiliary_loss": 0.009795720223337412,
      "weighted_relation_loss": 0.009795720223337412
    },
    {
      "action_query_batch_std": 0.45338426530361176,
      "action_query_norm": 44.38927726745605,
      "action_relation_loss": 0.010067136771976948,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.556594948550047,
      "flow_matching_loss": 0.055327174812555314,
      "grad_norm": 0.13309234380722046,
      "h_stock_valid_norm": 97.84673919677735,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.413270839783802e-05,
      "loss": 0.0654,
      "positive_teacher_pairs": 26741.4,
      "query_h_norm_ratio": 0.45366383492946627,
      "relation_loss": 0.010067136771976948,
      "selected_pairs": 26741.4,
      "step": 5940,
      "student_distance_mean": 29.89870414733887,
      "teacher_distance_mean": 0.8511324286460876,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.010067136771976948,
      "weighted_relation_loss": 0.010067136771976948
    },
    {
      "action_query_batch_std": 0.45861105620861053,
      "action_query_norm": 44.40301628112793,
      "action_relation_loss": 0.009710448607802392,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.565949485500468,
      "flow_matching_loss": 0.05624349340796471,
      "grad_norm": 0.1522451788187027,
      "h_stock_valid_norm": 98.33356018066407,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.41067759622895e-05,
      "loss": 0.066,
      "positive_teacher_pairs": 26776.6,
      "query_h_norm_ratio": 0.451559841632843,
      "relation_loss": 0.009710448607802392,
      "selected_pairs": 26776.6,
      "step": 5950,
      "student_distance_mean": 30.245254898071288,
      "teacher_distance_mean": 0.8503491759300232,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.009710448607802392,
      "weighted_relation_loss": 0.009710448607802392
    },
    {
      "action_query_batch_std": 0.46321536004543307,
      "action_query_norm": 44.40262794494629,
      "action_relation_loss": 0.009812956489622594,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.575304022450888,
      "flow_matching_loss": 0.05572629980742931,
      "grad_norm": 0.2080535739660263,
      "h_stock_valid_norm": 97.57323150634765,
      "invalid_tail_samples": 25.2,
      "learning_rate": 9.408078993284917e-05,
      "loss": 0.0655,
      "positive_teacher_pairs": 26529.1,
      "query_h_norm_ratio": 0.45508659183979033,
      "relation_loss": 0.009812956489622594,
      "selected_pairs": 26529.1,
      "step": 5960,
      "student_distance_mean": 30.467256355285645,
      "teacher_distance_mean": 0.854499101638794,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.009812956489622594,
      "weighted_relation_loss": 0.009812956489622594
    },
    {
      "action_query_batch_std": 0.4522635519504547,
      "action_query_norm": 44.3835391998291,
      "action_relation_loss": 0.010113280359655618,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.584658559401309,
      "flow_matching_loss": 0.057617134600877765,
      "grad_norm": 0.1390523463487625,
      "h_stock_valid_norm": 96.16315612792968,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.405475034109254e-05,
      "loss": 0.0677,
      "positive_teacher_pairs": 26644.7,
      "query_h_norm_ratio": 0.4615590959787369,
      "relation_loss": 0.010113280359655618,
      "selected_pairs": 26644.7,
      "step": 5970,
      "student_distance_mean": 29.890667152404784,
      "teacher_distance_mean": 0.8532105088233948,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.010113280359655618,
      "weighted_relation_loss": 0.010113280359655618
    },
    {
      "action_query_batch_std": 0.4637233555316925,
      "action_query_norm": 44.39803581237793,
      "action_relation_loss": 0.009525015950202942,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.5940130963517305,
      "flow_matching_loss": 0.05537530966103077,
      "grad_norm": 0.19020560383796692,
      "h_stock_valid_norm": 95.47241592407227,
      "invalid_tail_samples": 26.2,
      "learning_rate": 9.402865721866015e-05,
      "loss": 0.0649,
      "positive_teacher_pairs": 26301.0,
      "query_h_norm_ratio": 0.46503997445106504,
      "relation_loss": 0.009525015950202942,
      "selected_pairs": 26301.0,
      "step": 5980,
      "student_distance_mean": 30.54818992614746,
      "teacher_distance_mean": 0.856737595796585,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.009525015950202942,
      "weighted_relation_loss": 0.009525015950202942
    },
    {
      "action_query_batch_std": 0.469675350189209,
      "action_query_norm": 44.40114784240723,
      "action_relation_loss": 0.00949862003326416,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.603367633302152,
      "flow_matching_loss": 0.05509806536138058,
      "grad_norm": 0.12254655361175537,
      "h_stock_valid_norm": 97.09120407104493,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.400251059725762e-05,
      "loss": 0.0646,
      "positive_teacher_pairs": 26803.1,
      "query_h_norm_ratio": 0.4573262155056,
      "relation_loss": 0.00949862003326416,
      "selected_pairs": 26803.1,
      "step": 5990,
      "student_distance_mean": 30.855333137512208,
      "teacher_distance_mean": 0.8650400519371033,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.00949862003326416,
      "weighted_relation_loss": 0.00949862003326416
    },
    {
      "action_query_batch_std": 0.45354111194610597,
      "action_query_norm": 44.383988571166995,
      "action_relation_loss": 0.00966969970613718,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.612722170252573,
      "flow_matching_loss": 0.0564645629376173,
      "grad_norm": 0.16312046349048615,
      "h_stock_valid_norm": 97.69105072021485,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.397631050865554e-05,
      "loss": 0.0661,
      "positive_teacher_pairs": 26915.3,
      "query_h_norm_ratio": 0.45433710515499115,
      "relation_loss": 0.00966969970613718,
      "selected_pairs": 26915.3,
      "step": 6000,
      "student_distance_mean": 29.91804027557373,
      "teacher_distance_mean": 0.8534127831459045,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.00966969970613718,
      "weighted_relation_loss": 0.00966969970613718
    },
    {
      "action_query_batch_std": 0.4498323738574982,
      "action_query_norm": 44.37339820861816,
      "action_relation_loss": 0.009824927523732185,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.622076707202994,
      "flow_matching_loss": 0.05476266667246819,
      "grad_norm": 0.14432354271411896,
      "h_stock_valid_norm": 97.81100006103516,
      "invalid_tail_samples": 23.1,
      "learning_rate": 9.395005698468949e-05,
      "loss": 0.0646,
      "positive_teacher_pairs": 27026.6,
      "query_h_norm_ratio": 0.4536666214466095,
      "relation_loss": 0.009824927523732185,
      "selected_pairs": 27026.6,
      "step": 6010,
      "student_distance_mean": 29.687688636779786,
      "teacher_distance_mean": 0.8463877856731414,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.009824927523732185,
      "weighted_relation_loss": 0.009824927523732185
    },
    {
      "action_query_batch_std": 0.4553781390190125,
      "action_query_norm": 44.3751766204834,
      "action_relation_loss": 0.010238135978579522,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.631431244153415,
      "flow_matching_loss": 0.05504533052444458,
      "grad_norm": 0.19314324855804443,
      "h_stock_valid_norm": 96.22923431396484,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.392375005726e-05,
      "loss": 0.0653,
      "positive_teacher_pairs": 27200.2,
      "query_h_norm_ratio": 0.4611561566591263,
      "relation_loss": 0.010238135978579522,
      "selected_pairs": 27200.2,
      "step": 6020,
      "student_distance_mean": 29.904963111877443,
      "teacher_distance_mean": 0.8600810110569,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.010238135978579522,
      "weighted_relation_loss": 0.010238135978579522
    },
    {
      "action_query_batch_std": 0.4529807478189468,
      "action_query_norm": 44.37394065856934,
      "action_relation_loss": 0.009526697173714637,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.640785781103835,
      "flow_matching_loss": 0.05311926007270813,
      "grad_norm": 0.13627299666404724,
      "h_stock_valid_norm": 96.03200759887696,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.389738975833242e-05,
      "loss": 0.0626,
      "positive_teacher_pairs": 26696.7,
      "query_h_norm_ratio": 0.46207712292671205,
      "relation_loss": 0.009526697173714637,
      "selected_pairs": 26696.7,
      "step": 6030,
      "student_distance_mean": 29.786463737487793,
      "teacher_distance_mean": 0.850164270401001,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.009526697173714637,
      "weighted_relation_loss": 0.009526697173714637
    },
    {
      "action_query_batch_std": 0.4566999912261963,
      "action_query_norm": 44.379062271118165,
      "action_relation_loss": 0.01033007651567459,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.650140318054256,
      "flow_matching_loss": 0.055476054176688196,
      "grad_norm": 0.12786848843097687,
      "h_stock_valid_norm": 96.61761932373047,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.387097611993707e-05,
      "loss": 0.0658,
      "positive_teacher_pairs": 26919.0,
      "query_h_norm_ratio": 0.45932982861995697,
      "relation_loss": 0.01033007651567459,
      "selected_pairs": 26919.0,
      "step": 6040,
      "student_distance_mean": 29.903967666625977,
      "teacher_distance_mean": 0.8568325996398926,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.01033007651567459,
      "weighted_relation_loss": 0.01033007651567459
    },
    {
      "action_query_batch_std": 0.4587705433368683,
      "action_query_norm": 44.375949478149415,
      "action_relation_loss": 0.009854731056839228,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.659494855004677,
      "flow_matching_loss": 0.05635628402233124,
      "grad_norm": 0.10197443515062332,
      "h_stock_valid_norm": 97.28751983642579,
      "invalid_tail_samples": 23.1,
      "learning_rate": 9.384450917416894e-05,
      "loss": 0.0662,
      "positive_teacher_pairs": 27009.5,
      "query_h_norm_ratio": 0.45614399909973147,
      "relation_loss": 0.009854731056839228,
      "selected_pairs": 27009.5,
      "step": 6050,
      "student_distance_mean": 30.26084327697754,
      "teacher_distance_mean": 0.8578768312931061,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.009854731056839228,
      "weighted_relation_loss": 0.009854731056839228
    },
    {
      "action_query_batch_std": 0.45628006756305695,
      "action_query_norm": 44.37712364196777,
      "action_relation_loss": 0.009685199242085218,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.668849391955098,
      "flow_matching_loss": 0.056571245566010474,
      "grad_norm": 0.12213756144046783,
      "h_stock_valid_norm": 98.20686798095703,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.381798895318792e-05,
      "loss": 0.0663,
      "positive_teacher_pairs": 26828.1,
      "query_h_norm_ratio": 0.451879546046257,
      "relation_loss": 0.009685199242085218,
      "selected_pairs": 26828.1,
      "step": 6060,
      "student_distance_mean": 30.085576629638673,
      "teacher_distance_mean": 0.8656470656394959,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.009685199242085218,
      "weighted_relation_loss": 0.009685199242085218
    },
    {
      "action_query_batch_std": 0.46138680875301363,
      "action_query_norm": 44.38360176086426,
      "action_relation_loss": 0.009507138840854168,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.6782039289055195,
      "flow_matching_loss": 0.05344950184226036,
      "grad_norm": 0.1352711170911789,
      "h_stock_valid_norm": 97.50467681884766,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.379141548921853e-05,
      "loss": 0.063,
      "positive_teacher_pairs": 27220.6,
      "query_h_norm_ratio": 0.45520086884498595,
      "relation_loss": 0.009507138840854168,
      "selected_pairs": 27220.6,
      "step": 6070,
      "student_distance_mean": 30.236332130432128,
      "teacher_distance_mean": 0.851851636171341,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.009507138840854168,
      "weighted_relation_loss": 0.009507138840854168
    },
    {
      "action_query_batch_std": 0.44787871539592744,
      "action_query_norm": 44.36513214111328,
      "action_relation_loss": 0.009846639633178712,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.68755846585594,
      "flow_matching_loss": 0.05751098096370697,
      "grad_norm": 0.13760802149772644,
      "h_stock_valid_norm": 96.67490081787109,
      "invalid_tail_samples": 22.0,
      "learning_rate": 9.376478881455009e-05,
      "loss": 0.0674,
      "positive_teacher_pairs": 27269.6,
      "query_h_norm_ratio": 0.4589144766330719,
      "relation_loss": 0.009846639633178712,
      "selected_pairs": 27269.6,
      "step": 6080,
      "student_distance_mean": 29.494686317443847,
      "teacher_distance_mean": 0.8459052443504333,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.009846639633178712,
      "weighted_relation_loss": 0.009846639633178712
    },
    {
      "action_query_batch_std": 0.44998755753040315,
      "action_query_norm": 44.36122055053711,
      "action_relation_loss": 0.009441150911152362,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.696913002806361,
      "flow_matching_loss": 0.05463320203125477,
      "grad_norm": 0.1919383406639099,
      "h_stock_valid_norm": 97.54773406982422,
      "invalid_tail_samples": 21.7,
      "learning_rate": 9.373810896153647e-05,
      "loss": 0.0641,
      "positive_teacher_pairs": 27342.2,
      "query_h_norm_ratio": 0.45476830303668975,
      "relation_loss": 0.009441150911152362,
      "selected_pairs": 27342.2,
      "step": 6090,
      "student_distance_mean": 29.61108341217041,
      "teacher_distance_mean": 0.8596122324466705,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.009441150911152362,
      "weighted_relation_loss": 0.009441150911152362
    },
    {
      "action_query_batch_std": 0.46030148267745974,
      "action_query_norm": 44.37832336425781,
      "action_relation_loss": 0.009287714026868343,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.706267539756782,
      "flow_matching_loss": 0.056083124876022336,
      "grad_norm": 0.14063061773777008,
      "h_stock_valid_norm": 97.6355598449707,
      "invalid_tail_samples": 21.7,
      "learning_rate": 9.371137596259623e-05,
      "loss": 0.0654,
      "positive_teacher_pairs": 27346.2,
      "query_h_norm_ratio": 0.45453484654426574,
      "relation_loss": 0.009287714026868343,
      "selected_pairs": 27346.2,
      "step": 6100,
      "student_distance_mean": 30.131442832946778,
      "teacher_distance_mean": 0.8575326919555664,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.009287714026868343,
      "weighted_relation_loss": 0.009287714026868343
    },
    {
      "action_query_batch_std": 0.45786209106445314,
      "action_query_norm": 44.37336540222168,
      "action_relation_loss": 0.010028759296983481,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.715622076707203,
      "flow_matching_loss": 0.056398984789848325,
      "grad_norm": 0.15998320281505585,
      "h_stock_valid_norm": 97.6114631652832,
      "invalid_tail_samples": 26.3,
      "learning_rate": 9.368458985021248e-05,
      "loss": 0.0664,
      "positive_teacher_pairs": 26277.3,
      "query_h_norm_ratio": 0.45459275841712954,
      "relation_loss": 0.010028759296983481,
      "selected_pairs": 26277.3,
      "step": 6110,
      "student_distance_mean": 30.140233421325682,
      "teacher_distance_mean": 0.8567015826702118,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.010028759296983481,
      "weighted_relation_loss": 0.010028759296983481
    },
    {
      "action_query_batch_std": 0.4560219973325729,
      "action_query_norm": 44.37617225646973,
      "action_relation_loss": 0.009682938177138566,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.724976613657624,
      "flow_matching_loss": 0.05857052505016327,
      "grad_norm": 0.1280076801776886,
      "h_stock_valid_norm": 97.25746841430664,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.365775065693287e-05,
      "loss": 0.0683,
      "positive_teacher_pairs": 26673.4,
      "query_h_norm_ratio": 0.4562770456075668,
      "relation_loss": 0.009682938177138566,
      "selected_pairs": 26673.4,
      "step": 6120,
      "student_distance_mean": 30.07155990600586,
      "teacher_distance_mean": 0.8517943441867828,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.009682938177138566,
      "weighted_relation_loss": 0.009682938177138566
    },
    {
      "action_query_batch_std": 0.45490917563438416,
      "action_query_norm": 44.37902717590332,
      "action_relation_loss": 0.009731645416468381,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.734331150608045,
      "flow_matching_loss": 0.05463410131633282,
      "grad_norm": 0.16940100491046906,
      "h_stock_valid_norm": 96.95821914672851,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.363085841536957e-05,
      "loss": 0.0644,
      "positive_teacher_pairs": 27148.0,
      "query_h_norm_ratio": 0.45771568417549136,
      "relation_loss": 0.009731645416468381,
      "selected_pairs": 27148.0,
      "step": 6130,
      "student_distance_mean": 30.041409492492676,
      "teacher_distance_mean": 0.8469519674777984,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.009731645416468381,
      "weighted_relation_loss": 0.009731645416468381
    },
    {
      "action_query_batch_std": 0.45238925516605377,
      "action_query_norm": 44.361468124389646,
      "action_relation_loss": 0.009876966569572687,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.743685687558466,
      "flow_matching_loss": 0.05522089898586273,
      "grad_norm": 0.15927527844905853,
      "h_stock_valid_norm": 96.2902000427246,
      "invalid_tail_samples": 23.1,
      "learning_rate": 9.360391315819917e-05,
      "loss": 0.0651,
      "positive_teacher_pairs": 27009.9,
      "query_h_norm_ratio": 0.46070760786533355,
      "relation_loss": 0.009876966569572687,
      "selected_pairs": 27009.9,
      "step": 6140,
      "student_distance_mean": 29.79252185821533,
      "teacher_distance_mean": 0.8613130033016205,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.009876966569572687,
      "weighted_relation_loss": 0.009876966569572687
    },
    {
      "action_query_batch_std": 0.4552599757909775,
      "action_query_norm": 44.36059036254883,
      "action_relation_loss": 0.009962904453277587,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.7530402245088865,
      "flow_matching_loss": 0.05656501576304436,
      "grad_norm": 0.16517892479896545,
      "h_stock_valid_norm": 95.77883987426758,
      "invalid_tail_samples": 22.7,
      "learning_rate": 9.357691491816269e-05,
      "loss": 0.0665,
      "positive_teacher_pairs": 27107.1,
      "query_h_norm_ratio": 0.46316383481025697,
      "relation_loss": 0.009962904453277587,
      "selected_pairs": 27107.1,
      "step": 6150,
      "student_distance_mean": 29.92039222717285,
      "teacher_distance_mean": 0.8570758700370789,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.009962904453277587,
      "weighted_relation_loss": 0.009962904453277587
    },
    {
      "action_query_batch_std": 0.46674723029136655,
      "action_query_norm": 44.371482849121094,
      "action_relation_loss": 0.009646778926253319,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.762394761459308,
      "flow_matching_loss": 0.0553899846971035,
      "grad_norm": 0.1305738091468811,
      "h_stock_valid_norm": 96.7128662109375,
      "invalid_tail_samples": 26.8,
      "learning_rate": 9.354986372806557e-05,
      "loss": 0.065,
      "positive_teacher_pairs": 26159.3,
      "query_h_norm_ratio": 0.4588070511817932,
      "relation_loss": 0.009646778926253319,
      "selected_pairs": 26159.3,
      "step": 6160,
      "student_distance_mean": 30.61823101043701,
      "teacher_distance_mean": 0.855550330877304,
      "valid_action_samples": 229.2,
      "weighted_auxiliary_loss": 0.009646778926253319,
      "weighted_relation_loss": 0.009646778926253319
    },
    {
      "action_query_batch_std": 0.4544188857078552,
      "action_query_norm": 44.35261192321777,
      "action_relation_loss": 0.0099305909126997,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.771749298409729,
      "flow_matching_loss": 0.05475315451622009,
      "grad_norm": 0.14708614349365234,
      "h_stock_valid_norm": 96.87188110351562,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.352275962077752e-05,
      "loss": 0.0647,
      "positive_teacher_pairs": 26877.3,
      "query_h_norm_ratio": 0.45785175263881683,
      "relation_loss": 0.0099305909126997,
      "selected_pairs": 26877.3,
      "step": 6170,
      "student_distance_mean": 29.949305152893068,
      "teacher_distance_mean": 0.8516030073165893,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0099305909126997,
      "weighted_relation_loss": 0.0099305909126997
    },
    {
      "action_query_batch_std": 0.4628007560968399,
      "action_query_norm": 44.368323516845706,
      "action_relation_loss": 0.009629882127046584,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.78110383536015,
      "flow_matching_loss": 0.055851930379867555,
      "grad_norm": 0.14266280829906464,
      "h_stock_valid_norm": 96.46642379760742,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.349560262923262e-05,
      "loss": 0.0655,
      "positive_teacher_pairs": 26895.0,
      "query_h_norm_ratio": 0.4599380254745483,
      "relation_loss": 0.009629882127046584,
      "selected_pairs": 26895.0,
      "step": 6180,
      "student_distance_mean": 30.43447036743164,
      "teacher_distance_mean": 0.8568536281585694,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.009629882127046584,
      "weighted_relation_loss": 0.009629882127046584
    },
    {
      "action_query_batch_std": 0.45935666263103486,
      "action_query_norm": 44.359423828125,
      "action_relation_loss": 0.009423177596181631,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.790458372310571,
      "flow_matching_loss": 0.053613365069031714,
      "grad_norm": 0.1711619794368744,
      "h_stock_valid_norm": 96.89087982177735,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.346839278642913e-05,
      "loss": 0.063,
      "positive_teacher_pairs": 26651.4,
      "query_h_norm_ratio": 0.4578316658735275,
      "relation_loss": 0.009423177596181631,
      "selected_pairs": 26651.4,
      "step": 6190,
      "student_distance_mean": 30.352076530456543,
      "teacher_distance_mean": 0.8606139898300171,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.009423177596181631,
      "weighted_relation_loss": 0.009423177596181631
    },
    {
      "action_query_batch_std": 0.4507600039243698,
      "action_query_norm": 44.34857521057129,
      "action_relation_loss": 0.009686880186200142,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.799812909260991,
      "flow_matching_loss": 0.05463266670703888,
      "grad_norm": 0.12668104469776154,
      "h_stock_valid_norm": 97.38855667114258,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.344113012542963e-05,
      "loss": 0.0643,
      "positive_teacher_pairs": 26765.2,
      "query_h_norm_ratio": 0.4553804576396942,
      "relation_loss": 0.009686880186200142,
      "selected_pairs": 26765.2,
      "step": 6200,
      "student_distance_mean": 29.736144256591796,
      "teacher_distance_mean": 0.8588486850261688,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.009686880186200142,
      "weighted_relation_loss": 0.009686880186200142
    },
    {
      "action_query_batch_std": 0.4584068536758423,
      "action_query_norm": 44.35734825134277,
      "action_relation_loss": 0.009499807190150023,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.809167446211412,
      "flow_matching_loss": 0.05678979121148586,
      "grad_norm": 0.1554381400346756,
      "h_stock_valid_norm": 97.03462677001953,
      "invalid_tail_samples": 21.6,
      "learning_rate": 9.341381467936079e-05,
      "loss": 0.0663,
      "positive_teacher_pairs": 27359.0,
      "query_h_norm_ratio": 0.4571322500705719,
      "relation_loss": 0.009499807190150023,
      "selected_pairs": 27359.0,
      "step": 6210,
      "student_distance_mean": 30.223221206665038,
      "teacher_distance_mean": 0.8604921877384186,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.009499807190150023,
      "weighted_relation_loss": 0.009499807190150023
    },
    {
      "action_query_batch_std": 0.4511774033308029,
      "action_query_norm": 44.346034622192384,
      "action_relation_loss": 0.009463789686560631,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.818521983161833,
      "flow_matching_loss": 0.05397530943155289,
      "grad_norm": 0.1673876792192459,
      "h_stock_valid_norm": 96.97356948852538,
      "invalid_tail_samples": 26.0,
      "learning_rate": 9.338644648141346e-05,
      "loss": 0.0634,
      "positive_teacher_pairs": 26343.6,
      "query_h_norm_ratio": 0.4573042899370193,
      "relation_loss": 0.009463789686560631,
      "selected_pairs": 26343.6,
      "step": 6220,
      "student_distance_mean": 29.55519504547119,
      "teacher_distance_mean": 0.8498220920562745,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.009463789686560631,
      "weighted_relation_loss": 0.009463789686560631
    },
    {
      "action_query_batch_std": 0.4500803917646408,
      "action_query_norm": 44.34593391418457,
      "action_relation_loss": 0.00919077666476369,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.827876520112254,
      "flow_matching_loss": 0.055557917803525925,
      "grad_norm": 0.1505383402109146,
      "h_stock_valid_norm": 96.63172912597656,
      "invalid_tail_samples": 21.7,
      "learning_rate": 9.335902556484257e-05,
      "loss": 0.0647,
      "positive_teacher_pairs": 27335.0,
      "query_h_norm_ratio": 0.45891753733158114,
      "relation_loss": 0.00919077666476369,
      "selected_pairs": 27335.0,
      "step": 6230,
      "student_distance_mean": 29.644734382629395,
      "teacher_distance_mean": 0.8549777746200562,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.00919077666476369,
      "weighted_relation_loss": 0.00919077666476369
    },
    {
      "action_query_batch_std": 0.45008144378662107,
      "action_query_norm": 44.34001693725586,
      "action_relation_loss": 0.009313150122761727,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.8372310570626755,
      "flow_matching_loss": 0.055862539634108546,
      "grad_norm": 0.13303305208683014,
      "h_stock_valid_norm": 96.76777420043945,
      "invalid_tail_samples": 24.9,
      "learning_rate": 9.333155196296712e-05,
      "loss": 0.0652,
      "positive_teacher_pairs": 26590.5,
      "query_h_norm_ratio": 0.45821460485458376,
      "relation_loss": 0.009313150122761727,
      "selected_pairs": 26590.5,
      "step": 6240,
      "student_distance_mean": 29.790196800231932,
      "teacher_distance_mean": 0.8518109023571014,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.009313150122761727,
      "weighted_relation_loss": 0.009313150122761727
    },
    {
      "action_query_batch_std": 0.44999452531337736,
      "action_query_norm": 44.341860961914065,
      "action_relation_loss": 0.010151688382029534,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.846585594013097,
      "flow_matching_loss": 0.0570238970220089,
      "grad_norm": 0.16184470057487488,
      "h_stock_valid_norm": 96.02557601928712,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.330402570917016e-05,
      "loss": 0.0672,
      "positive_teacher_pairs": 27206.4,
      "query_h_norm_ratio": 0.46177425384521487,
      "relation_loss": 0.010151688382029534,
      "selected_pairs": 27206.4,
      "step": 6250,
      "student_distance_mean": 29.57438850402832,
      "teacher_distance_mean": 0.850264173746109,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.010151688382029534,
      "weighted_relation_loss": 0.010151688382029534
    },
    {
      "action_query_batch_std": 0.4524629235267639,
      "action_query_norm": 44.33946533203125,
      "action_relation_loss": 0.009770381730049849,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.855940130963518,
      "flow_matching_loss": 0.05652053467929363,
      "grad_norm": 0.15192380547523499,
      "h_stock_valid_norm": 96.30883178710937,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.327644683689865e-05,
      "loss": 0.0663,
      "positive_teacher_pairs": 26826.8,
      "query_h_norm_ratio": 0.46039644479751585,
      "relation_loss": 0.009770381730049849,
      "selected_pairs": 26826.8,
      "step": 6260,
      "student_distance_mean": 29.835947227478027,
      "teacher_distance_mean": 0.8625324726104736,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.009770381730049849,
      "weighted_relation_loss": 0.009770381730049849
    },
    {
      "action_query_batch_std": 0.455686217546463,
      "action_query_norm": 44.34403915405274,
      "action_relation_loss": 0.009540364146232605,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.865294667913938,
      "flow_matching_loss": 0.05425838902592659,
      "grad_norm": 0.1343858689069748,
      "h_stock_valid_norm": 97.89571609497071,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.324881537966354e-05,
      "loss": 0.0638,
      "positive_teacher_pairs": 26804.9,
      "query_h_norm_ratio": 0.45298130214214327,
      "relation_loss": 0.009540364146232605,
      "selected_pairs": 26804.9,
      "step": 6270,
      "student_distance_mean": 29.97092475891113,
      "teacher_distance_mean": 0.8572137832641602,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.009540364146232605,
      "weighted_relation_loss": 0.009540364146232605
    },
    {
      "action_query_batch_std": 0.46035686135292053,
      "action_query_norm": 44.3517261505127,
      "action_relation_loss": 0.009722617082297802,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.874649204864359,
      "flow_matching_loss": 0.05525132641196251,
      "grad_norm": 0.13445474207401276,
      "h_stock_valid_norm": 98.80554885864258,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.322113137103964e-05,
      "loss": 0.065,
      "positive_teacher_pairs": 26894.4,
      "query_h_norm_ratio": 0.4488810896873474,
      "relation_loss": 0.009722617082297802,
      "selected_pairs": 26894.4,
      "step": 6280,
      "student_distance_mean": 30.34520320892334,
      "teacher_distance_mean": 0.8599928200244904,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.009722617082297802,
      "weighted_relation_loss": 0.009722617082297802
    },
    {
      "action_query_batch_std": 0.4460755825042725,
      "action_query_norm": 44.32951889038086,
      "action_relation_loss": 0.009678256139159203,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.88400374181478,
      "flow_matching_loss": 0.05569230765104294,
      "grad_norm": 0.14761881530284882,
      "h_stock_valid_norm": 97.59016571044921,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.319339484466564e-05,
      "loss": 0.0654,
      "positive_teacher_pairs": 26984.7,
      "query_h_norm_ratio": 0.4542544811964035,
      "relation_loss": 0.009678256139159203,
      "selected_pairs": 26984.7,
      "step": 6290,
      "student_distance_mean": 29.44514102935791,
      "teacher_distance_mean": 0.8506622731685638,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.009678256139159203,
      "weighted_relation_loss": 0.009678256139159203
    },
    {
      "action_query_batch_std": 0.4438843995332718,
      "action_query_norm": 44.32933044433594,
      "action_relation_loss": 0.009599783550947904,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.893358278765201,
      "flow_matching_loss": 0.05377415753901005,
      "grad_norm": 0.1342644840478897,
      "h_stock_valid_norm": 97.24955291748047,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.316560583424403e-05,
      "loss": 0.0634,
      "positive_teacher_pairs": 26828.0,
      "query_h_norm_ratio": 0.45583259463310244,
      "relation_loss": 0.009599783550947904,
      "selected_pairs": 26828.0,
      "step": 6300,
      "student_distance_mean": 29.297483062744142,
      "teacher_distance_mean": 0.8473630011081695,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.009599783550947904,
      "weighted_relation_loss": 0.009599783550947904
    },
    {
      "action_query_batch_std": 0.44461953043937685,
      "action_query_norm": 44.32155075073242,
      "action_relation_loss": 0.009689613990485669,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.902712815715622,
      "flow_matching_loss": 0.054285848140716554,
      "grad_norm": 0.18048393726348877,
      "h_stock_valid_norm": 96.67662048339844,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.313776437354108e-05,
      "loss": 0.064,
      "positive_teacher_pairs": 27181.5,
      "query_h_norm_ratio": 0.45845342576503756,
      "relation_loss": 0.009689613990485669,
      "selected_pairs": 27181.5,
      "step": 6310,
      "student_distance_mean": 29.2223482131958,
      "teacher_distance_mean": 0.8457733213901519,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.009689613990485669,
      "weighted_relation_loss": 0.009689613990485669
    },
    {
      "action_query_batch_std": 0.4502940833568573,
      "action_query_norm": 44.33115119934082,
      "action_relation_loss": 0.009447744581848383,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.9120673526660426,
      "flow_matching_loss": 0.053387632593512535,
      "grad_norm": 0.16083021461963654,
      "h_stock_valid_norm": 96.51690444946288,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.31098704963868e-05,
      "loss": 0.0628,
      "positive_teacher_pairs": 26965.1,
      "query_h_norm_ratio": 0.45931562781333923,
      "relation_loss": 0.009447744581848383,
      "selected_pairs": 26965.1,
      "step": 6320,
      "student_distance_mean": 29.64188175201416,
      "teacher_distance_mean": 0.8556046307086944,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.009447744581848383,
      "weighted_relation_loss": 0.009447744581848383
    },
    {
      "action_query_batch_std": 0.4508246511220932,
      "action_query_norm": 44.331173706054685,
      "action_relation_loss": 0.009372196160256863,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.921421889616464,
      "flow_matching_loss": 0.05290091931819916,
      "grad_norm": 0.15913791954517365,
      "h_stock_valid_norm": 95.75844497680664,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.308192423667485e-05,
      "loss": 0.0623,
      "positive_teacher_pairs": 26893.4,
      "query_h_norm_ratio": 0.462956377863884,
      "relation_loss": 0.009372196160256863,
      "selected_pairs": 26893.4,
      "step": 6330,
      "student_distance_mean": 29.83780689239502,
      "teacher_distance_mean": 0.8560031950473785,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.009372196160256863,
      "weighted_relation_loss": 0.009372196160256863
    },
    {
      "action_query_batch_std": 0.44747276306152345,
      "action_query_norm": 44.32316627502441,
      "action_relation_loss": 0.009510185662657022,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.930776426566885,
      "flow_matching_loss": 0.055963240563869476,
      "grad_norm": 0.12600113451480865,
      "h_stock_valid_norm": 95.45037231445312,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.305392562836262e-05,
      "loss": 0.0655,
      "positive_teacher_pairs": 27224.4,
      "query_h_norm_ratio": 0.4643625020980835,
      "relation_loss": 0.009510185662657022,
      "selected_pairs": 27224.4,
      "step": 6340,
      "student_distance_mean": 29.513971900939943,
      "teacher_distance_mean": 0.8605177998542786,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.009510185662657022,
      "weighted_relation_loss": 0.009510185662657022
    },
    {
      "action_query_batch_std": 0.45120433568954466,
      "action_query_norm": 44.32526321411133,
      "action_relation_loss": 0.009296325873583556,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.940130963517306,
      "flow_matching_loss": 0.05464393943548203,
      "grad_norm": 0.19822031259536743,
      "h_stock_valid_norm": 96.69876022338867,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.3025874705471e-05,
      "loss": 0.0639,
      "positive_teacher_pairs": 26664.6,
      "query_h_norm_ratio": 0.45838811099529264,
      "relation_loss": 0.009296325873583556,
      "selected_pairs": 26664.6,
      "step": 6350,
      "student_distance_mean": 29.746771621704102,
      "teacher_distance_mean": 0.8572543323040008,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.009296325873583556,
      "weighted_relation_loss": 0.009296325873583556
    },
    {
      "action_query_batch_std": 0.4482727348804474,
      "action_query_norm": 44.31970977783203,
      "action_relation_loss": 0.009036409854888915,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.949485500467727,
      "flow_matching_loss": 0.05309858806431293,
      "grad_norm": 0.15559571981430054,
      "h_stock_valid_norm": 97.00207595825195,
      "invalid_tail_samples": 26.0,
      "learning_rate": 9.299777150208457e-05,
      "loss": 0.0621,
      "positive_teacher_pairs": 26341.9,
      "query_h_norm_ratio": 0.4569013833999634,
      "relation_loss": 0.009036409854888915,
      "selected_pairs": 26341.9,
      "step": 6360,
      "student_distance_mean": 29.47408046722412,
      "teacher_distance_mean": 0.8503810644149781,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.009036409854888915,
      "weighted_relation_loss": 0.009036409854888915
    },
    {
      "action_query_batch_std": 0.4520481705665588,
      "action_query_norm": 44.31479682922363,
      "action_relation_loss": 0.009346061386168003,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.958840037418148,
      "flow_matching_loss": 0.05544129870831967,
      "grad_norm": 0.1521129161119461,
      "h_stock_valid_norm": 97.88331680297851,
      "invalid_tail_samples": 21.2,
      "learning_rate": 9.296961605235133e-05,
      "loss": 0.0648,
      "positive_teacher_pairs": 27452.1,
      "query_h_norm_ratio": 0.45273537933826447,
      "relation_loss": 0.009346061386168003,
      "selected_pairs": 27452.1,
      "step": 6370,
      "student_distance_mean": 29.7933931350708,
      "teacher_distance_mean": 0.8527152895927429,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.009346061386168003,
      "weighted_relation_loss": 0.009346061386168003
    },
    {
      "action_query_batch_std": 0.4492044925689697,
      "action_query_norm": 44.31530838012695,
      "action_relation_loss": 0.009678217489272356,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.968194574368569,
      "flow_matching_loss": 0.05470740236341953,
      "grad_norm": 0.18281102180480957,
      "h_stock_valid_norm": 98.70950698852539,
      "invalid_tail_samples": 21.8,
      "learning_rate": 9.29414083904828e-05,
      "loss": 0.0644,
      "positive_teacher_pairs": 27312.7,
      "query_h_norm_ratio": 0.4489483952522278,
      "relation_loss": 0.009678217489272356,
      "selected_pairs": 27312.7,
      "step": 6380,
      "student_distance_mean": 29.675663566589357,
      "teacher_distance_mean": 0.8433324217796325,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.009678217489272356,
      "weighted_relation_loss": 0.009678217489272356
    },
    {
      "action_query_batch_std": 0.45179002583026884,
      "action_query_norm": 44.313046264648435,
      "action_relation_loss": 0.009034908656030894,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.977549111318989,
      "flow_matching_loss": 0.057301894202828406,
      "grad_norm": 0.1693127453327179,
      "h_stock_valid_norm": 99.10757446289062,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.291314855075397e-05,
      "loss": 0.0663,
      "positive_teacher_pairs": 26806.3,
      "query_h_norm_ratio": 0.44712608158588407,
      "relation_loss": 0.009034908656030894,
      "selected_pairs": 26806.3,
      "step": 6390,
      "student_distance_mean": 29.744729423522948,
      "teacher_distance_mean": 0.855410385131836,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.009034908656030894,
      "weighted_relation_loss": 0.009034908656030894
    },
    {
      "action_query_batch_std": 0.4599537789821625,
      "action_query_norm": 44.3301498413086,
      "action_relation_loss": 0.009421588387340307,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.9869036482694105,
      "flow_matching_loss": 0.056033405289053914,
      "grad_norm": 0.12881408631801605,
      "h_stock_valid_norm": 98.1771224975586,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.288483656750322e-05,
      "loss": 0.0655,
      "positive_teacher_pairs": 26798.4,
      "query_h_norm_ratio": 0.451537960767746,
      "relation_loss": 0.009421588387340307,
      "selected_pairs": 26798.4,
      "step": 6400,
      "student_distance_mean": 30.170096015930177,
      "teacher_distance_mean": 0.8618562161922455,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.009421588387340307,
      "weighted_relation_loss": 0.009421588387340307
    },
    {
      "action_query_batch_std": 0.4493618458509445,
      "action_query_norm": 44.31138305664062,
      "action_relation_loss": 0.009693940449506045,
      "auxiliary_loss_weight": 1.0,
      "epoch": 5.996258185219832,
      "flow_matching_loss": 0.05489300563931465,
      "grad_norm": 0.17674660682678223,
      "h_stock_valid_norm": 96.7187484741211,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.285647247513225e-05,
      "loss": 0.0646,
      "positive_teacher_pairs": 26619.8,
      "query_h_norm_ratio": 0.4581689566373825,
      "relation_loss": 0.009693940449506045,
      "selected_pairs": 26619.8,
      "step": 6410,
      "student_distance_mean": 29.551808738708495,
      "teacher_distance_mean": 0.8590123951435089,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.009693940449506045,
      "weighted_relation_loss": 0.009693940449506045
    },
    {
      "action_query_batch_std": 0.44089129865169524,
      "action_query_norm": 44.29500274658203,
      "action_relation_loss": 0.009345129504799843,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.005612722170253,
      "flow_matching_loss": 0.05356586165726185,
      "grad_norm": 0.1577019989490509,
      "h_stock_valid_norm": 95.7902000427246,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.282805630810614e-05,
      "loss": 0.0629,
      "positive_teacher_pairs": 23926.7,
      "query_h_norm_ratio": 0.46242385506629946,
      "relation_loss": 0.009345129504799843,
      "selected_pairs": 23926.7,
      "step": 6420,
      "student_distance_mean": 29.154207038879395,
      "teacher_distance_mean": 0.8479879677295685,
      "valid_action_samples": 212.7,
      "weighted_auxiliary_loss": 0.009345129504799843,
      "weighted_relation_loss": 0.009345129504799843
    },
    {
      "action_query_batch_std": 0.44068136215209963,
      "action_query_norm": 44.29694519042969,
      "action_relation_loss": 0.009865197539329528,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.014967259120674,
      "flow_matching_loss": 0.05441788174211979,
      "grad_norm": 0.16835245490074158,
      "h_stock_valid_norm": 96.28668670654297,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.279958810095317e-05,
      "loss": 0.0643,
      "positive_teacher_pairs": 26690.3,
      "query_h_norm_ratio": 0.4600584715604782,
      "relation_loss": 0.009865197539329528,
      "selected_pairs": 26690.3,
      "step": 6430,
      "student_distance_mean": 29.088146209716797,
      "teacher_distance_mean": 0.841916823387146,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.009865197539329528,
      "weighted_relation_loss": 0.009865197539329528
    },
    {
      "action_query_batch_std": 0.4454332321882248,
      "action_query_norm": 44.304602432250974,
      "action_relation_loss": 0.008969044545665383,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.024321796071095,
      "flow_matching_loss": 0.05352292135357857,
      "grad_norm": 0.17983143031597137,
      "h_stock_valid_norm": 97.38916854858398,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.277106788826494e-05,
      "loss": 0.0625,
      "positive_teacher_pairs": 26732.6,
      "query_h_norm_ratio": 0.4549351245164871,
      "relation_loss": 0.008969044545665383,
      "selected_pairs": 26732.6,
      "step": 6440,
      "student_distance_mean": 29.413277435302735,
      "teacher_distance_mean": 0.8517879605293274,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.008969044545665383,
      "weighted_relation_loss": 0.008969044545665383
    },
    {
      "action_query_batch_std": 0.44739130735397337,
      "action_query_norm": 44.31757278442383,
      "action_relation_loss": 0.009248221479356288,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.033676333021515,
      "flow_matching_loss": 0.05353090316057205,
      "grad_norm": 0.11858154088258743,
      "h_stock_valid_norm": 98.92691192626953,
      "invalid_tail_samples": 21.9,
      "learning_rate": 9.274249570469618e-05,
      "loss": 0.0628,
      "positive_teacher_pairs": 27291.4,
      "query_h_norm_ratio": 0.4479873329401016,
      "relation_loss": 0.009248221479356288,
      "selected_pairs": 27291.4,
      "step": 6450,
      "student_distance_mean": 29.54491271972656,
      "teacher_distance_mean": 0.8495012581348419,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.009248221479356288,
      "weighted_relation_loss": 0.009248221479356288
    },
    {
      "action_query_batch_std": 0.4490538626909256,
      "action_query_norm": 44.31227111816406,
      "action_relation_loss": 0.009153698943555355,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.043030869971936,
      "flow_matching_loss": 0.05626991391181946,
      "grad_norm": 0.17702873051166534,
      "h_stock_valid_norm": 98.77997970581055,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.271387158496476e-05,
      "loss": 0.0654,
      "positive_teacher_pairs": 27152.3,
      "query_h_norm_ratio": 0.4486000776290894,
      "relation_loss": 0.009153698943555355,
      "selected_pairs": 27152.3,
      "step": 6460,
      "student_distance_mean": 29.56557426452637,
      "teacher_distance_mean": 0.8569265902042389,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.009153698943555355,
      "weighted_relation_loss": 0.009153698943555355
    },
    {
      "action_query_batch_std": 0.4470068454742432,
      "action_query_norm": 44.30155029296875,
      "action_relation_loss": 0.009198356233537198,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.052385406922357,
      "flow_matching_loss": 0.053302351757884024,
      "grad_norm": 0.17696736752986908,
      "h_stock_valid_norm": 97.5861587524414,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.268519556385173e-05,
      "loss": 0.0625,
      "positive_teacher_pairs": 27086.6,
      "query_h_norm_ratio": 0.4539866387844086,
      "relation_loss": 0.009198356233537198,
      "selected_pairs": 27086.6,
      "step": 6470,
      "student_distance_mean": 29.301386260986327,
      "teacher_distance_mean": 0.8512375712394714,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.009198356233537198,
      "weighted_relation_loss": 0.009198356233537198
    },
    {
      "action_query_batch_std": 0.4417905569076538,
      "action_query_norm": 44.295884704589845,
      "action_relation_loss": 0.009125436935573817,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.061739943872778,
      "flow_matching_loss": 0.054224394634366034,
      "grad_norm": 0.12912380695343018,
      "h_stock_valid_norm": 96.60820388793945,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.265646767620113e-05,
      "loss": 0.0633,
      "positive_teacher_pairs": 27080.1,
      "query_h_norm_ratio": 0.45851689875125884,
      "relation_loss": 0.009125436935573817,
      "selected_pairs": 27080.1,
      "step": 6480,
      "student_distance_mean": 29.03719139099121,
      "teacher_distance_mean": 0.8514952301979065,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.009125436935573817,
      "weighted_relation_loss": 0.009125436935573817
    },
    {
      "action_query_batch_std": 0.44391669929027555,
      "action_query_norm": 44.30072288513183,
      "action_relation_loss": 0.009034386277198792,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.0710944808231995,
      "flow_matching_loss": 0.05337479822337628,
      "grad_norm": 0.18259704113006592,
      "h_stock_valid_norm": 96.25340118408204,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.262768795692006e-05,
      "loss": 0.0624,
      "positive_teacher_pairs": 26996.7,
      "query_h_norm_ratio": 0.46025561094284057,
      "relation_loss": 0.009034386277198792,
      "selected_pairs": 26996.7,
      "step": 6490,
      "student_distance_mean": 29.20440216064453,
      "teacher_distance_mean": 0.8598646581172943,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.009034386277198792,
      "weighted_relation_loss": 0.009034386277198792
    },
    {
      "action_query_batch_std": 0.44326700568199157,
      "action_query_norm": 44.302120208740234,
      "action_relation_loss": 0.009085557842627167,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.080449017773621,
      "flow_matching_loss": 0.053595545887947085,
      "grad_norm": 0.22618746757507324,
      "h_stock_valid_norm": 96.41011428833008,
      "invalid_tail_samples": 23.1,
      "learning_rate": 9.259885644097861e-05,
      "loss": 0.0627,
      "positive_teacher_pairs": 27008.8,
      "query_h_norm_ratio": 0.45952064394950864,
      "relation_loss": 0.009085557842627167,
      "selected_pairs": 27008.8,
      "step": 6500,
      "student_distance_mean": 29.37670307159424,
      "teacher_distance_mean": 0.8626942038536072,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.009085557842627167,
      "weighted_relation_loss": 0.009085557842627167
    },
    {
      "action_query_batch_std": 0.4386919528245926,
      "action_query_norm": 44.28744392395019,
      "action_relation_loss": 0.008821534737944604,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.089803554724041,
      "flow_matching_loss": 0.05388713106513023,
      "grad_norm": 0.22161643207073212,
      "h_stock_valid_norm": 96.87714157104492,
      "invalid_tail_samples": 25.2,
      "learning_rate": 9.256997316340976e-05,
      "loss": 0.0627,
      "positive_teacher_pairs": 26528.5,
      "query_h_norm_ratio": 0.45715562999248505,
      "relation_loss": 0.008821534737944604,
      "selected_pairs": 26528.5,
      "step": 6510,
      "student_distance_mean": 28.98535270690918,
      "teacher_distance_mean": 0.8552859127521515,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.008821534737944604,
      "weighted_relation_loss": 0.008821534737944604
    },
    {
      "action_query_batch_std": 0.433867883682251,
      "action_query_norm": 44.27996559143067,
      "action_relation_loss": 0.00876768296584487,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.099158091674462,
      "flow_matching_loss": 0.05201970748603344,
      "grad_norm": 0.13616028428077698,
      "h_stock_valid_norm": 98.13627624511719,
      "invalid_tail_samples": 25.3,
      "learning_rate": 9.254103815930943e-05,
      "loss": 0.0608,
      "positive_teacher_pairs": 26506.6,
      "query_h_norm_ratio": 0.45121638774871825,
      "relation_loss": 0.00876768296584487,
      "selected_pairs": 26506.6,
      "step": 6520,
      "student_distance_mean": 28.780602264404298,
      "teacher_distance_mean": 0.8472282290458679,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.00876768296584487,
      "weighted_relation_loss": 0.00876768296584487
    },
    {
      "action_query_batch_std": 0.444094580411911,
      "action_query_norm": 44.287496185302736,
      "action_relation_loss": 0.00869555319659412,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.108512628624883,
      "flow_matching_loss": 0.05167555771768093,
      "grad_norm": 0.13677889108657837,
      "h_stock_valid_norm": 98.22092132568359,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.251205146383636e-05,
      "loss": 0.0604,
      "positive_teacher_pairs": 26625.1,
      "query_h_norm_ratio": 0.4509012281894684,
      "relation_loss": 0.00869555319659412,
      "selected_pairs": 26625.1,
      "step": 6530,
      "student_distance_mean": 29.28966293334961,
      "teacher_distance_mean": 0.8602219820022583,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.00869555319659412,
      "weighted_relation_loss": 0.00869555319659412
    },
    {
      "action_query_batch_std": 0.4396430402994156,
      "action_query_norm": 44.28163566589355,
      "action_relation_loss": 0.00879757753573358,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.117867165575304,
      "flow_matching_loss": 0.053728736191987994,
      "grad_norm": 0.14784863591194153,
      "h_stock_valid_norm": 98.1902359008789,
      "invalid_tail_samples": 21.0,
      "learning_rate": 9.248301311221216e-05,
      "loss": 0.0625,
      "positive_teacher_pairs": 27504.8,
      "query_h_norm_ratio": 0.4509816110134125,
      "relation_loss": 0.00879757753573358,
      "selected_pairs": 27504.8,
      "step": 6540,
      "student_distance_mean": 28.989124870300294,
      "teacher_distance_mean": 0.8511854946613312,
      "valid_action_samples": 235.0,
      "weighted_auxiliary_loss": 0.00879757753573358,
      "weighted_relation_loss": 0.00879757753573358
    },
    {
      "action_query_batch_std": 0.44118720293045044,
      "action_query_norm": 44.289105606079104,
      "action_relation_loss": 0.008840575907379388,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.127221702525725,
      "flow_matching_loss": 0.05356688052415848,
      "grad_norm": 0.16541464626789093,
      "h_stock_valid_norm": 98.17381057739257,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.245392313972115e-05,
      "loss": 0.0624,
      "positive_teacher_pairs": 26917.4,
      "query_h_norm_ratio": 0.45113309323787687,
      "relation_loss": 0.008840575907379388,
      "selected_pairs": 26917.4,
      "step": 6550,
      "student_distance_mean": 29.096364974975586,
      "teacher_distance_mean": 0.8512877702713013,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.008840575907379388,
      "weighted_relation_loss": 0.008840575907379388
    },
    {
      "action_query_batch_std": 0.4322665959596634,
      "action_query_norm": 44.27258110046387,
      "action_relation_loss": 0.009481490310281514,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.136576239476146,
      "flow_matching_loss": 0.0530180498957634,
      "grad_norm": 0.12730672955513,
      "h_stock_valid_norm": 97.4210220336914,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.242478158171037e-05,
      "loss": 0.0625,
      "positive_teacher_pairs": 26735.5,
      "query_h_norm_ratio": 0.4544536620378494,
      "relation_loss": 0.009481490310281514,
      "selected_pairs": 26735.5,
      "step": 6560,
      "student_distance_mean": 28.577195739746095,
      "teacher_distance_mean": 0.8459981083869934,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.009481490310281514,
      "weighted_relation_loss": 0.009481490310281514
    },
    {
      "action_query_batch_std": 0.44469837546348573,
      "action_query_norm": 44.284904861450194,
      "action_relation_loss": 0.008729384792968631,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.1459307764265665,
      "flow_matching_loss": 0.05441866554319859,
      "grad_norm": 0.13375592231750488,
      "h_stock_valid_norm": 96.91442184448242,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.239558847358957e-05,
      "loss": 0.0631,
      "positive_teacher_pairs": 26903.7,
      "query_h_norm_ratio": 0.45695269107818604,
      "relation_loss": 0.008729384792968631,
      "selected_pairs": 26903.7,
      "step": 6570,
      "student_distance_mean": 29.32769012451172,
      "teacher_distance_mean": 0.8596959233283996,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.008729384792968631,
      "weighted_relation_loss": 0.008729384792968631
    },
    {
      "action_query_batch_std": 0.437423437833786,
      "action_query_norm": 44.27326202392578,
      "action_relation_loss": 0.009077069256454707,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.155285313376988,
      "flow_matching_loss": 0.05291253812611103,
      "grad_norm": 0.16614572703838348,
      "h_stock_valid_norm": 97.68408432006837,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.236634385083114e-05,
      "loss": 0.062,
      "positive_teacher_pairs": 26816.6,
      "query_h_norm_ratio": 0.4532407283782959,
      "relation_loss": 0.009077069256454707,
      "selected_pairs": 26816.6,
      "step": 6580,
      "student_distance_mean": 28.82962760925293,
      "teacher_distance_mean": 0.849156504869461,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.009077069256454707,
      "weighted_relation_loss": 0.009077069256454707
    },
    {
      "action_query_batch_std": 0.449979704618454,
      "action_query_norm": 44.292816162109375,
      "action_relation_loss": 0.008825626596808434,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.164639850327409,
      "flow_matching_loss": 0.05407427698373794,
      "grad_norm": 0.16739726066589355,
      "h_stock_valid_norm": 97.385205078125,
      "invalid_tail_samples": 21.9,
      "learning_rate": 9.233704774897005e-05,
      "loss": 0.0629,
      "positive_teacher_pairs": 27292.0,
      "query_h_norm_ratio": 0.45482778549194336,
      "relation_loss": 0.008825626596808434,
      "selected_pairs": 27292.0,
      "step": 6590,
      "student_distance_mean": 29.640609359741212,
      "teacher_distance_mean": 0.8634037733078003,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.008825626596808434,
      "weighted_relation_loss": 0.008825626596808434
    },
    {
      "action_query_batch_std": 0.437526148557663,
      "action_query_norm": 44.265322875976565,
      "action_relation_loss": 0.008950219955295324,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.17399438727783,
      "flow_matching_loss": 0.052157673984766006,
      "grad_norm": 0.15318897366523743,
      "h_stock_valid_norm": 97.25172653198243,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.230770020360382e-05,
      "loss": 0.0611,
      "positive_teacher_pairs": 26785.9,
      "query_h_norm_ratio": 0.4551661044359207,
      "relation_loss": 0.008950219955295324,
      "selected_pairs": 26785.9,
      "step": 6600,
      "student_distance_mean": 28.857127571105956,
      "teacher_distance_mean": 0.8507076501846313,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.008950219955295324,
      "weighted_relation_loss": 0.008950219955295324
    },
    {
      "action_query_batch_std": 0.4372692883014679,
      "action_query_norm": 44.26692504882813,
      "action_relation_loss": 0.009595932066440582,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.183348924228251,
      "flow_matching_loss": 0.052270345762372016,
      "grad_norm": 0.11262659728527069,
      "h_stock_valid_norm": 97.66726837158203,
      "invalid_tail_samples": 22.7,
      "learning_rate": 9.227830125039247e-05,
      "loss": 0.0619,
      "positive_teacher_pairs": 27116.3,
      "query_h_norm_ratio": 0.45325559079647065,
      "relation_loss": 0.009595932066440582,
      "selected_pairs": 27116.3,
      "step": 6610,
      "student_distance_mean": 28.843610763549805,
      "teacher_distance_mean": 0.8506549656391144,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.009595932066440582,
      "weighted_relation_loss": 0.009595932066440582
    },
    {
      "action_query_batch_std": 0.4437212198972702,
      "action_query_norm": 44.27828483581543,
      "action_relation_loss": 0.008766455482691527,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.192703461178672,
      "flow_matching_loss": 0.05037235356867313,
      "grad_norm": 0.10410986095666885,
      "h_stock_valid_norm": 98.54184494018554,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.224885092505853e-05,
      "loss": 0.0591,
      "positive_teacher_pairs": 26898.4,
      "query_h_norm_ratio": 0.4493390381336212,
      "relation_loss": 0.008766455482691527,
      "selected_pairs": 26898.4,
      "step": 6620,
      "student_distance_mean": 29.415042304992674,
      "teacher_distance_mean": 0.849870240688324,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.008766455482691527,
      "weighted_relation_loss": 0.008766455482691527
    },
    {
      "action_query_batch_std": 0.4523605018854141,
      "action_query_norm": 44.29473533630371,
      "action_relation_loss": 0.008765799133107067,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.202057998129092,
      "flow_matching_loss": 0.05398422293365002,
      "grad_norm": 0.14703547954559326,
      "h_stock_valid_norm": 98.25149688720703,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.221934926338691e-05,
      "loss": 0.0628,
      "positive_teacher_pairs": 26701.0,
      "query_h_norm_ratio": 0.4508359581232071,
      "relation_loss": 0.008765799133107067,
      "selected_pairs": 26701.0,
      "step": 6630,
      "student_distance_mean": 30.028085899353027,
      "teacher_distance_mean": 0.8683351933956146,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.008765799133107067,
      "weighted_relation_loss": 0.008765799133107067
    },
    {
      "action_query_batch_std": 0.4404929608106613,
      "action_query_norm": 44.27612457275391,
      "action_relation_loss": 0.008688420709222554,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.211412535079513,
      "flow_matching_loss": 0.05261509604752064,
      "grad_norm": 0.13641957938671112,
      "h_stock_valid_norm": 96.97928390502929,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.21897963012249e-05,
      "loss": 0.0613,
      "positive_teacher_pairs": 26737.8,
      "query_h_norm_ratio": 0.4565568506717682,
      "relation_loss": 0.008688420709222554,
      "selected_pairs": 26737.8,
      "step": 6640,
      "student_distance_mean": 29.21709442138672,
      "teacher_distance_mean": 0.8559660732746124,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.008688420709222554,
      "weighted_relation_loss": 0.008688420709222554
    },
    {
      "action_query_batch_std": 0.4377728670835495,
      "action_query_norm": 44.265200805664065,
      "action_relation_loss": 0.009078176971524955,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.220767072029934,
      "flow_matching_loss": 0.054487187415361404,
      "grad_norm": 0.10950438678264618,
      "h_stock_valid_norm": 96.41681060791015,
      "invalid_tail_samples": 26.3,
      "learning_rate": 9.216019207448217e-05,
      "loss": 0.0636,
      "positive_teacher_pairs": 26277.3,
      "query_h_norm_ratio": 0.4591147989034653,
      "relation_loss": 0.009078176971524955,
      "selected_pairs": 26277.3,
      "step": 6650,
      "student_distance_mean": 28.79334888458252,
      "teacher_distance_mean": 0.839885151386261,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.009078176971524955,
      "weighted_relation_loss": 0.009078176971524955
    },
    {
      "action_query_batch_std": 0.44253145158290863,
      "action_query_norm": 44.2754264831543,
      "action_relation_loss": 0.008538389671593904,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.2301216089803555,
      "flow_matching_loss": 0.05130433887243271,
      "grad_norm": 0.14694878458976746,
      "h_stock_valid_norm": 96.65998611450195,
      "invalid_tail_samples": 23.0,
      "learning_rate": 9.21305366191306e-05,
      "loss": 0.0598,
      "positive_teacher_pairs": 27033.8,
      "query_h_norm_ratio": 0.45805882215499877,
      "relation_loss": 0.008538389671593904,
      "selected_pairs": 27033.8,
      "step": 6660,
      "student_distance_mean": 29.150013732910157,
      "teacher_distance_mean": 0.8467274844646454,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.008538389671593904,
      "weighted_relation_loss": 0.008538389671593904
    },
    {
      "action_query_batch_std": 0.4434982091188431,
      "action_query_norm": 44.27036590576172,
      "action_relation_loss": 0.008971612714231015,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.239476145930777,
      "flow_matching_loss": 0.055218834057450296,
      "grad_norm": 0.19357255101203918,
      "h_stock_valid_norm": 97.3760383605957,
      "invalid_tail_samples": 25.0,
      "learning_rate": 9.21008299712044e-05,
      "loss": 0.0642,
      "positive_teacher_pairs": 26571.6,
      "query_h_norm_ratio": 0.45464242696762086,
      "relation_loss": 0.008971612714231015,
      "selected_pairs": 26571.6,
      "step": 6670,
      "student_distance_mean": 29.275905036926268,
      "teacher_distance_mean": 0.8563144922256469,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.008971612714231015,
      "weighted_relation_loss": 0.008971612714231015
    },
    {
      "action_query_batch_std": 0.4423085987567902,
      "action_query_norm": 44.27137451171875,
      "action_relation_loss": 0.008946001529693604,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.248830682881198,
      "flow_matching_loss": 0.05235964730381966,
      "grad_norm": 0.12655094265937805,
      "h_stock_valid_norm": 98.69392166137695,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.207107216679993e-05,
      "loss": 0.0613,
      "positive_teacher_pairs": 26779.7,
      "query_h_norm_ratio": 0.4485768526792526,
      "relation_loss": 0.008946001529693604,
      "selected_pairs": 26779.7,
      "step": 6680,
      "student_distance_mean": 29.192632484436036,
      "teacher_distance_mean": 0.84547917842865,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.008946001529693604,
      "weighted_relation_loss": 0.008946001529693604
    },
    {
      "action_query_batch_std": 0.442666631937027,
      "action_query_norm": 44.26778907775879,
      "action_relation_loss": 0.0087840071413666,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.258185219831619,
      "flow_matching_loss": 0.05324157550930977,
      "grad_norm": 0.12746559083461761,
      "h_stock_valid_norm": 98.80434951782226,
      "invalid_tail_samples": 25.7,
      "learning_rate": 9.204126324207574e-05,
      "loss": 0.062,
      "positive_teacher_pairs": 26410.5,
      "query_h_norm_ratio": 0.44803925454616544,
      "relation_loss": 0.0087840071413666,
      "selected_pairs": 26410.5,
      "step": 6690,
      "student_distance_mean": 29.312471008300783,
      "teacher_distance_mean": 0.8483158469200134,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0087840071413666,
      "weighted_relation_loss": 0.0087840071413666
    },
    {
      "action_query_batch_std": 0.4567137658596039,
      "action_query_norm": 44.28996658325195,
      "action_relation_loss": 0.008841872680932283,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.267539756782039,
      "flow_matching_loss": 0.05366474092006683,
      "grad_norm": 0.17215242981910706,
      "h_stock_valid_norm": 98.24026489257812,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.201140323325248e-05,
      "loss": 0.0625,
      "positive_teacher_pairs": 26851.0,
      "query_h_norm_ratio": 0.4508392572402954,
      "relation_loss": 0.008841872680932283,
      "selected_pairs": 26851.0,
      "step": 6700,
      "student_distance_mean": 30.151272583007813,
      "teacher_distance_mean": 0.8693130373954773,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.008841872680932283,
      "weighted_relation_loss": 0.008841872680932283
    },
    {
      "action_query_batch_std": 0.4498042553663254,
      "action_query_norm": 44.28241539001465,
      "action_relation_loss": 0.00906903725117445,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.27689429373246,
      "flow_matching_loss": 0.053287023305892946,
      "grad_norm": 0.2174975872039795,
      "h_stock_valid_norm": 97.41573562622071,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.198149217661287e-05,
      "loss": 0.0624,
      "positive_teacher_pairs": 26691.3,
      "query_h_norm_ratio": 0.45457992851734164,
      "relation_loss": 0.00906903725117445,
      "selected_pairs": 26691.3,
      "step": 6710,
      "student_distance_mean": 29.746109390258788,
      "teacher_distance_mean": 0.855083417892456,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.00906903725117445,
      "weighted_relation_loss": 0.00906903725117445
    },
    {
      "action_query_batch_std": 0.4431190103292465,
      "action_query_norm": 44.26788673400879,
      "action_relation_loss": 0.008859155513346196,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.286248830682881,
      "flow_matching_loss": 0.05270777009427548,
      "grad_norm": 0.1900254189968109,
      "h_stock_valid_norm": 96.68904647827148,
      "invalid_tail_samples": 22.7,
      "learning_rate": 9.195153010850167e-05,
      "loss": 0.0616,
      "positive_teacher_pairs": 27102.4,
      "query_h_norm_ratio": 0.45784045159816744,
      "relation_loss": 0.008859155513346196,
      "selected_pairs": 27102.4,
      "step": 6720,
      "student_distance_mean": 29.17789249420166,
      "teacher_distance_mean": 0.8574794054031372,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.008859155513346196,
      "weighted_relation_loss": 0.008859155513346196
    },
    {
      "action_query_batch_std": 0.4505601167678833,
      "action_query_norm": 44.277470397949216,
      "action_relation_loss": 0.00881953863427043,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.295603367633302,
      "flow_matching_loss": 0.05283698663115501,
      "grad_norm": 0.19523130357265472,
      "h_stock_valid_norm": 96.82881164550781,
      "invalid_tail_samples": 21.8,
      "learning_rate": 9.192151706532562e-05,
      "loss": 0.0617,
      "positive_teacher_pairs": 27313.7,
      "query_h_norm_ratio": 0.4572781980037689,
      "relation_loss": 0.00881953863427043,
      "selected_pairs": 27313.7,
      "step": 6730,
      "student_distance_mean": 29.785982131958008,
      "teacher_distance_mean": 0.8644701898097992,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.00881953863427043,
      "weighted_relation_loss": 0.00881953863427043
    },
    {
      "action_query_batch_std": 0.4481358706951141,
      "action_query_norm": 44.26958808898926,
      "action_relation_loss": 0.009029234759509563,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.304957904583723,
      "flow_matching_loss": 0.05127544105052948,
      "grad_norm": 0.20114199817180634,
      "h_stock_valid_norm": 97.19377746582032,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.189145308355337e-05,
      "loss": 0.0603,
      "positive_teacher_pairs": 27173.3,
      "query_h_norm_ratio": 0.4554807782173157,
      "relation_loss": 0.009029234759509563,
      "selected_pairs": 27173.3,
      "step": 6740,
      "student_distance_mean": 29.460418701171875,
      "teacher_distance_mean": 0.8457156419754028,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.009029234759509563,
      "weighted_relation_loss": 0.009029234759509563
    },
    {
      "action_query_batch_std": 0.44253367483615874,
      "action_query_norm": 44.259057235717776,
      "action_relation_loss": 0.008533418597653507,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.314312441534144,
      "flow_matching_loss": 0.05204342044889927,
      "grad_norm": 0.13493317365646362,
      "h_stock_valid_norm": 97.90139770507812,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.186133819971555e-05,
      "loss": 0.0606,
      "positive_teacher_pairs": 26939.5,
      "query_h_norm_ratio": 0.4520833283662796,
      "relation_loss": 0.008533418597653507,
      "selected_pairs": 26939.5,
      "step": 6750,
      "student_distance_mean": 29.330468940734864,
      "teacher_distance_mean": 0.8550687611103058,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.008533418597653507,
      "weighted_relation_loss": 0.008533418597653507
    },
    {
      "action_query_batch_std": 0.437457463145256,
      "action_query_norm": 44.25081253051758,
      "action_relation_loss": 0.008566580619663,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.323666978484565,
      "flow_matching_loss": 0.05260408222675324,
      "grad_norm": 0.10973859578371048,
      "h_stock_valid_norm": 98.91091461181641,
      "invalid_tail_samples": 24.4,
      "learning_rate": 9.183117245040455e-05,
      "loss": 0.0612,
      "positive_teacher_pairs": 26715.1,
      "query_h_norm_ratio": 0.4473870784044266,
      "relation_loss": 0.008566580619663,
      "selected_pairs": 26715.1,
      "step": 6760,
      "student_distance_mean": 28.863204956054688,
      "teacher_distance_mean": 0.8513049781322479,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.008566580619663,
      "weighted_relation_loss": 0.008566580619663
    },
    {
      "action_query_batch_std": 0.4420220792293549,
      "action_query_norm": 44.26460571289063,
      "action_relation_loss": 0.009175017196685075,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.333021515434986,
      "flow_matching_loss": 0.05399457067251205,
      "grad_norm": 0.1257781982421875,
      "h_stock_valid_norm": 99.21550064086914,
      "invalid_tail_samples": 22.9,
      "learning_rate": 9.180095587227458e-05,
      "loss": 0.0632,
      "positive_teacher_pairs": 27057.8,
      "query_h_norm_ratio": 0.44615405797958374,
      "relation_loss": 0.009175017196685075,
      "selected_pairs": 27057.8,
      "step": 6770,
      "student_distance_mean": 29.22618179321289,
      "teacher_distance_mean": 0.8549359261989593,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.009175017196685075,
      "weighted_relation_loss": 0.009175017196685075
    },
    {
      "action_query_batch_std": 0.4444139778614044,
      "action_query_norm": 44.2572998046875,
      "action_relation_loss": 0.008748041838407517,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.342376052385407,
      "flow_matching_loss": 0.05450657680630684,
      "grad_norm": 0.15427684783935547,
      "h_stock_valid_norm": 97.86286087036133,
      "invalid_tail_samples": 25.7,
      "learning_rate": 9.177068850204167e-05,
      "loss": 0.0633,
      "positive_teacher_pairs": 26408.9,
      "query_h_norm_ratio": 0.4522469699382782,
      "relation_loss": 0.008748041838407517,
      "selected_pairs": 26408.9,
      "step": 6780,
      "student_distance_mean": 29.422729110717775,
      "teacher_distance_mean": 0.854955130815506,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.008748041838407517,
      "weighted_relation_loss": 0.008748041838407517
    },
    {
      "action_query_batch_std": 0.44526179134845734,
      "action_query_norm": 44.26539649963379,
      "action_relation_loss": 0.008601219905540348,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.351730589335828,
      "flow_matching_loss": 0.05159456729888916,
      "grad_norm": 0.12918411195278168,
      "h_stock_valid_norm": 98.37557830810547,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.174037037648351e-05,
      "loss": 0.0602,
      "positive_teacher_pairs": 26945.2,
      "query_h_norm_ratio": 0.4499727338552475,
      "relation_loss": 0.008601219905540348,
      "selected_pairs": 26945.2,
      "step": 6790,
      "student_distance_mean": 29.443947982788085,
      "teacher_distance_mean": 0.8580003499984741,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.008601219905540348,
      "weighted_relation_loss": 0.008601219905540348
    },
    {
      "action_query_batch_std": 0.4365649580955505,
      "action_query_norm": 44.24614295959473,
      "action_relation_loss": 0.008516621123999358,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.361085126286249,
      "flow_matching_loss": 0.05099165365099907,
      "grad_norm": 0.1425229161977768,
      "h_stock_valid_norm": 98.37546463012696,
      "invalid_tail_samples": 24.4,
      "learning_rate": 9.171000153243947e-05,
      "loss": 0.0595,
      "positive_teacher_pairs": 26706.7,
      "query_h_norm_ratio": 0.4497712105512619,
      "relation_loss": 0.008516621123999358,
      "selected_pairs": 26706.7,
      "step": 6800,
      "student_distance_mean": 28.877797317504882,
      "teacher_distance_mean": 0.8530425488948822,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.008516621123999358,
      "weighted_relation_loss": 0.008516621123999358
    },
    {
      "action_query_batch_std": 0.4386147290468216,
      "action_query_norm": 44.24618644714356,
      "action_relation_loss": 0.009356401488184929,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.37043966323667,
      "flow_matching_loss": 0.05436481647193432,
      "grad_norm": 0.14358802139759064,
      "h_stock_valid_norm": 97.16958999633789,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.167958200681057e-05,
      "loss": 0.0637,
      "positive_teacher_pairs": 26779.7,
      "query_h_norm_ratio": 0.4553598940372467,
      "relation_loss": 0.009356401488184929,
      "selected_pairs": 26779.7,
      "step": 6810,
      "student_distance_mean": 28.824372482299804,
      "teacher_distance_mean": 0.8522985935211181,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.009356401488184929,
      "weighted_relation_loss": 0.009356401488184929
    },
    {
      "action_query_batch_std": 0.44643549919128417,
      "action_query_norm": 44.24988136291504,
      "action_relation_loss": 0.00874393773265183,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.37979420018709,
      "flow_matching_loss": 0.05403813868761063,
      "grad_norm": 0.17585420608520508,
      "h_stock_valid_norm": 97.73366088867188,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.164911183655943e-05,
      "loss": 0.0628,
      "positive_teacher_pairs": 26875.8,
      "query_h_norm_ratio": 0.45277959406375884,
      "relation_loss": 0.00874393773265183,
      "selected_pairs": 26875.8,
      "step": 6820,
      "student_distance_mean": 29.650488662719727,
      "teacher_distance_mean": 0.8600635349750518,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.00874393773265183,
      "weighted_relation_loss": 0.00874393773265183
    },
    {
      "action_query_batch_std": 0.4475900262594223,
      "action_query_norm": 44.254309844970706,
      "action_relation_loss": 0.008735425351187587,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.3891487371375115,
      "flow_matching_loss": 0.05125780031085014,
      "grad_norm": 0.1574002206325531,
      "h_stock_valid_norm": 98.84162826538086,
      "invalid_tail_samples": 21.2,
      "learning_rate": 9.161859105871013e-05,
      "loss": 0.06,
      "positive_teacher_pairs": 27458.2,
      "query_h_norm_ratio": 0.44773334860801695,
      "relation_loss": 0.008735425351187587,
      "selected_pairs": 27458.2,
      "step": 6830,
      "student_distance_mean": 29.628078842163085,
      "teacher_distance_mean": 0.8695252418518067,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.008735425351187587,
      "weighted_relation_loss": 0.008735425351187587
    },
    {
      "action_query_batch_std": 0.4418612688779831,
      "action_query_norm": 44.24819984436035,
      "action_relation_loss": 0.008672963455319404,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.398503274087933,
      "flow_matching_loss": 0.051576196774840356,
      "grad_norm": 0.16251599788665771,
      "h_stock_valid_norm": 97.35794982910156,
      "invalid_tail_samples": 22.9,
      "learning_rate": 9.158801971034832e-05,
      "loss": 0.0602,
      "positive_teacher_pairs": 27055.6,
      "query_h_norm_ratio": 0.45449860394001007,
      "relation_loss": 0.008672963455319404,
      "selected_pairs": 27055.6,
      "step": 6840,
      "student_distance_mean": 29.289017105102538,
      "teacher_distance_mean": 0.8576358258724213,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.008672963455319404,
      "weighted_relation_loss": 0.008672963455319404
    },
    {
      "action_query_batch_std": 0.44874094128608705,
      "action_query_norm": 44.26075820922851,
      "action_relation_loss": 0.008744553383439779,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.407857811038354,
      "flow_matching_loss": 0.050693927332758904,
      "grad_norm": 0.11939522624015808,
      "h_stock_valid_norm": 96.58600006103515,
      "invalid_tail_samples": 25.1,
      "learning_rate": 9.155739782862107e-05,
      "loss": 0.0594,
      "positive_teacher_pairs": 26551.6,
      "query_h_norm_ratio": 0.45825617015361786,
      "relation_loss": 0.008744553383439779,
      "selected_pairs": 26551.6,
      "step": 6850,
      "student_distance_mean": 29.608242416381835,
      "teacher_distance_mean": 0.8486410975456238,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.008744553383439779,
      "weighted_relation_loss": 0.008744553383439779
    },
    {
      "action_query_batch_std": 0.44391530454158784,
      "action_query_norm": 44.23755950927735,
      "action_relation_loss": 0.00870600314810872,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.417212347988775,
      "flow_matching_loss": 0.05160349160432816,
      "grad_norm": 0.10990530252456665,
      "h_stock_valid_norm": 96.42538757324219,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.152672545073687e-05,
      "loss": 0.0603,
      "positive_teacher_pairs": 26784.5,
      "query_h_norm_ratio": 0.45877947509288786,
      "relation_loss": 0.00870600314810872,
      "selected_pairs": 26784.5,
      "step": 6860,
      "student_distance_mean": 29.30751132965088,
      "teacher_distance_mean": 0.8546960532665253,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.00870600314810872,
      "weighted_relation_loss": 0.00870600314810872
    },
    {
      "action_query_batch_std": 0.43649756014347074,
      "action_query_norm": 44.233263397216795,
      "action_relation_loss": 0.009348135534673929,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.426566884939196,
      "flow_matching_loss": 0.05384690575301647,
      "grad_norm": 0.14509646594524384,
      "h_stock_valid_norm": 96.52845916748046,
      "invalid_tail_samples": 25.1,
      "learning_rate": 9.149600261396551e-05,
      "loss": 0.0632,
      "positive_teacher_pairs": 26546.5,
      "query_h_norm_ratio": 0.4582437962293625,
      "relation_loss": 0.009348135534673929,
      "selected_pairs": 26546.5,
      "step": 6870,
      "student_distance_mean": 28.903252792358398,
      "teacher_distance_mean": 0.8486383199691773,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.009348135534673929,
      "weighted_relation_loss": 0.009348135534673929
    },
    {
      "action_query_batch_std": 0.44296315908432005,
      "action_query_norm": 44.24061508178711,
      "action_relation_loss": 0.009071934269741178,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.435921421889616,
      "flow_matching_loss": 0.049703953042626384,
      "grad_norm": 0.15774260461330414,
      "h_stock_valid_norm": 97.0033332824707,
      "invalid_tail_samples": 21.3,
      "learning_rate": 9.146522935563817e-05,
      "loss": 0.0588,
      "positive_teacher_pairs": 27434.6,
      "query_h_norm_ratio": 0.4560748666524887,
      "relation_loss": 0.009071934269741178,
      "selected_pairs": 27434.6,
      "step": 6880,
      "student_distance_mean": 29.131878280639647,
      "teacher_distance_mean": 0.8474416851997375,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.009071934269741178,
      "weighted_relation_loss": 0.009071934269741178
    },
    {
      "action_query_batch_std": 0.44106880724430086,
      "action_query_norm": 44.233177947998044,
      "action_relation_loss": 0.00842552799731493,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.445275958840037,
      "flow_matching_loss": 0.05093399360775948,
      "grad_norm": 0.0957622304558754,
      "h_stock_valid_norm": 97.28865356445313,
      "invalid_tail_samples": 24.6,
      "learning_rate": 9.143440571314723e-05,
      "loss": 0.0594,
      "positive_teacher_pairs": 26661.8,
      "query_h_norm_ratio": 0.45466333627700806,
      "relation_loss": 0.00842552799731493,
      "selected_pairs": 26661.8,
      "step": 6890,
      "student_distance_mean": 29.17599582672119,
      "teacher_distance_mean": 0.8564040064811707,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.00842552799731493,
      "weighted_relation_loss": 0.00842552799731493
    },
    {
      "action_query_batch_std": 0.44068956673145293,
      "action_query_norm": 44.23021545410156,
      "action_relation_loss": 0.008667147811502218,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.454630495790458,
      "flow_matching_loss": 0.04969879724085331,
      "grad_norm": 0.13952909409999847,
      "h_stock_valid_norm": 96.49378967285156,
      "invalid_tail_samples": 24.4,
      "learning_rate": 9.140353172394636e-05,
      "loss": 0.0584,
      "positive_teacher_pairs": 26711.1,
      "query_h_norm_ratio": 0.4583763301372528,
      "relation_loss": 0.008667147811502218,
      "selected_pairs": 26711.1,
      "step": 6900,
      "student_distance_mean": 29.060432052612306,
      "teacher_distance_mean": 0.8574673056602478,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.008667147811502218,
      "weighted_relation_loss": 0.008667147811502218
    },
    {
      "action_query_batch_std": 0.4408844292163849,
      "action_query_norm": 44.23374786376953,
      "action_relation_loss": 0.008682109322398901,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.463985032740879,
      "flow_matching_loss": 0.05292470864951611,
      "grad_norm": 0.1314915418624878,
      "h_stock_valid_norm": 96.4317413330078,
      "invalid_tail_samples": 22.0,
      "learning_rate": 9.137260742555032e-05,
      "loss": 0.0616,
      "positive_teacher_pairs": 27269.6,
      "query_h_norm_ratio": 0.45870902836322786,
      "relation_loss": 0.008682109322398901,
      "selected_pairs": 27269.6,
      "step": 6910,
      "student_distance_mean": 29.221406936645508,
      "teacher_distance_mean": 0.8487402915954589,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.008682109322398901,
      "weighted_relation_loss": 0.008682109322398901
    },
    {
      "action_query_batch_std": 0.44990605413913726,
      "action_query_norm": 44.24423408508301,
      "action_relation_loss": 0.008886068128049374,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.4733395696913005,
      "flow_matching_loss": 0.05383641719818115,
      "grad_norm": 0.11760301887989044,
      "h_stock_valid_norm": 95.6635238647461,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.13416328555351e-05,
      "loss": 0.0627,
      "positive_teacher_pairs": 26964.6,
      "query_h_norm_ratio": 0.4625041335821152,
      "relation_loss": 0.008886068128049374,
      "selected_pairs": 26964.6,
      "step": 6920,
      "student_distance_mean": 29.763829612731932,
      "teacher_distance_mean": 0.8639849543571472,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.008886068128049374,
      "weighted_relation_loss": 0.008886068128049374
    },
    {
      "action_query_batch_std": 0.4406514555215836,
      "action_query_norm": 44.22418212890625,
      "action_relation_loss": 0.008640496898442507,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.482694106641722,
      "flow_matching_loss": 0.05070470049977303,
      "grad_norm": 0.12292206287384033,
      "h_stock_valid_norm": 96.48811569213868,
      "invalid_tail_samples": 25.7,
      "learning_rate": 9.131060805153769e-05,
      "loss": 0.0593,
      "positive_teacher_pairs": 26418.6,
      "query_h_norm_ratio": 0.45835670828819275,
      "relation_loss": 0.008640496898442507,
      "selected_pairs": 26418.6,
      "step": 6930,
      "student_distance_mean": 29.04921703338623,
      "teacher_distance_mean": 0.851640272140503,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.008640496898442507,
      "weighted_relation_loss": 0.008640496898442507
    },
    {
      "action_query_batch_std": 0.4379824548959732,
      "action_query_norm": 44.22017707824707,
      "action_relation_loss": 0.008656568126752973,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.492048643592142,
      "flow_matching_loss": 0.05284300334751606,
      "grad_norm": 0.10742802917957306,
      "h_stock_valid_norm": 97.47661972045898,
      "invalid_tail_samples": 26.4,
      "learning_rate": 9.127953305125618e-05,
      "loss": 0.0615,
      "positive_teacher_pairs": 26249.2,
      "query_h_norm_ratio": 0.4536513119935989,
      "relation_loss": 0.008656568126752973,
      "selected_pairs": 26249.2,
      "step": 6940,
      "student_distance_mean": 28.967234992980956,
      "teacher_distance_mean": 0.8458339810371399,
      "valid_action_samples": 229.6,
      "weighted_auxiliary_loss": 0.008656568126752973,
      "weighted_relation_loss": 0.008656568126752973
    },
    {
      "action_query_batch_std": 0.4381188124418259,
      "action_query_norm": 44.22311515808106,
      "action_relation_loss": 0.008419681107625365,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.501403180542563,
      "flow_matching_loss": 0.052573902904987334,
      "grad_norm": 0.15436045825481415,
      "h_stock_valid_norm": 96.89059066772461,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.124840789244958e-05,
      "loss": 0.061,
      "positive_teacher_pairs": 26640.2,
      "query_h_norm_ratio": 0.45642639994621276,
      "relation_loss": 0.008419681107625365,
      "selected_pairs": 26640.2,
      "step": 6950,
      "student_distance_mean": 29.015458488464354,
      "teacher_distance_mean": 0.853215891122818,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.008419681107625365,
      "weighted_relation_loss": 0.008419681107625365
    },
    {
      "action_query_batch_std": 0.43925362527370454,
      "action_query_norm": 44.225911712646486,
      "action_relation_loss": 0.008859189227223396,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.510757717492984,
      "flow_matching_loss": 0.05216008424758911,
      "grad_norm": 0.10617944598197937,
      "h_stock_valid_norm": 96.04967498779297,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.121723261293794e-05,
      "loss": 0.061,
      "positive_teacher_pairs": 26802.6,
      "query_h_norm_ratio": 0.46045048236846925,
      "relation_loss": 0.008859189227223396,
      "selected_pairs": 26802.6,
      "step": 6960,
      "student_distance_mean": 29.042463493347167,
      "teacher_distance_mean": 0.854384309053421,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.008859189227223396,
      "weighted_relation_loss": 0.008859189227223396
    },
    {
      "action_query_batch_std": 0.4419611841440201,
      "action_query_norm": 44.23112182617187,
      "action_relation_loss": 0.0085248498711735,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.520112254443405,
      "flow_matching_loss": 0.049695099145174025,
      "grad_norm": 0.1221766322851181,
      "h_stock_valid_norm": 96.31312789916993,
      "invalid_tail_samples": 22.7,
      "learning_rate": 9.118600725060214e-05,
      "loss": 0.0582,
      "positive_teacher_pairs": 27102.4,
      "query_h_norm_ratio": 0.4592470794916153,
      "relation_loss": 0.0085248498711735,
      "selected_pairs": 27102.4,
      "step": 6970,
      "student_distance_mean": 29.3161376953125,
      "teacher_distance_mean": 0.8533631980419158,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0085248498711735,
      "weighted_relation_loss": 0.0085248498711735
    },
    {
      "action_query_batch_std": 0.44284839034080503,
      "action_query_norm": 44.227582931518555,
      "action_relation_loss": 0.008618235448375344,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.529466791393826,
      "flow_matching_loss": 0.05140262767672539,
      "grad_norm": 0.1583244651556015,
      "h_stock_valid_norm": 97.01598815917968,
      "invalid_tail_samples": 22.6,
      "learning_rate": 9.115473184338392e-05,
      "loss": 0.06,
      "positive_teacher_pairs": 27131.1,
      "query_h_norm_ratio": 0.4558833986520767,
      "relation_loss": 0.008618235448375344,
      "selected_pairs": 27131.1,
      "step": 6980,
      "student_distance_mean": 29.37299346923828,
      "teacher_distance_mean": 0.8609455466270447,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.008618235448375344,
      "weighted_relation_loss": 0.008618235448375344
    },
    {
      "action_query_batch_std": 0.44179455637931825,
      "action_query_norm": 44.22355003356934,
      "action_relation_loss": 0.008610341511666775,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.538821328344247,
      "flow_matching_loss": 0.05269462466239929,
      "grad_norm": 0.21702559292316437,
      "h_stock_valid_norm": 96.61185455322266,
      "invalid_tail_samples": 26.3,
      "learning_rate": 9.11234064292859e-05,
      "loss": 0.0613,
      "positive_teacher_pairs": 26278.2,
      "query_h_norm_ratio": 0.45775860249996186,
      "relation_loss": 0.008610341511666775,
      "selected_pairs": 26278.2,
      "step": 6990,
      "student_distance_mean": 29.278630256652832,
      "teacher_distance_mean": 0.8627460360527038,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.008610341511666775,
      "weighted_relation_loss": 0.008610341511666775
    },
    {
      "action_query_batch_std": 0.4395809382200241,
      "action_query_norm": 44.21658134460449,
      "action_relation_loss": 0.008364163944497705,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.5481758652946676,
      "flow_matching_loss": 0.05260836072266102,
      "grad_norm": 0.22020678222179413,
      "h_stock_valid_norm": 95.69695587158203,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.109203104637137e-05,
      "loss": 0.061,
      "positive_teacher_pairs": 26967.1,
      "query_h_norm_ratio": 0.46205319166183473,
      "relation_loss": 0.008364163944497705,
      "selected_pairs": 26967.1,
      "step": 7000,
      "student_distance_mean": 29.010591888427733,
      "teacher_distance_mean": 0.8577425003051757,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.008364163944497705,
      "weighted_relation_loss": 0.008364163944497705
    },
    {
      "action_query_batch_std": 0.4412956088781357,
      "action_query_norm": 44.21586265563965,
      "action_relation_loss": 0.009346392378211022,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.557530402245089,
      "flow_matching_loss": 0.05428835898637772,
      "grad_norm": 0.19826306402683258,
      "h_stock_valid_norm": 96.54583892822265,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.10606057327644e-05,
      "loss": 0.0636,
      "positive_teacher_pairs": 27085.2,
      "query_h_norm_ratio": 0.45798684656620026,
      "relation_loss": 0.009346392378211022,
      "selected_pairs": 27085.2,
      "step": 7010,
      "student_distance_mean": 29.219304847717286,
      "teacher_distance_mean": 0.853103518486023,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.009346392378211022,
      "weighted_relation_loss": 0.009346392378211022
    },
    {
      "action_query_batch_std": 0.43896146714687345,
      "action_query_norm": 44.20886993408203,
      "action_relation_loss": 0.008775789476931094,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.56688493919551,
      "flow_matching_loss": 0.05228128507733345,
      "grad_norm": 0.16231323778629303,
      "h_stock_valid_norm": 97.64037399291992,
      "invalid_tail_samples": 25.0,
      "learning_rate": 9.10291305266497e-05,
      "loss": 0.0611,
      "positive_teacher_pairs": 26570.0,
      "query_h_norm_ratio": 0.4527783960103989,
      "relation_loss": 0.008775789476931094,
      "selected_pairs": 26570.0,
      "step": 7020,
      "student_distance_mean": 29.081136322021486,
      "teacher_distance_mean": 0.8486098945140839,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.008775789476931094,
      "weighted_relation_loss": 0.008775789476931094
    },
    {
      "action_query_batch_std": 0.4395421415567398,
      "action_query_norm": 44.215359115600585,
      "action_relation_loss": 0.008694610558450222,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.576239476145931,
      "flow_matching_loss": 0.05470748357474804,
      "grad_norm": 0.2701050937175751,
      "h_stock_valid_norm": 98.29678955078126,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.099760546627261e-05,
      "loss": 0.0634,
      "positive_teacher_pairs": 26781.5,
      "query_h_norm_ratio": 0.4498197019100189,
      "relation_loss": 0.008694610558450222,
      "selected_pairs": 26781.5,
      "step": 7030,
      "student_distance_mean": 29.17637825012207,
      "teacher_distance_mean": 0.8653688728809357,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.008694610558450222,
      "weighted_relation_loss": 0.008694610558450222
    },
    {
      "action_query_batch_std": 0.4450228691101074,
      "action_query_norm": 44.222334671020505,
      "action_relation_loss": 0.008441106416285038,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.585594013096352,
      "flow_matching_loss": 0.05428258031606674,
      "grad_norm": 0.1824967861175537,
      "h_stock_valid_norm": 97.54067687988281,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.096603058993907e-05,
      "loss": 0.0627,
      "positive_teacher_pairs": 26627.0,
      "query_h_norm_ratio": 0.45337872505187987,
      "relation_loss": 0.008441106416285038,
      "selected_pairs": 26627.0,
      "step": 7040,
      "student_distance_mean": 29.327387046813964,
      "teacher_distance_mean": 0.8645540356636048,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.008441106416285038,
      "weighted_relation_loss": 0.008441106416285038
    },
    {
      "action_query_batch_std": 0.4403833210468292,
      "action_query_norm": 44.21575088500977,
      "action_relation_loss": 0.009113406017422675,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.594948550046773,
      "flow_matching_loss": 0.053664276376366615,
      "grad_norm": 0.16252760589122772,
      "h_stock_valid_norm": 97.50662384033203,
      "invalid_tail_samples": 25.4,
      "learning_rate": 9.093440593601552e-05,
      "loss": 0.0628,
      "positive_teacher_pairs": 26485.3,
      "query_h_norm_ratio": 0.45346611738204956,
      "relation_loss": 0.009113406017422675,
      "selected_pairs": 26485.3,
      "step": 7050,
      "student_distance_mean": 29.189105987548828,
      "teacher_distance_mean": 0.853467607498169,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.009113406017422675,
      "weighted_relation_loss": 0.009113406017422675
    },
    {
      "action_query_batch_std": 0.4422205984592438,
      "action_query_norm": 44.21607894897461,
      "action_relation_loss": 0.008913565194234252,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.604303086997193,
      "flow_matching_loss": 0.05050086528062821,
      "grad_norm": 0.12688355147838593,
      "h_stock_valid_norm": 96.48232879638672,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.090273154292889e-05,
      "loss": 0.0594,
      "positive_teacher_pairs": 26897.0,
      "query_h_norm_ratio": 0.4583066403865814,
      "relation_loss": 0.008913565194234252,
      "selected_pairs": 26897.0,
      "step": 7060,
      "student_distance_mean": 29.355212020874024,
      "teacher_distance_mean": 0.8547001838684082,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.008913565194234252,
      "weighted_relation_loss": 0.008913565194234252
    },
    {
      "action_query_batch_std": 0.43688441812992096,
      "action_query_norm": 44.20667991638184,
      "action_relation_loss": 0.00878028110601008,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.613657623947614,
      "flow_matching_loss": 0.05355775617063045,
      "grad_norm": 0.16619345545768738,
      "h_stock_valid_norm": 94.75764999389648,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.087100744916656e-05,
      "loss": 0.0623,
      "positive_teacher_pairs": 26925.3,
      "query_h_norm_ratio": 0.46652652323246,
      "relation_loss": 0.00878028110601008,
      "selected_pairs": 26925.3,
      "step": 7070,
      "student_distance_mean": 28.811701583862305,
      "teacher_distance_mean": 0.8494760572910309,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.00878028110601008,
      "weighted_relation_loss": 0.00878028110601008
    },
    {
      "action_query_batch_std": 0.44023153781890867,
      "action_query_norm": 44.21115837097168,
      "action_relation_loss": 0.00917202839627862,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.6230121608980355,
      "flow_matching_loss": 0.05324503444135189,
      "grad_norm": 0.11618804186582565,
      "h_stock_valid_norm": 95.53150787353516,
      "invalid_tail_samples": 20.8,
      "learning_rate": 9.08392336932763e-05,
      "loss": 0.0624,
      "positive_teacher_pairs": 27554.9,
      "query_h_norm_ratio": 0.46279831528663634,
      "relation_loss": 0.00917202839627862,
      "selected_pairs": 27554.9,
      "step": 7080,
      "student_distance_mean": 29.106574058532715,
      "teacher_distance_mean": 0.8470839262008667,
      "valid_action_samples": 235.2,
      "weighted_auxiliary_loss": 0.00917202839627862,
      "weighted_relation_loss": 0.00917202839627862
    },
    {
      "action_query_batch_std": 0.4477941423654556,
      "action_query_norm": 44.21496810913086,
      "action_relation_loss": 0.00874933172017336,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.632366697848457,
      "flow_matching_loss": 0.051841671019792555,
      "grad_norm": 0.1327243447303772,
      "h_stock_valid_norm": 96.36996612548828,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.080741031386619e-05,
      "loss": 0.0606,
      "positive_teacher_pairs": 26898.2,
      "query_h_norm_ratio": 0.45881599485874175,
      "relation_loss": 0.00874933172017336,
      "selected_pairs": 26898.2,
      "step": 7090,
      "student_distance_mean": 29.526216125488283,
      "teacher_distance_mean": 0.8530238330364227,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.00874933172017336,
      "weighted_relation_loss": 0.00874933172017336
    },
    {
      "action_query_batch_std": 0.4495520681142807,
      "action_query_norm": 44.22331619262695,
      "action_relation_loss": 0.008660761825740337,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.641721234798878,
      "flow_matching_loss": 0.052873161435127256,
      "grad_norm": 0.1492944210767746,
      "h_stock_valid_norm": 97.76702880859375,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.077553734960468e-05,
      "loss": 0.0615,
      "positive_teacher_pairs": 26696.2,
      "query_h_norm_ratio": 0.4523352414369583,
      "relation_loss": 0.008660761825740337,
      "selected_pairs": 26696.2,
      "step": 7100,
      "student_distance_mean": 29.719826889038085,
      "teacher_distance_mean": 0.8603653967380523,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.008660761825740337,
      "weighted_relation_loss": 0.008660761825740337
    },
    {
      "action_query_batch_std": 0.44041421711444856,
      "action_query_norm": 44.21024169921875,
      "action_relation_loss": 0.00867093582637608,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.651075771749299,
      "flow_matching_loss": 0.04823221787810326,
      "grad_norm": 0.1265983134508133,
      "h_stock_valid_norm": 97.56317672729492,
      "invalid_tail_samples": 22.5,
      "learning_rate": 9.074361483922041e-05,
      "loss": 0.0569,
      "positive_teacher_pairs": 27149.6,
      "query_h_norm_ratio": 0.4531489938497543,
      "relation_loss": 0.00867093582637608,
      "selected_pairs": 27149.6,
      "step": 7110,
      "student_distance_mean": 29.192882347106934,
      "teacher_distance_mean": 0.8527377426624299,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.00867093582637608,
      "weighted_relation_loss": 0.00867093582637608
    },
    {
      "action_query_batch_std": 0.4431263953447342,
      "action_query_norm": 44.214561462402344,
      "action_relation_loss": 0.009035386936739087,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.66043030869972,
      "flow_matching_loss": 0.05097108483314514,
      "grad_norm": 0.14957892894744873,
      "h_stock_valid_norm": 97.40048294067383,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.071164282150224e-05,
      "loss": 0.06,
      "positive_teacher_pairs": 26895.7,
      "query_h_norm_ratio": 0.4539535820484161,
      "relation_loss": 0.009035386936739087,
      "selected_pairs": 26895.7,
      "step": 7120,
      "student_distance_mean": 29.298222732543945,
      "teacher_distance_mean": 0.8515543282032013,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.009035386936739087,
      "weighted_relation_loss": 0.009035386936739087
    },
    {
      "action_query_batch_std": 0.4403069466352463,
      "action_query_norm": 44.19898300170898,
      "action_relation_loss": 0.008753289189189672,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.66978484565014,
      "flow_matching_loss": 0.05126370079815388,
      "grad_norm": 0.14292430877685547,
      "h_stock_valid_norm": 96.9168197631836,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.067962133529919e-05,
      "loss": 0.06,
      "positive_teacher_pairs": 26642.1,
      "query_h_norm_ratio": 0.4560529291629791,
      "relation_loss": 0.008753289189189672,
      "selected_pairs": 26642.1,
      "step": 7130,
      "student_distance_mean": 28.96078567504883,
      "teacher_distance_mean": 0.8549824714660644,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.008753289189189672,
      "weighted_relation_loss": 0.008753289189189672
    },
    {
      "action_query_batch_std": 0.4396529346704483,
      "action_query_norm": 44.19223518371582,
      "action_relation_loss": 0.008521308656781912,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.679139382600561,
      "flow_matching_loss": 0.05087776333093643,
      "grad_norm": 0.11764844506978989,
      "h_stock_valid_norm": 96.75809631347656,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.064755041952036e-05,
      "loss": 0.0594,
      "positive_teacher_pairs": 26922.8,
      "query_h_norm_ratio": 0.45673137307167055,
      "relation_loss": 0.008521308656781912,
      "selected_pairs": 26922.8,
      "step": 7140,
      "student_distance_mean": 28.914690017700195,
      "teacher_distance_mean": 0.8526542007923126,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.008521308656781912,
      "weighted_relation_loss": 0.008521308656781912
    },
    {
      "action_query_batch_std": 0.43898730874061587,
      "action_query_norm": 44.19542274475098,
      "action_relation_loss": 0.00867573292925954,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.688493919550982,
      "flow_matching_loss": 0.05173314288258553,
      "grad_norm": 0.17218901216983795,
      "h_stock_valid_norm": 97.31585922241212,
      "invalid_tail_samples": 25.8,
      "learning_rate": 9.061543011313497e-05,
      "loss": 0.0604,
      "positive_teacher_pairs": 26389.7,
      "query_h_norm_ratio": 0.4541501492261887,
      "relation_loss": 0.00867573292925954,
      "selected_pairs": 26389.7,
      "step": 7150,
      "student_distance_mean": 29.112705802917482,
      "teacher_distance_mean": 0.8460339367389679,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.00867573292925954,
      "weighted_relation_loss": 0.00867573292925954
    },
    {
      "action_query_batch_std": 0.444243860244751,
      "action_query_norm": 44.20910377502442,
      "action_relation_loss": 0.00855292221531272,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.697848456501403,
      "flow_matching_loss": 0.0520255871117115,
      "grad_norm": 0.13473181426525116,
      "h_stock_valid_norm": 97.74615478515625,
      "invalid_tail_samples": 21.9,
      "learning_rate": 9.05832604551722e-05,
      "loss": 0.0606,
      "positive_teacher_pairs": 27289.1,
      "query_h_norm_ratio": 0.4522894859313965,
      "relation_loss": 0.00855292221531272,
      "selected_pairs": 27289.1,
      "step": 7160,
      "student_distance_mean": 29.38282413482666,
      "teacher_distance_mean": 0.8550559639930725,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.00855292221531272,
      "weighted_relation_loss": 0.00855292221531272
    },
    {
      "action_query_batch_std": 0.44329107403755186,
      "action_query_norm": 44.193006896972655,
      "action_relation_loss": 0.008259504707530141,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.7072029934518245,
      "flow_matching_loss": 0.04987519755959511,
      "grad_norm": 0.10841154307126999,
      "h_stock_valid_norm": 98.34363250732422,
      "invalid_tail_samples": 24.4,
      "learning_rate": 9.055104148472124e-05,
      "loss": 0.0581,
      "positive_teacher_pairs": 26709.2,
      "query_h_norm_ratio": 0.44937609434127807,
      "relation_loss": 0.008259504707530141,
      "selected_pairs": 26709.2,
      "step": 7170,
      "student_distance_mean": 29.279901695251464,
      "teacher_distance_mean": 0.8607962787151336,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.008259504707530141,
      "weighted_relation_loss": 0.008259504707530141
    },
    {
      "action_query_batch_std": 0.4441526889801025,
      "action_query_norm": 44.19748191833496,
      "action_relation_loss": 0.008675164962187409,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.716557530402245,
      "flow_matching_loss": 0.05242014154791832,
      "grad_norm": 0.1670106053352356,
      "h_stock_valid_norm": 97.8732666015625,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.051877324093114e-05,
      "loss": 0.0611,
      "positive_teacher_pairs": 26963.6,
      "query_h_norm_ratio": 0.4515895128250122,
      "relation_loss": 0.008675164962187409,
      "selected_pairs": 26963.6,
      "step": 7180,
      "student_distance_mean": 29.483790969848634,
      "teacher_distance_mean": 0.8587196946144104,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.008675164962187409,
      "weighted_relation_loss": 0.008675164962187409
    },
    {
      "action_query_batch_std": 0.4440714240074158,
      "action_query_norm": 44.19906349182129,
      "action_relation_loss": 0.00871074958704412,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.725912067352666,
      "flow_matching_loss": 0.05016368143260479,
      "grad_norm": 0.19184184074401855,
      "h_stock_valid_norm": 97.51814727783203,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.04864557630109e-05,
      "loss": 0.0589,
      "positive_teacher_pairs": 26618.7,
      "query_h_norm_ratio": 0.4532456696033478,
      "relation_loss": 0.00871074958704412,
      "selected_pairs": 26618.7,
      "step": 7190,
      "student_distance_mean": 29.550149154663085,
      "teacher_distance_mean": 0.8560801565647125,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.00871074958704412,
      "weighted_relation_loss": 0.00871074958704412
    },
    {
      "action_query_batch_std": 0.4456844002008438,
      "action_query_norm": 44.20073890686035,
      "action_relation_loss": 0.008473641239106655,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.735266604303087,
      "flow_matching_loss": 0.055013862624764445,
      "grad_norm": 0.13642318546772003,
      "h_stock_valid_norm": 96.5126350402832,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.045408909022928e-05,
      "loss": 0.0635,
      "positive_teacher_pairs": 27004.5,
      "query_h_norm_ratio": 0.45798390209674833,
      "relation_loss": 0.008473641239106655,
      "selected_pairs": 27004.5,
      "step": 7200,
      "student_distance_mean": 29.26423759460449,
      "teacher_distance_mean": 0.8516120970249176,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.008473641239106655,
      "weighted_relation_loss": 0.008473641239106655
    },
    {
      "action_query_batch_std": 0.4442174553871155,
      "action_query_norm": 44.197356414794925,
      "action_relation_loss": 0.008599287364631891,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.744621141253508,
      "flow_matching_loss": 0.05219379439949989,
      "grad_norm": 0.23608790338039398,
      "h_stock_valid_norm": 97.8097640991211,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.042167326191484e-05,
      "loss": 0.0608,
      "positive_teacher_pairs": 26989.9,
      "query_h_norm_ratio": 0.4518829256296158,
      "relation_loss": 0.008599287364631891,
      "selected_pairs": 26989.9,
      "step": 7210,
      "student_distance_mean": 29.34976272583008,
      "teacher_distance_mean": 0.868370521068573,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.008599287364631891,
      "weighted_relation_loss": 0.008599287364631891
    },
    {
      "action_query_batch_std": 0.43649911880493164,
      "action_query_norm": 44.18864097595215,
      "action_relation_loss": 0.00841552740894258,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.753975678203929,
      "flow_matching_loss": 0.05279930718243122,
      "grad_norm": 0.19169799983501434,
      "h_stock_valid_norm": 98.54966583251954,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.038920831745587e-05,
      "loss": 0.0612,
      "positive_teacher_pairs": 27083.9,
      "query_h_norm_ratio": 0.4483931750059128,
      "relation_loss": 0.00841552740894258,
      "selected_pairs": 27083.9,
      "step": 7220,
      "student_distance_mean": 28.89986400604248,
      "teacher_distance_mean": 0.8585586786270142,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.00841552740894258,
      "weighted_relation_loss": 0.00841552740894258
    },
    {
      "action_query_batch_std": 0.44807552397251127,
      "action_query_norm": 44.20150299072266,
      "action_relation_loss": 0.008012371882796288,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.76333021515435,
      "flow_matching_loss": 0.052678913623094556,
      "grad_norm": 0.13545222580432892,
      "h_stock_valid_norm": 98.36515197753906,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.035669429630036e-05,
      "loss": 0.0607,
      "positive_teacher_pairs": 26840.5,
      "query_h_norm_ratio": 0.4493707627058029,
      "relation_loss": 0.008012371882796288,
      "selected_pairs": 26840.5,
      "step": 7230,
      "student_distance_mean": 29.62383270263672,
      "teacher_distance_mean": 0.8658746004104614,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.008012371882796288,
      "weighted_relation_loss": 0.008012371882796288
    },
    {
      "action_query_batch_std": 0.4349328398704529,
      "action_query_norm": 44.17747840881348,
      "action_relation_loss": 0.00860048308968544,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.772684752104771,
      "flow_matching_loss": 0.05262295939028263,
      "grad_norm": 0.14825735986232758,
      "h_stock_valid_norm": 96.98480377197265,
      "invalid_tail_samples": 25.4,
      "learning_rate": 9.032413123795588e-05,
      "loss": 0.0612,
      "positive_teacher_pairs": 26490.1,
      "query_h_norm_ratio": 0.45551320910453796,
      "relation_loss": 0.00860048308968544,
      "selected_pairs": 26490.1,
      "step": 7240,
      "student_distance_mean": 28.78893051147461,
      "teacher_distance_mean": 0.8435587108135223,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.00860048308968544,
      "weighted_relation_loss": 0.00860048308968544
    },
    {
      "action_query_batch_std": 0.4403835266828537,
      "action_query_norm": 44.18132972717285,
      "action_relation_loss": 0.00806795763783157,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.7820392890551915,
      "flow_matching_loss": 0.05012371204793453,
      "grad_norm": 0.15300075709819794,
      "h_stock_valid_norm": 96.44356384277344,
      "invalid_tail_samples": 25.5,
      "learning_rate": 9.029151918198962e-05,
      "loss": 0.0582,
      "positive_teacher_pairs": 26466.0,
      "query_h_norm_ratio": 0.4581089287996292,
      "relation_loss": 0.00806795763783157,
      "selected_pairs": 26466.0,
      "step": 7250,
      "student_distance_mean": 29.000444221496583,
      "teacher_distance_mean": 0.8555651009082794,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.00806795763783157,
      "weighted_relation_loss": 0.00806795763783157
    },
    {
      "action_query_batch_std": 0.44137921929359436,
      "action_query_norm": 44.17427635192871,
      "action_relation_loss": 0.008453379664570094,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.791393826005613,
      "flow_matching_loss": 0.050535840913653374,
      "grad_norm": 0.1934881955385208,
      "h_stock_valid_norm": 96.59046783447266,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.025885816802833e-05,
      "loss": 0.059,
      "positive_teacher_pairs": 26639.9,
      "query_h_norm_ratio": 0.4573396533727646,
      "relation_loss": 0.008453379664570094,
      "selected_pairs": 26639.9,
      "step": 7260,
      "student_distance_mean": 29.17438907623291,
      "teacher_distance_mean": 0.8731890916824341,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.008453379664570094,
      "weighted_relation_loss": 0.008453379664570094
    },
    {
      "action_query_batch_std": 0.4390820562839508,
      "action_query_norm": 44.17417678833008,
      "action_relation_loss": 0.008814779808744788,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.800748362956034,
      "flow_matching_loss": 0.0526904046535492,
      "grad_norm": 0.16231106221675873,
      "h_stock_valid_norm": 96.31715545654296,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.02261482357582e-05,
      "loss": 0.0615,
      "positive_teacher_pairs": 26826.3,
      "query_h_norm_ratio": 0.45863622725009917,
      "relation_loss": 0.008814779808744788,
      "selected_pairs": 26826.3,
      "step": 7270,
      "student_distance_mean": 29.0660249710083,
      "teacher_distance_mean": 0.8575051426887512,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.008814779808744788,
      "weighted_relation_loss": 0.008814779808744788
    },
    {
      "action_query_batch_std": 0.4353328347206116,
      "action_query_norm": 44.17005996704101,
      "action_relation_loss": 0.008461070572957396,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.810102899906455,
      "flow_matching_loss": 0.05105659291148186,
      "grad_norm": 0.15421125292778015,
      "h_stock_valid_norm": 95.79933395385743,
      "invalid_tail_samples": 25.6,
      "learning_rate": 9.019338942492485e-05,
      "loss": 0.0595,
      "positive_teacher_pairs": 26438.1,
      "query_h_norm_ratio": 0.46107260286808016,
      "relation_loss": 0.008461070572957396,
      "selected_pairs": 26438.1,
      "step": 7280,
      "student_distance_mean": 28.885638809204103,
      "teacher_distance_mean": 0.8554852068424225,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.008461070572957396,
      "weighted_relation_loss": 0.008461070572957396
    },
    {
      "action_query_batch_std": 0.4346194326877594,
      "action_query_norm": 44.16378669738769,
      "action_relation_loss": 0.008328199991956354,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.819457436856876,
      "flow_matching_loss": 0.05283939465880394,
      "grad_norm": 0.1565053015947342,
      "h_stock_valid_norm": 95.74167709350586,
      "invalid_tail_samples": 22.6,
      "learning_rate": 9.01605817753334e-05,
      "loss": 0.0612,
      "positive_teacher_pairs": 27134.6,
      "query_h_norm_ratio": 0.4612903445959091,
      "relation_loss": 0.008328199991956354,
      "selected_pairs": 27134.6,
      "step": 7290,
      "student_distance_mean": 28.8101167678833,
      "teacher_distance_mean": 0.8571621119976044,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.008328199991956354,
      "weighted_relation_loss": 0.008328199991956354
    },
    {
      "action_query_batch_std": 0.44094646275043486,
      "action_query_norm": 44.17249069213867,
      "action_relation_loss": 0.008185478765517474,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.828811973807296,
      "flow_matching_loss": 0.05349581092596054,
      "grad_norm": 0.13473045825958252,
      "h_stock_valid_norm": 97.49799575805665,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.012772532684818e-05,
      "loss": 0.0617,
      "positive_teacher_pairs": 26850.7,
      "query_h_norm_ratio": 0.4530664384365082,
      "relation_loss": 0.008185478765517474,
      "selected_pairs": 26850.7,
      "step": 7300,
      "student_distance_mean": 29.140887832641603,
      "teacher_distance_mean": 0.8628462553024292,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.008185478765517474,
      "weighted_relation_loss": 0.008185478765517474
    },
    {
      "action_query_batch_std": 0.43693803548812865,
      "action_query_norm": 44.15802574157715,
      "action_relation_loss": 0.008544413838535547,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.838166510757717,
      "flow_matching_loss": 0.05344742499291897,
      "grad_norm": 0.17297491431236267,
      "h_stock_valid_norm": 97.58992462158203,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.00948201193929e-05,
      "loss": 0.062,
      "positive_teacher_pairs": 26733.4,
      "query_h_norm_ratio": 0.4524871438741684,
      "relation_loss": 0.008544413838535547,
      "selected_pairs": 26733.4,
      "step": 7310,
      "student_distance_mean": 28.867044830322264,
      "teacher_distance_mean": 0.8543272972106933,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.008544413838535547,
      "weighted_relation_loss": 0.008544413838535547
    },
    {
      "action_query_batch_std": 0.4334615796804428,
      "action_query_norm": 44.156056594848636,
      "action_relation_loss": 0.008433082513511181,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.847521047708138,
      "flow_matching_loss": 0.05297682918608189,
      "grad_norm": 0.11472944170236588,
      "h_stock_valid_norm": 98.16139907836914,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.006186619295047e-05,
      "loss": 0.0614,
      "positive_teacher_pairs": 26828.8,
      "query_h_norm_ratio": 0.4498389035463333,
      "relation_loss": 0.008433082513511181,
      "selected_pairs": 26828.8,
      "step": 7320,
      "student_distance_mean": 28.74657917022705,
      "teacher_distance_mean": 0.8523162484169007,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.008433082513511181,
      "weighted_relation_loss": 0.008433082513511181
    },
    {
      "action_query_batch_std": 0.4288680851459503,
      "action_query_norm": 44.14013671875,
      "action_relation_loss": 0.008844668045639992,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.856875584658559,
      "flow_matching_loss": 0.052962854504585266,
      "grad_norm": 0.16362856328487396,
      "h_stock_valid_norm": 97.42459869384766,
      "invalid_tail_samples": 20.7,
      "learning_rate": 9.002886358756305e-05,
      "loss": 0.0618,
      "positive_teacher_pairs": 27571.3,
      "query_h_norm_ratio": 0.4530769675970078,
      "relation_loss": 0.008844668045639992,
      "selected_pairs": 27571.3,
      "step": 7330,
      "student_distance_mean": 28.37291259765625,
      "teacher_distance_mean": 0.851145476102829,
      "valid_action_samples": 235.3,
      "weighted_auxiliary_loss": 0.008844668045639992,
      "weighted_relation_loss": 0.008844668045639992
    },
    {
      "action_query_batch_std": 0.4362403929233551,
      "action_query_norm": 44.16250305175781,
      "action_relation_loss": 0.008726124465465546,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.8662301216089805,
      "flow_matching_loss": 0.05263004563748837,
      "grad_norm": 0.14750133454799652,
      "h_stock_valid_norm": 97.0621109008789,
      "invalid_tail_samples": 22.9,
      "learning_rate": 8.999581234333188e-05,
      "loss": 0.0614,
      "positive_teacher_pairs": 27056.8,
      "query_h_norm_ratio": 0.4549941152334213,
      "relation_loss": 0.008726124465465546,
      "selected_pairs": 27056.8,
      "step": 7340,
      "student_distance_mean": 28.818189811706542,
      "teacher_distance_mean": 0.8561428844928741,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.008726124465465546,
      "weighted_relation_loss": 0.008726124465465546
    },
    {
      "action_query_batch_std": 0.441314822435379,
      "action_query_norm": 44.1675106048584,
      "action_relation_loss": 0.00819530845619738,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.875584658559402,
      "flow_matching_loss": 0.05017702616751194,
      "grad_norm": 0.17104731500148773,
      "h_stock_valid_norm": 97.54132995605468,
      "invalid_tail_samples": 24.0,
      "learning_rate": 8.996271250041734e-05,
      "loss": 0.0584,
      "positive_teacher_pairs": 26802.9,
      "query_h_norm_ratio": 0.4528115808963776,
      "relation_loss": 0.00819530845619738,
      "selected_pairs": 26802.9,
      "step": 7350,
      "student_distance_mean": 29.18293113708496,
      "teacher_distance_mean": 0.8626627922058105,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.00819530845619738,
      "weighted_relation_loss": 0.00819530845619738
    },
    {
      "action_query_batch_std": 0.44293877184391023,
      "action_query_norm": 44.16696319580078,
      "action_relation_loss": 0.008987723011523485,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.884939195509823,
      "flow_matching_loss": 0.05273823514580726,
      "grad_norm": 0.1517804116010666,
      "h_stock_valid_norm": 97.48389739990235,
      "invalid_tail_samples": 20.6,
      "learning_rate": 8.99295640990389e-05,
      "loss": 0.0617,
      "positive_teacher_pairs": 27599.3,
      "query_h_norm_ratio": 0.453074249625206,
      "relation_loss": 0.008987723011523485,
      "selected_pairs": 27599.3,
      "step": 7360,
      "student_distance_mean": 29.26665515899658,
      "teacher_distance_mean": 0.8593522787094117,
      "valid_action_samples": 235.4,
      "weighted_auxiliary_loss": 0.008987723011523485,
      "weighted_relation_loss": 0.008987723011523485
    },
    {
      "action_query_batch_std": 0.4452000766992569,
      "action_query_norm": 44.169752502441405,
      "action_relation_loss": 0.00862079942598939,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.894293732460243,
      "flow_matching_loss": 0.05291054099798202,
      "grad_norm": 0.15155866742134094,
      "h_stock_valid_norm": 96.95364456176758,
      "invalid_tail_samples": 21.7,
      "learning_rate": 8.989636717947494e-05,
      "loss": 0.0615,
      "positive_teacher_pairs": 27337.1,
      "query_h_norm_ratio": 0.45557873547077177,
      "relation_loss": 0.00862079942598939,
      "selected_pairs": 27337.1,
      "step": 7370,
      "student_distance_mean": 29.45073356628418,
      "teacher_distance_mean": 0.8672034978866577,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.00862079942598939,
      "weighted_relation_loss": 0.00862079942598939
    },
    {
      "action_query_batch_std": 0.4309679478406906,
      "action_query_norm": 44.14190254211426,
      "action_relation_loss": 0.008460921375080943,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.903648269410664,
      "flow_matching_loss": 0.0503770686686039,
      "grad_norm": 0.1332094520330429,
      "h_stock_valid_norm": 97.87970733642578,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.986312178206291e-05,
      "loss": 0.0588,
      "positive_teacher_pairs": 26689.2,
      "query_h_norm_ratio": 0.45099141299724577,
      "relation_loss": 0.008460921375080943,
      "selected_pairs": 26689.2,
      "step": 7380,
      "student_distance_mean": 28.534205055236818,
      "teacher_distance_mean": 0.8453955292701721,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.008460921375080943,
      "weighted_relation_loss": 0.008460921375080943
    },
    {
      "action_query_batch_std": 0.4283880203962326,
      "action_query_norm": 44.137223434448245,
      "action_relation_loss": 0.008834674069657922,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.913002806361085,
      "flow_matching_loss": 0.051387951150536536,
      "grad_norm": 0.1561374068260193,
      "h_stock_valid_norm": 98.85432434082031,
      "invalid_tail_samples": 26.5,
      "learning_rate": 8.982982794719904e-05,
      "loss": 0.0602,
      "positive_teacher_pairs": 26234.2,
      "query_h_norm_ratio": 0.4464918851852417,
      "relation_loss": 0.008834674069657922,
      "selected_pairs": 26234.2,
      "step": 7390,
      "student_distance_mean": 28.428202819824218,
      "teacher_distance_mean": 0.8474992156028748,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.008834674069657922,
      "weighted_relation_loss": 0.008834674069657922
    },
    {
      "action_query_batch_std": 0.4379212588071823,
      "action_query_norm": 44.15916786193848,
      "action_relation_loss": 0.008182195015251637,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.922357343311506,
      "flow_matching_loss": 0.053545324876904485,
      "grad_norm": 0.1522766649723053,
      "h_stock_valid_norm": 98.70885925292968,
      "invalid_tail_samples": 21.4,
      "learning_rate": 8.979648571533852e-05,
      "loss": 0.0617,
      "positive_teacher_pairs": 27414.5,
      "query_h_norm_ratio": 0.4473766297101974,
      "relation_loss": 0.008182195015251637,
      "selected_pairs": 27414.5,
      "step": 7400,
      "student_distance_mean": 29.078770256042482,
      "teacher_distance_mean": 0.8621039688587189,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.008182195015251637,
      "weighted_relation_loss": 0.008182195015251637
    },
    {
      "action_query_batch_std": 0.4346734881401062,
      "action_query_norm": 44.144558334350585,
      "action_relation_loss": 0.008303504390642046,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.931711880261927,
      "flow_matching_loss": 0.049214692786335945,
      "grad_norm": 0.12121692299842834,
      "h_stock_valid_norm": 97.87622680664063,
      "invalid_tail_samples": 23.4,
      "learning_rate": 8.976309512699528e-05,
      "loss": 0.0575,
      "positive_teacher_pairs": 26938.9,
      "query_h_norm_ratio": 0.45103061497211455,
      "relation_loss": 0.008303504390642046,
      "selected_pairs": 26938.9,
      "step": 7410,
      "student_distance_mean": 28.862138938903808,
      "teacher_distance_mean": 0.8527838170528412,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.008303504390642046,
      "weighted_relation_loss": 0.008303504390642046
    },
    {
      "action_query_batch_std": 0.4386875331401825,
      "action_query_norm": 44.15615119934082,
      "action_relation_loss": 0.008813885506242514,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.9410664172123475,
      "flow_matching_loss": 0.05151863619685173,
      "grad_norm": 0.16541215777397156,
      "h_stock_valid_norm": 96.6830062866211,
      "invalid_tail_samples": 24.4,
      "learning_rate": 8.972965622274205e-05,
      "loss": 0.0603,
      "positive_teacher_pairs": 26712.5,
      "query_h_norm_ratio": 0.4567244440317154,
      "relation_loss": 0.008813885506242514,
      "selected_pairs": 26712.5,
      "step": 7420,
      "student_distance_mean": 29.186228942871093,
      "teacher_distance_mean": 0.8595913290977478,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.008813885506242514,
      "weighted_relation_loss": 0.008813885506242514
    },
    {
      "action_query_batch_std": 0.4329139232635498,
      "action_query_norm": 44.13747291564941,
      "action_relation_loss": 0.008153799548745155,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.950420954162769,
      "flow_matching_loss": 0.05032750628888607,
      "grad_norm": 0.14921218156814575,
      "h_stock_valid_norm": 96.7430046081543,
      "invalid_tail_samples": 23.9,
      "learning_rate": 8.969616904321026e-05,
      "loss": 0.0585,
      "positive_teacher_pairs": 26823.7,
      "query_h_norm_ratio": 0.45624141693115233,
      "relation_loss": 0.008153799548745155,
      "selected_pairs": 26823.7,
      "step": 7430,
      "student_distance_mean": 28.608967971801757,
      "teacher_distance_mean": 0.853182464838028,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.008153799548745155,
      "weighted_relation_loss": 0.008153799548745155
    },
    {
      "action_query_batch_std": 0.425823250412941,
      "action_query_norm": 44.13098373413086,
      "action_relation_loss": 0.009034632612019778,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.95977549111319,
      "flow_matching_loss": 0.04973046071827412,
      "grad_norm": 0.14008526504039764,
      "h_stock_valid_norm": 97.73678131103516,
      "invalid_tail_samples": 24.4,
      "learning_rate": 8.966263362908998e-05,
      "loss": 0.0588,
      "positive_teacher_pairs": 26712.4,
      "query_h_norm_ratio": 0.4515332967042923,
      "relation_loss": 0.009034632612019778,
      "selected_pairs": 26712.4,
      "step": 7440,
      "student_distance_mean": 28.12914409637451,
      "teacher_distance_mean": 0.8470549046993255,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.009034632612019778,
      "weighted_relation_loss": 0.009034632612019778
    },
    {
      "action_query_batch_std": 0.42350941598415376,
      "action_query_norm": 44.12957611083984,
      "action_relation_loss": 0.008516955515369773,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.969130028063611,
      "flow_matching_loss": 0.052967633679509164,
      "grad_norm": 0.1257512867450714,
      "h_stock_valid_norm": 97.49144744873047,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.962905002112989e-05,
      "loss": 0.0615,
      "positive_teacher_pairs": 26691.8,
      "query_h_norm_ratio": 0.4526558041572571,
      "relation_loss": 0.008516955515369773,
      "selected_pairs": 26691.8,
      "step": 7450,
      "student_distance_mean": 28.17362117767334,
      "teacher_distance_mean": 0.8445034801959992,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.008516955515369773,
      "weighted_relation_loss": 0.008516955515369773
    },
    {
      "action_query_batch_std": 0.4277544140815735,
      "action_query_norm": 44.130378341674806,
      "action_relation_loss": 0.008340831147506833,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.978484565014032,
      "flow_matching_loss": 0.05083308406174183,
      "grad_norm": 0.09968865662813187,
      "h_stock_valid_norm": 98.7834487915039,
      "invalid_tail_samples": 24.3,
      "learning_rate": 8.959541826013724e-05,
      "loss": 0.0592,
      "positive_teacher_pairs": 26746.0,
      "query_h_norm_ratio": 0.44674440622329714,
      "relation_loss": 0.008340831147506833,
      "selected_pairs": 26746.0,
      "step": 7460,
      "student_distance_mean": 28.240957069396973,
      "teacher_distance_mean": 0.8456895530223847,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.008340831147506833,
      "weighted_relation_loss": 0.008340831147506833
    },
    {
      "action_query_batch_std": 0.4330342262983322,
      "action_query_norm": 44.14130058288574,
      "action_relation_loss": 0.008127935184165835,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.987839101964453,
      "flow_matching_loss": 0.04884027577936649,
      "grad_norm": 0.14349417388439178,
      "h_stock_valid_norm": 100.1117950439453,
      "invalid_tail_samples": 23.6,
      "learning_rate": 8.956173838697781e-05,
      "loss": 0.057,
      "positive_teacher_pairs": 26899.5,
      "query_h_norm_ratio": 0.44092499315738676,
      "relation_loss": 0.008127935184165835,
      "selected_pairs": 26899.5,
      "step": 7470,
      "student_distance_mean": 28.78705539703369,
      "teacher_distance_mean": 0.857307106256485,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.008127935184165835,
      "weighted_relation_loss": 0.008127935184165835
    },
    {
      "action_query_batch_std": 0.4391139030456543,
      "action_query_norm": 44.14263381958008,
      "action_relation_loss": 0.00828099506907165,
      "auxiliary_loss_weight": 1.0,
      "epoch": 6.997193638914874,
      "flow_matching_loss": 0.052058383077383044,
      "grad_norm": 0.12701468169689178,
      "h_stock_valid_norm": 99.19424285888672,
      "invalid_tail_samples": 25.2,
      "learning_rate": 8.952801044257581e-05,
      "loss": 0.0603,
      "positive_teacher_pairs": 26529.9,
      "query_h_norm_ratio": 0.44501557350158694,
      "relation_loss": 0.00828099506907165,
      "selected_pairs": 26529.9,
      "step": 7480,
      "student_distance_mean": 29.124262619018555,
      "teacher_distance_mean": 0.8587952494621277,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.00828099506907165,
      "weighted_relation_loss": 0.00828099506907165
    },
    {
      "action_query_batch_std": 0.4347840517759323,
      "action_query_norm": 44.132637405395506,
      "action_relation_loss": 0.00841062916442752,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.006548175865294,
      "flow_matching_loss": 0.05232340507209301,
      "grad_norm": 0.14586171507835388,
      "h_stock_valid_norm": 97.83281021118164,
      "invalid_tail_samples": 20.9,
      "learning_rate": 8.949423446791387e-05,
      "loss": 0.0607,
      "positive_teacher_pairs": 24532.8,
      "query_h_norm_ratio": 0.45111219584941864,
      "relation_loss": 0.00841062916442752,
      "selected_pairs": 24532.8,
      "step": 7490,
      "student_distance_mean": 28.73592643737793,
      "teacher_distance_mean": 0.8570819735527039,
      "valid_action_samples": 215.2,
      "weighted_auxiliary_loss": 0.00841062916442752,
      "weighted_relation_loss": 0.00841062916442752
    },
    {
      "action_query_batch_std": 0.4367103695869446,
      "action_query_norm": 44.140320205688475,
      "action_relation_loss": 0.007799294730648399,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.015902712815715,
      "flow_matching_loss": 0.048636849224567416,
      "grad_norm": 0.12313555181026459,
      "h_stock_valid_norm": 96.99043731689453,
      "invalid_tail_samples": 24.4,
      "learning_rate": 8.9460410504033e-05,
      "loss": 0.0564,
      "positive_teacher_pairs": 26719.0,
      "query_h_norm_ratio": 0.4551053047180176,
      "relation_loss": 0.007799294730648399,
      "selected_pairs": 26719.0,
      "step": 7500,
      "student_distance_mean": 28.899776268005372,
      "teacher_distance_mean": 0.856358277797699,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.007799294730648399,
      "weighted_relation_loss": 0.007799294730648399
    },
    {
      "action_query_batch_std": 0.43847929537296293,
      "action_query_norm": 44.14625511169434,
      "action_relation_loss": 0.007876633573323489,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.0252572497661365,
      "flow_matching_loss": 0.04881029948592186,
      "grad_norm": 0.10531182587146759,
      "h_stock_valid_norm": 96.63299942016602,
      "invalid_tail_samples": 22.9,
      "learning_rate": 8.942653859203248e-05,
      "loss": 0.0567,
      "positive_teacher_pairs": 27063.8,
      "query_h_norm_ratio": 0.45685330033302307,
      "relation_loss": 0.007876633573323489,
      "selected_pairs": 27063.8,
      "step": 7510,
      "student_distance_mean": 29.137535858154298,
      "teacher_distance_mean": 0.8593722403049469,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.007876633573323489,
      "weighted_relation_loss": 0.007876633573323489
    },
    {
      "action_query_batch_std": 0.4332710027694702,
      "action_query_norm": 44.13542022705078,
      "action_relation_loss": 0.007905677473172546,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.034611786716558,
      "flow_matching_loss": 0.04868492409586907,
      "grad_norm": 0.1476520597934723,
      "h_stock_valid_norm": 96.37669448852539,
      "invalid_tail_samples": 20.9,
      "learning_rate": 8.93926187730699e-05,
      "loss": 0.0566,
      "positive_teacher_pairs": 27530.2,
      "query_h_norm_ratio": 0.4579510509967804,
      "relation_loss": 0.007905677473172546,
      "selected_pairs": 27530.2,
      "step": 7520,
      "student_distance_mean": 28.694157218933107,
      "teacher_distance_mean": 0.8550886690616608,
      "valid_action_samples": 235.1,
      "weighted_auxiliary_loss": 0.007905677473172546,
      "weighted_relation_loss": 0.007905677473172546
    },
    {
      "action_query_batch_std": 0.43009486198425295,
      "action_query_norm": 44.121988677978514,
      "action_relation_loss": 0.008043897850438953,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.043966323666979,
      "flow_matching_loss": 0.05148736871778965,
      "grad_norm": 0.15279723703861237,
      "h_stock_valid_norm": 96.7673095703125,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.935865108836102e-05,
      "loss": 0.0595,
      "positive_teacher_pairs": 26868.9,
      "query_h_norm_ratio": 0.45596094727516173,
      "relation_loss": 0.008043897850438953,
      "selected_pairs": 26868.9,
      "step": 7530,
      "student_distance_mean": 28.530482292175293,
      "teacher_distance_mean": 0.8577612221240998,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.008043897850438953,
      "weighted_relation_loss": 0.008043897850438953
    },
    {
      "action_query_batch_std": 0.4304381042718887,
      "action_query_norm": 44.12126579284668,
      "action_relation_loss": 0.008193028811365365,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.0533208606174,
      "flow_matching_loss": 0.05241321250796318,
      "grad_norm": 0.18198832869529724,
      "h_stock_valid_norm": 97.34071807861328,
      "invalid_tail_samples": 22.9,
      "learning_rate": 8.932463557917981e-05,
      "loss": 0.0606,
      "positive_teacher_pairs": 27058.6,
      "query_h_norm_ratio": 0.4532685250043869,
      "relation_loss": 0.008193028811365365,
      "selected_pairs": 27058.6,
      "step": 7540,
      "student_distance_mean": 28.47421760559082,
      "teacher_distance_mean": 0.8548420250415802,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.008193028811365365,
      "weighted_relation_loss": 0.008193028811365365
    },
    {
      "action_query_batch_std": 0.4332943707704544,
      "action_query_norm": 44.12953681945801,
      "action_relation_loss": 0.008075301116332412,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.06267539756782,
      "flow_matching_loss": 0.04947250746190548,
      "grad_norm": 0.1526663601398468,
      "h_stock_valid_norm": 97.2102424621582,
      "invalid_tail_samples": 26.8,
      "learning_rate": 8.929057228685829e-05,
      "loss": 0.0575,
      "positive_teacher_pairs": 26162.2,
      "query_h_norm_ratio": 0.45396162271499635,
      "relation_loss": 0.008075301116332412,
      "selected_pairs": 26162.2,
      "step": 7550,
      "student_distance_mean": 28.7947847366333,
      "teacher_distance_mean": 0.8587847054004669,
      "valid_action_samples": 229.2,
      "weighted_auxiliary_loss": 0.008075301116332412,
      "weighted_relation_loss": 0.008075301116332412
    },
    {
      "action_query_batch_std": 0.4335036039352417,
      "action_query_norm": 44.129570388793944,
      "action_relation_loss": 0.007644564937800169,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.072029934518241,
      "flow_matching_loss": 0.04691052176058293,
      "grad_norm": 0.21431320905685425,
      "h_stock_valid_norm": 96.7514862060547,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.925646125278658e-05,
      "loss": 0.0546,
      "positive_teacher_pairs": 26683.5,
      "query_h_norm_ratio": 0.4561160236597061,
      "relation_loss": 0.007644564937800169,
      "selected_pairs": 26683.5,
      "step": 7560,
      "student_distance_mean": 28.835444450378418,
      "teacher_distance_mean": 0.8527034282684326,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.007644564937800169,
      "weighted_relation_loss": 0.007644564937800169
    },
    {
      "action_query_batch_std": 0.4378224521875381,
      "action_query_norm": 44.13126831054687,
      "action_relation_loss": 0.007843216741457581,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.081384471468662,
      "flow_matching_loss": 0.05212226435542107,
      "grad_norm": 0.11764055490493774,
      "h_stock_valid_norm": 96.86578140258788,
      "invalid_tail_samples": 23.8,
      "learning_rate": 8.92223025184128e-05,
      "loss": 0.06,
      "positive_teacher_pairs": 26848.8,
      "query_h_norm_ratio": 0.45559405982494355,
      "relation_loss": 0.007843216741457581,
      "selected_pairs": 26848.8,
      "step": 7570,
      "student_distance_mean": 29.041106986999512,
      "teacher_distance_mean": 0.8588057518005371,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.007843216741457581,
      "weighted_relation_loss": 0.007843216741457581
    },
    {
      "action_query_batch_std": 0.438396018743515,
      "action_query_norm": 44.12265205383301,
      "action_relation_loss": 0.0076701796613633634,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.090739008419083,
      "flow_matching_loss": 0.04867291301488876,
      "grad_norm": 0.11004704982042313,
      "h_stock_valid_norm": 97.48420104980468,
      "invalid_tail_samples": 25.0,
      "learning_rate": 8.918809612524303e-05,
      "loss": 0.0563,
      "positive_teacher_pairs": 26575.9,
      "query_h_norm_ratio": 0.4526176810264587,
      "relation_loss": 0.0076701796613633634,
      "selected_pairs": 26575.9,
      "step": 7580,
      "student_distance_mean": 29.156683921813965,
      "teacher_distance_mean": 0.8566790401935578,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0076701796613633634,
      "weighted_relation_loss": 0.0076701796613633634
    },
    {
      "action_query_batch_std": 0.43607905209064485,
      "action_query_norm": 44.12245674133301,
      "action_relation_loss": 0.008135629910975694,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.100093545369504,
      "flow_matching_loss": 0.04865141585469246,
      "grad_norm": 0.14807899296283722,
      "h_stock_valid_norm": 97.63217468261719,
      "invalid_tail_samples": 25.7,
      "learning_rate": 8.91538421148413e-05,
      "loss": 0.0568,
      "positive_teacher_pairs": 26421.2,
      "query_h_norm_ratio": 0.4519317239522934,
      "relation_loss": 0.008135629910975694,
      "selected_pairs": 26421.2,
      "step": 7590,
      "student_distance_mean": 28.85023651123047,
      "teacher_distance_mean": 0.8504078447818756,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.008135629910975694,
      "weighted_relation_loss": 0.008135629910975694
    },
    {
      "action_query_batch_std": 0.4415286362171173,
      "action_query_norm": 44.13322830200195,
      "action_relation_loss": 0.008382557751610875,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.1094480823199255,
      "flow_matching_loss": 0.04912329725921154,
      "grad_norm": 0.09515794366598129,
      "h_stock_valid_norm": 97.5796142578125,
      "invalid_tail_samples": 24.0,
      "learning_rate": 8.91195405288294e-05,
      "loss": 0.0575,
      "positive_teacher_pairs": 26797.9,
      "query_h_norm_ratio": 0.4522823691368103,
      "relation_loss": 0.008382557751610875,
      "selected_pairs": 26797.9,
      "step": 7600,
      "student_distance_mean": 29.17276725769043,
      "teacher_distance_mean": 0.8709393799304962,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.008382557751610875,
      "weighted_relation_loss": 0.008382557751610875
    },
    {
      "action_query_batch_std": 0.4387216538190842,
      "action_query_norm": 44.127641677856445,
      "action_relation_loss": 0.008265334647148848,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.118802619270346,
      "flow_matching_loss": 0.05037583149969578,
      "grad_norm": 0.10193216800689697,
      "h_stock_valid_norm": 96.99755554199218,
      "invalid_tail_samples": 23.5,
      "learning_rate": 8.908519140888705e-05,
      "loss": 0.0586,
      "positive_teacher_pairs": 26929.9,
      "query_h_norm_ratio": 0.45493960082530976,
      "relation_loss": 0.008265334647148848,
      "selected_pairs": 26929.9,
      "step": 7610,
      "student_distance_mean": 29.081796073913573,
      "teacher_distance_mean": 0.8508171319961548,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.008265334647148848,
      "weighted_relation_loss": 0.008265334647148848
    },
    {
      "action_query_batch_std": 0.430875352025032,
      "action_query_norm": 44.11287879943848,
      "action_relation_loss": 0.007893224386498332,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.128157156220767,
      "flow_matching_loss": 0.04898877590894699,
      "grad_norm": 0.1381828784942627,
      "h_stock_valid_norm": 97.87546081542969,
      "invalid_tail_samples": 26.0,
      "learning_rate": 8.905079479675164e-05,
      "loss": 0.0569,
      "positive_teacher_pairs": 26339.0,
      "query_h_norm_ratio": 0.4507075667381287,
      "relation_loss": 0.007893224386498332,
      "selected_pairs": 26339.0,
      "step": 7620,
      "student_distance_mean": 28.601742362976076,
      "teacher_distance_mean": 0.8481608033180237,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.007893224386498332,
      "weighted_relation_loss": 0.007893224386498332
    },
    {
      "action_query_batch_std": 0.4337263762950897,
      "action_query_norm": 44.12110023498535,
      "action_relation_loss": 0.00803438234142959,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.137511693171188,
      "flow_matching_loss": 0.04751188531517982,
      "grad_norm": 0.15156756341457367,
      "h_stock_valid_norm": 97.62546081542969,
      "invalid_tail_samples": 22.8,
      "learning_rate": 8.901635073421831e-05,
      "loss": 0.0555,
      "positive_teacher_pairs": 27088.2,
      "query_h_norm_ratio": 0.45195107758045194,
      "relation_loss": 0.00803438234142959,
      "selected_pairs": 27088.2,
      "step": 7630,
      "student_distance_mean": 28.689328956604005,
      "teacher_distance_mean": 0.8598795354366302,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.00803438234142959,
      "weighted_relation_loss": 0.00803438234142959
    },
    {
      "action_query_batch_std": 0.43538332283496856,
      "action_query_norm": 44.1237075805664,
      "action_relation_loss": 0.007877619285136462,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.146866230121609,
      "flow_matching_loss": 0.04910760037600994,
      "grad_norm": 0.10990666598081589,
      "h_stock_valid_norm": 96.76364135742188,
      "invalid_tail_samples": 23.9,
      "learning_rate": 8.898185926313983e-05,
      "loss": 0.057,
      "positive_teacher_pairs": 26827.2,
      "query_h_norm_ratio": 0.4559985786676407,
      "relation_loss": 0.007877619285136462,
      "selected_pairs": 26827.2,
      "step": 7640,
      "student_distance_mean": 29.010419845581055,
      "teacher_distance_mean": 0.8600528359413147,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.007877619285136462,
      "weighted_relation_loss": 0.007877619285136462
    },
    {
      "action_query_batch_std": 0.4335306704044342,
      "action_query_norm": 44.11739730834961,
      "action_relation_loss": 0.007723823236301541,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.15622076707203,
      "flow_matching_loss": 0.0476244892925024,
      "grad_norm": 0.11697989702224731,
      "h_stock_valid_norm": 96.99431228637695,
      "invalid_tail_samples": 22.0,
      "learning_rate": 8.894732042542659e-05,
      "loss": 0.0553,
      "positive_teacher_pairs": 27271.9,
      "query_h_norm_ratio": 0.45485183894634246,
      "relation_loss": 0.007723823236301541,
      "selected_pairs": 27271.9,
      "step": 7650,
      "student_distance_mean": 28.638451385498048,
      "teacher_distance_mean": 0.8574818968772888,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.007723823236301541,
      "weighted_relation_loss": 0.007723823236301541
    },
    {
      "action_query_batch_std": 0.4323799580335617,
      "action_query_norm": 44.11108207702637,
      "action_relation_loss": 0.007887036725878715,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.165575304022451,
      "flow_matching_loss": 0.050549739971756935,
      "grad_norm": 0.12149971723556519,
      "h_stock_valid_norm": 97.66837844848632,
      "invalid_tail_samples": 23.4,
      "learning_rate": 8.891273426304654e-05,
      "loss": 0.0584,
      "positive_teacher_pairs": 26946.4,
      "query_h_norm_ratio": 0.4516490936279297,
      "relation_loss": 0.007887036725878715,
      "selected_pairs": 26946.4,
      "step": 7660,
      "student_distance_mean": 28.595350074768067,
      "teacher_distance_mean": 0.8527041554450989,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.007887036725878715,
      "weighted_relation_loss": 0.007887036725878715
    },
    {
      "action_query_batch_std": 0.4334577232599258,
      "action_query_norm": 44.11230697631836,
      "action_relation_loss": 0.007839092798531056,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.1749298409728715,
      "flow_matching_loss": 0.048204905912280084,
      "grad_norm": 0.15132136642932892,
      "h_stock_valid_norm": 96.80984573364258,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.887810081802515e-05,
      "loss": 0.056,
      "positive_teacher_pairs": 26694.4,
      "query_h_norm_ratio": 0.45566812455654143,
      "relation_loss": 0.007839092798531056,
      "selected_pairs": 26694.4,
      "step": 7670,
      "student_distance_mean": 28.775454902648924,
      "teacher_distance_mean": 0.8607877910137176,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.007839092798531056,
      "weighted_relation_loss": 0.007839092798531056
    },
    {
      "action_query_batch_std": 0.4253101259469986,
      "action_query_norm": 44.09513549804687,
      "action_relation_loss": 0.008021394396200776,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.1842843779232926,
      "flow_matching_loss": 0.04857560396194458,
      "grad_norm": 0.1629246473312378,
      "h_stock_valid_norm": 96.66489486694336,
      "invalid_tail_samples": 21.8,
      "learning_rate": 8.884342013244529e-05,
      "loss": 0.0566,
      "positive_teacher_pairs": 27314.0,
      "query_h_norm_ratio": 0.4561684012413025,
      "relation_loss": 0.008021394396200776,
      "selected_pairs": 27314.0,
      "step": 7680,
      "student_distance_mean": 28.223523330688476,
      "teacher_distance_mean": 0.8511075496673584,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.008021394396200776,
      "weighted_relation_loss": 0.008021394396200776
    },
    {
      "action_query_batch_std": 0.42864450216293337,
      "action_query_norm": 44.10158271789551,
      "action_relation_loss": 0.008002108801156283,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.193638914873714,
      "flow_matching_loss": 0.051811840012669565,
      "grad_norm": 0.17802739143371582,
      "h_stock_valid_norm": 96.7121452331543,
      "invalid_tail_samples": 23.6,
      "learning_rate": 8.880869224844725e-05,
      "loss": 0.0598,
      "positive_teacher_pairs": 26902.5,
      "query_h_norm_ratio": 0.45601141154766084,
      "relation_loss": 0.008002108801156283,
      "selected_pairs": 26902.5,
      "step": 7690,
      "student_distance_mean": 28.458311462402342,
      "teacher_distance_mean": 0.8519628703594208,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.008002108801156283,
      "weighted_relation_loss": 0.008002108801156283
    },
    {
      "action_query_batch_std": 0.4266175925731659,
      "action_query_norm": 44.085560989379886,
      "action_relation_loss": 0.008165821805596352,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.202993451824135,
      "flow_matching_loss": 0.049896720051765445,
      "grad_norm": 0.14638835191726685,
      "h_stock_valid_norm": 96.94295272827148,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.877391720822875e-05,
      "loss": 0.0581,
      "positive_teacher_pairs": 26876.9,
      "query_h_norm_ratio": 0.45476765036582945,
      "relation_loss": 0.008165821805596352,
      "selected_pairs": 26876.9,
      "step": 7700,
      "student_distance_mean": 28.237960815429688,
      "teacher_distance_mean": 0.8496497869491577,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.008165821805596352,
      "weighted_relation_loss": 0.008165821805596352
    },
    {
      "action_query_batch_std": 0.42876687049865725,
      "action_query_norm": 44.089155960083005,
      "action_relation_loss": 0.008080608118325472,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.212347988774556,
      "flow_matching_loss": 0.049904496595263484,
      "grad_norm": 0.1554315835237503,
      "h_stock_valid_norm": 98.09532470703125,
      "invalid_tail_samples": 23.8,
      "learning_rate": 8.873909505404468e-05,
      "loss": 0.058,
      "positive_teacher_pairs": 26846.2,
      "query_h_norm_ratio": 0.4494554340839386,
      "relation_loss": 0.008080608118325472,
      "selected_pairs": 26846.2,
      "step": 7710,
      "student_distance_mean": 28.336377143859863,
      "teacher_distance_mean": 0.8485854625701904,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.008080608118325472,
      "weighted_relation_loss": 0.008080608118325472
    },
    {
      "action_query_batch_std": 0.4311320424079895,
      "action_query_norm": 44.0950267791748,
      "action_relation_loss": 0.0077295079827308655,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.221702525724977,
      "flow_matching_loss": 0.04726284332573414,
      "grad_norm": 0.1318487972021103,
      "h_stock_valid_norm": 97.66045379638672,
      "invalid_tail_samples": 24.0,
      "learning_rate": 8.870422582820726e-05,
      "loss": 0.055,
      "positive_teacher_pairs": 26801.9,
      "query_h_norm_ratio": 0.45152067244052885,
      "relation_loss": 0.0077295079827308655,
      "selected_pairs": 26801.9,
      "step": 7720,
      "student_distance_mean": 28.47724552154541,
      "teacher_distance_mean": 0.8509435176849365,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0077295079827308655,
      "weighted_relation_loss": 0.0077295079827308655
    },
    {
      "action_query_batch_std": 0.43565390110015867,
      "action_query_norm": 44.099005126953124,
      "action_relation_loss": 0.00778433745726943,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.231057062675397,
      "flow_matching_loss": 0.0502527441829443,
      "grad_norm": 0.1837989091873169,
      "h_stock_valid_norm": 96.69836730957032,
      "invalid_tail_samples": 23.4,
      "learning_rate": 8.866930957308588e-05,
      "loss": 0.058,
      "positive_teacher_pairs": 26943.4,
      "query_h_norm_ratio": 0.4560542404651642,
      "relation_loss": 0.00778433745726943,
      "selected_pairs": 26943.4,
      "step": 7730,
      "student_distance_mean": 28.994941329956056,
      "teacher_distance_mean": 0.8618034958839417,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.00778433745726943,
      "weighted_relation_loss": 0.00778433745726943
    },
    {
      "action_query_batch_std": 0.4339591920375824,
      "action_query_norm": 44.09550743103027,
      "action_relation_loss": 0.007431101892143488,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.240411599625818,
      "flow_matching_loss": 0.04656772427260876,
      "grad_norm": 0.15951886773109436,
      "h_stock_valid_norm": 96.59734573364258,
      "invalid_tail_samples": 23.2,
      "learning_rate": 8.86343463311071e-05,
      "loss": 0.054,
      "positive_teacher_pairs": 26987.8,
      "query_h_norm_ratio": 0.45649510622024536,
      "relation_loss": 0.007431101892143488,
      "selected_pairs": 26987.8,
      "step": 7740,
      "student_distance_mean": 28.64826774597168,
      "teacher_distance_mean": 0.8617575526237488,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.007431101892143488,
      "weighted_relation_loss": 0.007431101892143488
    },
    {
      "action_query_batch_std": 0.4336211860179901,
      "action_query_norm": 44.10275764465332,
      "action_relation_loss": 0.008260813495144247,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.249766136576239,
      "flow_matching_loss": 0.0492363553494215,
      "grad_norm": 0.17548272013664246,
      "h_stock_valid_norm": 97.1036865234375,
      "invalid_tail_samples": 22.3,
      "learning_rate": 8.859933614475452e-05,
      "loss": 0.0575,
      "positive_teacher_pairs": 27200.3,
      "query_h_norm_ratio": 0.45419538021087646,
      "relation_loss": 0.008260813495144247,
      "selected_pairs": 27200.3,
      "step": 7750,
      "student_distance_mean": 28.630481338500978,
      "teacher_distance_mean": 0.8617521822452545,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.008260813495144247,
      "weighted_relation_loss": 0.008260813495144247
    },
    {
      "action_query_batch_std": 0.4251717567443848,
      "action_query_norm": 44.083744049072266,
      "action_relation_loss": 0.00796448439359665,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.2591206735266605,
      "flow_matching_loss": 0.049583807960152625,
      "grad_norm": 0.14716804027557373,
      "h_stock_valid_norm": 98.2431427001953,
      "invalid_tail_samples": 21.8,
      "learning_rate": 8.85642790565689e-05,
      "loss": 0.0575,
      "positive_teacher_pairs": 27312.9,
      "query_h_norm_ratio": 0.4487248778343201,
      "relation_loss": 0.00796448439359665,
      "selected_pairs": 27312.9,
      "step": 7760,
      "student_distance_mean": 28.28025074005127,
      "teacher_distance_mean": 0.8386616170406341,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.00796448439359665,
      "weighted_relation_loss": 0.00796448439359665
    },
    {
      "action_query_batch_std": 0.4180431395769119,
      "action_query_norm": 44.06364479064941,
      "action_relation_loss": 0.007836223999038338,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.268475210477082,
      "flow_matching_loss": 0.051042893156409266,
      "grad_norm": 0.12758398056030273,
      "h_stock_valid_norm": 97.77899398803712,
      "invalid_tail_samples": 20.8,
      "learning_rate": 8.852917510914782e-05,
      "loss": 0.0589,
      "positive_teacher_pairs": 27549.8,
      "query_h_norm_ratio": 0.4506488472223282,
      "relation_loss": 0.007836223999038338,
      "selected_pairs": 27549.8,
      "step": 7770,
      "student_distance_mean": 27.717968559265138,
      "teacher_distance_mean": 0.8384555280208588,
      "valid_action_samples": 235.2,
      "weighted_auxiliary_loss": 0.007836223999038338,
      "weighted_relation_loss": 0.007836223999038338
    },
    {
      "action_query_batch_std": 0.42721388041973113,
      "action_query_norm": 44.083465576171875,
      "action_relation_loss": 0.007939303340390325,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.277829747427503,
      "flow_matching_loss": 0.04993043802678585,
      "grad_norm": 0.12072066962718964,
      "h_stock_valid_norm": 96.7360824584961,
      "invalid_tail_samples": 22.7,
      "learning_rate": 8.84940243451459e-05,
      "loss": 0.0579,
      "positive_teacher_pairs": 27103.3,
      "query_h_norm_ratio": 0.45571265518665316,
      "relation_loss": 0.007939303340390325,
      "selected_pairs": 27103.3,
      "step": 7780,
      "student_distance_mean": 28.21179828643799,
      "teacher_distance_mean": 0.8538496315479278,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.007939303340390325,
      "weighted_relation_loss": 0.007939303340390325
    },
    {
      "action_query_batch_std": 0.4355536550283432,
      "action_query_norm": 44.092123413085936,
      "action_relation_loss": 0.008095852239057422,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.287184284377924,
      "flow_matching_loss": 0.05176020599901676,
      "grad_norm": 0.12196404486894608,
      "h_stock_valid_norm": 96.24624710083008,
      "invalid_tail_samples": 22.9,
      "learning_rate": 8.845882680727469e-05,
      "loss": 0.0599,
      "positive_teacher_pairs": 27064.9,
      "query_h_norm_ratio": 0.45812330543994906,
      "relation_loss": 0.008095852239057422,
      "selected_pairs": 27064.9,
      "step": 7790,
      "student_distance_mean": 28.941535186767577,
      "teacher_distance_mean": 0.8697809100151062,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.008095852239057422,
      "weighted_relation_loss": 0.008095852239057422
    },
    {
      "action_query_batch_std": 0.42345704436302184,
      "action_query_norm": 44.066292190551756,
      "action_relation_loss": 0.008024262217804789,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.296538821328344,
      "flow_matching_loss": 0.049895339831709865,
      "grad_norm": 0.15543580055236816,
      "h_stock_valid_norm": 96.6311752319336,
      "invalid_tail_samples": 24.6,
      "learning_rate": 8.842358253830246e-05,
      "loss": 0.0579,
      "positive_teacher_pairs": 26664.8,
      "query_h_norm_ratio": 0.4560320645570755,
      "relation_loss": 0.008024262217804789,
      "selected_pairs": 26664.8,
      "step": 7800,
      "student_distance_mean": 28.060860061645506,
      "teacher_distance_mean": 0.8543617367744446,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.008024262217804789,
      "weighted_relation_loss": 0.008024262217804789
    },
    {
      "action_query_batch_std": 0.42506236732006075,
      "action_query_norm": 44.064200973510744,
      "action_relation_loss": 0.008136632898822427,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.305893358278765,
      "flow_matching_loss": 0.05078279227018356,
      "grad_norm": 0.17196977138519287,
      "h_stock_valid_norm": 95.91879196166992,
      "invalid_tail_samples": 25.1,
      "learning_rate": 8.838829158105434e-05,
      "loss": 0.0589,
      "positive_teacher_pairs": 26546.3,
      "query_h_norm_ratio": 0.45939733982086184,
      "relation_loss": 0.008136632898822427,
      "selected_pairs": 26546.3,
      "step": 7810,
      "student_distance_mean": 28.147967720031737,
      "teacher_distance_mean": 0.8528271198272706,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.008136632898822427,
      "weighted_relation_loss": 0.008136632898822427
    },
    {
      "action_query_batch_std": 0.4285069555044174,
      "action_query_norm": 44.07096557617187,
      "action_relation_loss": 0.00774828651919961,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.315247895229186,
      "flow_matching_loss": 0.04735487923026085,
      "grad_norm": 0.14582014083862305,
      "h_stock_valid_norm": 97.73545684814454,
      "invalid_tail_samples": 25.0,
      "learning_rate": 8.835295397841217e-05,
      "loss": 0.0551,
      "positive_teacher_pairs": 26577.2,
      "query_h_norm_ratio": 0.45094879567623136,
      "relation_loss": 0.00774828651919961,
      "selected_pairs": 26577.2,
      "step": 7820,
      "student_distance_mean": 28.41762065887451,
      "teacher_distance_mean": 0.8550146818161011,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.00774828651919961,
      "weighted_relation_loss": 0.00774828651919961
    },
    {
      "action_query_batch_std": 0.43343760073184967,
      "action_query_norm": 44.08457374572754,
      "action_relation_loss": 0.007927614590153098,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.324602432179607,
      "flow_matching_loss": 0.04709854871034622,
      "grad_norm": 0.12104713916778564,
      "h_stock_valid_norm": 98.57700576782227,
      "invalid_tail_samples": 25.8,
      "learning_rate": 8.831756977331447e-05,
      "loss": 0.055,
      "positive_teacher_pairs": 26386.2,
      "query_h_norm_ratio": 0.44721212089061735,
      "relation_loss": 0.007927614590153098,
      "selected_pairs": 26386.2,
      "step": 7830,
      "student_distance_mean": 28.670578384399413,
      "teacher_distance_mean": 0.8520548164844512,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.007927614590153098,
      "weighted_relation_loss": 0.007927614590153098
    },
    {
      "action_query_batch_std": 0.43446349203586576,
      "action_query_norm": 44.07275466918945,
      "action_relation_loss": 0.007843274483457207,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.333956969130028,
      "flow_matching_loss": 0.05283229984343052,
      "grad_norm": 0.22083155810832977,
      "h_stock_valid_norm": 97.87069778442383,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.828213900875638e-05,
      "loss": 0.0607,
      "positive_teacher_pairs": 26689.0,
      "query_h_norm_ratio": 0.4503180295228958,
      "relation_loss": 0.007843274483457207,
      "selected_pairs": 26689.0,
      "step": 7840,
      "student_distance_mean": 28.791570091247557,
      "teacher_distance_mean": 0.8622426629066468,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.007843274483457207,
      "weighted_relation_loss": 0.007843274483457207
    },
    {
      "action_query_batch_std": 0.4301185667514801,
      "action_query_norm": 44.06818199157715,
      "action_relation_loss": 0.007944981334730983,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.343311506080449,
      "flow_matching_loss": 0.05002827458083629,
      "grad_norm": 0.1275738626718521,
      "h_stock_valid_norm": 97.58326034545898,
      "invalid_tail_samples": 24.2,
      "learning_rate": 8.824666172778964e-05,
      "loss": 0.058,
      "positive_teacher_pairs": 26765.9,
      "query_h_norm_ratio": 0.45160029232501986,
      "relation_loss": 0.007944981334730983,
      "selected_pairs": 26765.9,
      "step": 7850,
      "student_distance_mean": 28.406728744506836,
      "teacher_distance_mean": 0.8551561117172242,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.007944981334730983,
      "weighted_relation_loss": 0.007944981334730983
    },
    {
      "action_query_batch_std": 0.42871464490890504,
      "action_query_norm": 44.06652755737305,
      "action_relation_loss": 0.007883745385333896,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.35266604303087,
      "flow_matching_loss": 0.050895480811595915,
      "grad_norm": 0.13588717579841614,
      "h_stock_valid_norm": 98.04396362304688,
      "invalid_tail_samples": 23.8,
      "learning_rate": 8.821113797352245e-05,
      "loss": 0.0588,
      "positive_teacher_pairs": 26852.2,
      "query_h_norm_ratio": 0.44946223497390747,
      "relation_loss": 0.007883745385333896,
      "selected_pairs": 26852.2,
      "step": 7860,
      "student_distance_mean": 28.47870864868164,
      "teacher_distance_mean": 0.8552723348140716,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.007883745385333896,
      "weighted_relation_loss": 0.007883745385333896
    },
    {
      "action_query_batch_std": 0.4286540776491165,
      "action_query_norm": 44.0652515411377,
      "action_relation_loss": 0.008076582243666053,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.362020579981291,
      "flow_matching_loss": 0.04936905130743981,
      "grad_norm": 0.14688491821289062,
      "h_stock_valid_norm": 98.21607894897461,
      "invalid_tail_samples": 24.6,
      "learning_rate": 8.817556778911956e-05,
      "loss": 0.0574,
      "positive_teacher_pairs": 26671.0,
      "query_h_norm_ratio": 0.4486613094806671,
      "relation_loss": 0.008076582243666053,
      "selected_pairs": 26671.0,
      "step": 7870,
      "student_distance_mean": 28.4790828704834,
      "teacher_distance_mean": 0.8564915597438812,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.008076582243666053,
      "weighted_relation_loss": 0.008076582243666053
    },
    {
      "action_query_batch_std": 0.4258475065231323,
      "action_query_norm": 44.0680534362793,
      "action_relation_loss": 0.007465877896174788,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.371375116931712,
      "flow_matching_loss": 0.046393153443932536,
      "grad_norm": 0.15750126540660858,
      "h_stock_valid_norm": 99.05047607421875,
      "invalid_tail_samples": 24.2,
      "learning_rate": 8.81399512178021e-05,
      "loss": 0.0539,
      "positive_teacher_pairs": 26755.8,
      "query_h_norm_ratio": 0.44490925371646883,
      "relation_loss": 0.007465877896174788,
      "selected_pairs": 26755.8,
      "step": 7880,
      "student_distance_mean": 28.2560546875,
      "teacher_distance_mean": 0.8499222218990325,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.007465877896174788,
      "weighted_relation_loss": 0.007465877896174788
    },
    {
      "action_query_batch_std": 0.42902224957942964,
      "action_query_norm": 44.07440338134766,
      "action_relation_loss": 0.007921129371970892,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.380729653882133,
      "flow_matching_loss": 0.04749383181333542,
      "grad_norm": 0.13943913578987122,
      "h_stock_valid_norm": 98.46097793579102,
      "invalid_tail_samples": 25.3,
      "learning_rate": 8.810428830284752e-05,
      "loss": 0.0554,
      "positive_teacher_pairs": 26500.4,
      "query_h_norm_ratio": 0.4476432234048843,
      "relation_loss": 0.007921129371970892,
      "selected_pairs": 26500.4,
      "step": 7890,
      "student_distance_mean": 28.475897598266602,
      "teacher_distance_mean": 0.8555309116840363,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.007921129371970892,
      "weighted_relation_loss": 0.007921129371970892
    },
    {
      "action_query_batch_std": 0.4299160033464432,
      "action_query_norm": 44.07305335998535,
      "action_relation_loss": 0.00789542282000184,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.390084190832554,
      "flow_matching_loss": 0.04795755632221699,
      "grad_norm": 0.12066502869129181,
      "h_stock_valid_norm": 97.67577590942383,
      "invalid_tail_samples": 25.2,
      "learning_rate": 8.806857908758967e-05,
      "loss": 0.0559,
      "positive_teacher_pairs": 26536.6,
      "query_h_norm_ratio": 0.4512239247560501,
      "relation_loss": 0.00789542282000184,
      "selected_pairs": 26536.6,
      "step": 7900,
      "student_distance_mean": 28.418256187438963,
      "teacher_distance_mean": 0.8517780601978302,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.00789542282000184,
      "weighted_relation_loss": 0.00789542282000184
    },
    {
      "action_query_batch_std": 0.4276371330022812,
      "action_query_norm": 44.06443710327149,
      "action_relation_loss": 0.007599850697442889,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.399438727782975,
      "flow_matching_loss": 0.04773058705031872,
      "grad_norm": 0.11173338443040848,
      "h_stock_valid_norm": 97.62870025634766,
      "invalid_tail_samples": 25.1,
      "learning_rate": 8.80328236154186e-05,
      "loss": 0.0553,
      "positive_teacher_pairs": 26556.1,
      "query_h_norm_ratio": 0.4513497710227966,
      "relation_loss": 0.007599850697442889,
      "selected_pairs": 26556.1,
      "step": 7910,
      "student_distance_mean": 28.483184814453125,
      "teacher_distance_mean": 0.8528826892375946,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.007599850697442889,
      "weighted_relation_loss": 0.007599850697442889
    },
    {
      "action_query_batch_std": 0.42670516669750214,
      "action_query_norm": 44.055580139160156,
      "action_relation_loss": 0.00779494047164917,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.408793264733395,
      "flow_matching_loss": 0.048922672122716906,
      "grad_norm": 0.12149310857057571,
      "h_stock_valid_norm": 97.73736190795898,
      "invalid_tail_samples": 27.6,
      "learning_rate": 8.799702192978056e-05,
      "loss": 0.0567,
      "positive_teacher_pairs": 25996.9,
      "query_h_norm_ratio": 0.4507573783397675,
      "relation_loss": 0.00779494047164917,
      "selected_pairs": 25996.9,
      "step": 7920,
      "student_distance_mean": 28.180914878845215,
      "teacher_distance_mean": 0.8616521954536438,
      "valid_action_samples": 228.4,
      "weighted_auxiliary_loss": 0.00779494047164917,
      "weighted_relation_loss": 0.00779494047164917
    },
    {
      "action_query_batch_std": 0.42991114854812623,
      "action_query_norm": 44.056145095825194,
      "action_relation_loss": 0.008076891070231795,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.4181478016838165,
      "flow_matching_loss": 0.04815867953002453,
      "grad_norm": 0.12544234097003937,
      "h_stock_valid_norm": 97.6689697265625,
      "invalid_tail_samples": 24.4,
      "learning_rate": 8.7961174074178e-05,
      "loss": 0.0562,
      "positive_teacher_pairs": 26708.8,
      "query_h_norm_ratio": 0.4510804980993271,
      "relation_loss": 0.008076891070231795,
      "selected_pairs": 26708.8,
      "step": 7930,
      "student_distance_mean": 28.57270450592041,
      "teacher_distance_mean": 0.847359162569046,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.008076891070231795,
      "weighted_relation_loss": 0.008076891070231795
    },
    {
      "action_query_batch_std": 0.43116057813167574,
      "action_query_norm": 44.06350860595703,
      "action_relation_loss": 0.00784586500376463,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.427502338634238,
      "flow_matching_loss": 0.04656365439295769,
      "grad_norm": 0.14138519763946533,
      "h_stock_valid_norm": 97.83234024047852,
      "invalid_tail_samples": 24.3,
      "learning_rate": 8.792528009216942e-05,
      "loss": 0.0544,
      "positive_teacher_pairs": 26739.5,
      "query_h_norm_ratio": 0.4504032015800476,
      "relation_loss": 0.00784586500376463,
      "selected_pairs": 26739.5,
      "step": 7940,
      "student_distance_mean": 28.445284271240233,
      "teacher_distance_mean": 0.8463572025299072,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.00784586500376463,
      "weighted_relation_loss": 0.00784586500376463
    },
    {
      "action_query_batch_std": 0.4280575066804886,
      "action_query_norm": 44.05140151977539,
      "action_relation_loss": 0.007890864741057158,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.436856875584659,
      "flow_matching_loss": 0.04944423101842403,
      "grad_norm": 0.16519682109355927,
      "h_stock_valid_norm": 97.6106948852539,
      "invalid_tail_samples": 24.6,
      "learning_rate": 8.788934002736943e-05,
      "loss": 0.0573,
      "positive_teacher_pairs": 26670.3,
      "query_h_norm_ratio": 0.451303094625473,
      "relation_loss": 0.007890864741057158,
      "selected_pairs": 26670.3,
      "step": 7950,
      "student_distance_mean": 28.45250835418701,
      "teacher_distance_mean": 0.8533509850502015,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.007890864741057158,
      "weighted_relation_loss": 0.007890864741057158
    },
    {
      "action_query_batch_std": 0.4319839745759964,
      "action_query_norm": 44.06271896362305,
      "action_relation_loss": 0.008071517059579491,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.44621141253508,
      "flow_matching_loss": 0.0502660870552063,
      "grad_norm": 0.12003787606954575,
      "h_stock_valid_norm": 97.2046012878418,
      "invalid_tail_samples": 24.7,
      "learning_rate": 8.785335392344857e-05,
      "loss": 0.0583,
      "positive_teacher_pairs": 26644.4,
      "query_h_norm_ratio": 0.45330557227134705,
      "relation_loss": 0.008071517059579491,
      "selected_pairs": 26644.4,
      "step": 7960,
      "student_distance_mean": 28.50519790649414,
      "teacher_distance_mean": 0.8493892967700958,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.008071517059579491,
      "weighted_relation_loss": 0.008071517059579491
    },
    {
      "action_query_batch_std": 0.42808517813682556,
      "action_query_norm": 44.05615730285645,
      "action_relation_loss": 0.008104386925697326,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.4555659494855,
      "flow_matching_loss": 0.0497784037142992,
      "grad_norm": 0.1371244192123413,
      "h_stock_valid_norm": 97.87072143554687,
      "invalid_tail_samples": 24.1,
      "learning_rate": 8.781732182413336e-05,
      "loss": 0.0579,
      "positive_teacher_pairs": 26787.1,
      "query_h_norm_ratio": 0.4501479774713516,
      "relation_loss": 0.008104386925697326,
      "selected_pairs": 26787.1,
      "step": 7970,
      "student_distance_mean": 28.4625452041626,
      "teacher_distance_mean": 0.846110999584198,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.008104386925697326,
      "weighted_relation_loss": 0.008104386925697326
    },
    {
      "action_query_batch_std": 0.43188410103321073,
      "action_query_norm": 44.05945472717285,
      "action_relation_loss": 0.008085222728550435,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.464920486435921,
      "flow_matching_loss": 0.05054552927613258,
      "grad_norm": 0.13787010312080383,
      "h_stock_valid_norm": 97.48901290893555,
      "invalid_tail_samples": 26.3,
      "learning_rate": 8.778124377320618e-05,
      "loss": 0.0586,
      "positive_teacher_pairs": 26275.4,
      "query_h_norm_ratio": 0.45194695293903353,
      "relation_loss": 0.008085222728550435,
      "selected_pairs": 26275.4,
      "step": 7980,
      "student_distance_mean": 28.568597412109376,
      "teacher_distance_mean": 0.8540644109249115,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.008085222728550435,
      "weighted_relation_loss": 0.008085222728550435
    },
    {
      "action_query_batch_std": 0.4377309799194336,
      "action_query_norm": 44.06265029907227,
      "action_relation_loss": 0.00782334003597498,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.474275023386342,
      "flow_matching_loss": 0.04880519881844521,
      "grad_norm": 0.12498936057090759,
      "h_stock_valid_norm": 98.26786727905274,
      "invalid_tail_samples": 23.2,
      "learning_rate": 8.774511981450529e-05,
      "loss": 0.0566,
      "positive_teacher_pairs": 26989.2,
      "query_h_norm_ratio": 0.4483991116285324,
      "relation_loss": 0.00782334003597498,
      "selected_pairs": 26989.2,
      "step": 7990,
      "student_distance_mean": 28.889323806762697,
      "teacher_distance_mean": 0.8646834194660187,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.00782334003597498,
      "weighted_relation_loss": 0.00782334003597498
    },
    {
      "action_query_batch_std": 0.4324367940425873,
      "action_query_norm": 44.05207633972168,
      "action_relation_loss": 0.008065864630043506,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.483629560336763,
      "flow_matching_loss": 0.05200971402227879,
      "grad_norm": 0.10871385782957077,
      "h_stock_valid_norm": 98.25606002807618,
      "invalid_tail_samples": 24.3,
      "learning_rate": 8.770894999192467e-05,
      "loss": 0.0601,
      "positive_teacher_pairs": 26742.7,
      "query_h_norm_ratio": 0.4483447641134262,
      "relation_loss": 0.008065864630043506,
      "selected_pairs": 26742.7,
      "step": 8000,
      "student_distance_mean": 28.63080997467041,
      "teacher_distance_mean": 0.8639601111412049,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.008065864630043506,
      "weighted_relation_loss": 0.008065864630043506
    },
    {
      "action_query_batch_std": 0.4343517035245895,
      "action_query_norm": 44.057580184936526,
      "action_relation_loss": 0.008222323097288608,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.492984097287184,
      "flow_matching_loss": 0.049227886646986005,
      "grad_norm": 0.13957427442073822,
      "h_stock_valid_norm": 97.9587158203125,
      "invalid_tail_samples": 23.3,
      "learning_rate": 8.767273434941412e-05,
      "loss": 0.0575,
      "positive_teacher_pairs": 26966.4,
      "query_h_norm_ratio": 0.4497607558965683,
      "relation_loss": 0.008222323097288608,
      "selected_pairs": 26966.4,
      "step": 8010,
      "student_distance_mean": 28.599139595031737,
      "teacher_distance_mean": 0.8618746221065521,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.008222323097288608,
      "weighted_relation_loss": 0.008222323097288608
    },
    {
      "action_query_batch_std": 0.4367122709751129,
      "action_query_norm": 44.06407699584961,
      "action_relation_loss": 0.008430072478950024,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.5023386342376055,
      "flow_matching_loss": 0.0514911700040102,
      "grad_norm": 0.13082022964954376,
      "h_stock_valid_norm": 98.8245460510254,
      "invalid_tail_samples": 24.3,
      "learning_rate": 8.7636472930979e-05,
      "loss": 0.0599,
      "positive_teacher_pairs": 26736.2,
      "query_h_norm_ratio": 0.4458902895450592,
      "relation_loss": 0.008430072478950024,
      "selected_pairs": 26736.2,
      "step": 8020,
      "student_distance_mean": 29.054520034790038,
      "teacher_distance_mean": 0.8548044621944427,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.008430072478950024,
      "weighted_relation_loss": 0.008430072478950024
    },
    {
      "action_query_batch_std": 0.432910281419754,
      "action_query_norm": 44.05283088684082,
      "action_relation_loss": 0.007623405661433935,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.511693171188027,
      "flow_matching_loss": 0.04659348428249359,
      "grad_norm": 0.1356101781129837,
      "h_stock_valid_norm": 98.18447875976562,
      "invalid_tail_samples": 22.8,
      "learning_rate": 8.76001657806804e-05,
      "loss": 0.0542,
      "positive_teacher_pairs": 27082.3,
      "query_h_norm_ratio": 0.44868754148483275,
      "relation_loss": 0.007623405661433935,
      "selected_pairs": 27082.3,
      "step": 8030,
      "student_distance_mean": 28.915715408325195,
      "teacher_distance_mean": 0.8633624374866485,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.007623405661433935,
      "weighted_relation_loss": 0.007623405661433935
    },
    {
      "action_query_batch_std": 0.4254444450139999,
      "action_query_norm": 44.039401626586915,
      "action_relation_loss": 0.008143804594874382,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.521047708138447,
      "flow_matching_loss": 0.048436645045876504,
      "grad_norm": 0.13189874589443207,
      "h_stock_valid_norm": 96.79920806884766,
      "invalid_tail_samples": 24.9,
      "learning_rate": 8.756381294263491e-05,
      "loss": 0.0566,
      "positive_teacher_pairs": 26593.9,
      "query_h_norm_ratio": 0.4549610257148743,
      "relation_loss": 0.008143804594874382,
      "selected_pairs": 26593.9,
      "step": 8040,
      "student_distance_mean": 28.139229774475098,
      "teacher_distance_mean": 0.8498658299446106,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.008143804594874382,
      "weighted_relation_loss": 0.008143804594874382
    },
    {
      "action_query_batch_std": 0.4305803716182709,
      "action_query_norm": 44.049377822875975,
      "action_relation_loss": 0.00805154494009912,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.530402245088868,
      "flow_matching_loss": 0.047983264178037645,
      "grad_norm": 0.1241263821721077,
      "h_stock_valid_norm": 96.51574172973633,
      "invalid_tail_samples": 23.2,
      "learning_rate": 8.752741446101464e-05,
      "loss": 0.056,
      "positive_teacher_pairs": 26991.3,
      "query_h_norm_ratio": 0.45640315413475036,
      "relation_loss": 0.00805154494009912,
      "selected_pairs": 26991.3,
      "step": 8050,
      "student_distance_mean": 28.556801986694335,
      "teacher_distance_mean": 0.8608804285526276,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.00805154494009912,
      "weighted_relation_loss": 0.00805154494009912
    },
    {
      "action_query_batch_std": 0.4313523262739182,
      "action_query_norm": 44.040015029907224,
      "action_relation_loss": 0.007748089963570237,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.539756782039289,
      "flow_matching_loss": 0.04874866604804993,
      "grad_norm": 0.1686355620622635,
      "h_stock_valid_norm": 96.67733306884766,
      "invalid_tail_samples": 21.2,
      "learning_rate": 8.749097038004722e-05,
      "loss": 0.0565,
      "positive_teacher_pairs": 27456.2,
      "query_h_norm_ratio": 0.4555389106273651,
      "relation_loss": 0.007748089963570237,
      "selected_pairs": 27456.2,
      "step": 8060,
      "student_distance_mean": 28.618613243103027,
      "teacher_distance_mean": 0.8588360607624054,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.007748089963570237,
      "weighted_relation_loss": 0.007748089963570237
    },
    {
      "action_query_batch_std": 0.4331334501504898,
      "action_query_norm": 44.04793128967285,
      "action_relation_loss": 0.007639320474117995,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.54911131898971,
      "flow_matching_loss": 0.04858467429876327,
      "grad_norm": 0.14315804839134216,
      "h_stock_valid_norm": 96.88815612792969,
      "invalid_tail_samples": 24.0,
      "learning_rate": 8.745448074401561e-05,
      "loss": 0.0562,
      "positive_teacher_pairs": 26803.1,
      "query_h_norm_ratio": 0.45462973713874816,
      "relation_loss": 0.007639320474117995,
      "selected_pairs": 26803.1,
      "step": 8070,
      "student_distance_mean": 28.682634353637695,
      "teacher_distance_mean": 0.8577327787876129,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.007639320474117995,
      "weighted_relation_loss": 0.007639320474117995
    },
    {
      "action_query_batch_std": 0.43072444200515747,
      "action_query_norm": 44.046311950683595,
      "action_relation_loss": 0.007858684612438082,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.558465855940131,
      "flow_matching_loss": 0.04999744445085526,
      "grad_norm": 0.17809021472930908,
      "h_stock_valid_norm": 96.92450637817383,
      "invalid_tail_samples": 20.4,
      "learning_rate": 8.741794559725818e-05,
      "loss": 0.0579,
      "positive_teacher_pairs": 27639.3,
      "query_h_norm_ratio": 0.4544407606124878,
      "relation_loss": 0.007858684612438082,
      "selected_pairs": 27639.3,
      "step": 8080,
      "student_distance_mean": 28.645397758483888,
      "teacher_distance_mean": 0.8565585970878601,
      "valid_action_samples": 235.6,
      "weighted_auxiliary_loss": 0.007858684612438082,
      "weighted_relation_loss": 0.007858684612438082
    },
    {
      "action_query_batch_std": 0.43036467432975767,
      "action_query_norm": 44.037390518188474,
      "action_relation_loss": 0.008061423571780324,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.567820392890552,
      "flow_matching_loss": 0.04984950385987759,
      "grad_norm": 0.13149143755435944,
      "h_stock_valid_norm": 98.06483993530273,
      "invalid_tail_samples": 25.7,
      "learning_rate": 8.738136498416856e-05,
      "loss": 0.0579,
      "positive_teacher_pairs": 26410.0,
      "query_h_norm_ratio": 0.4490795314311981,
      "relation_loss": 0.008061423571780324,
      "selected_pairs": 26410.0,
      "step": 8090,
      "student_distance_mean": 28.60809917449951,
      "teacher_distance_mean": 0.8579094707965851,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.008061423571780324,
      "weighted_relation_loss": 0.008061423571780324
    },
    {
      "action_query_batch_std": 0.42952040731906893,
      "action_query_norm": 44.04198684692383,
      "action_relation_loss": 0.007967438222840429,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.5771749298409725,
      "flow_matching_loss": 0.04821065366268158,
      "grad_norm": 0.13860121369361877,
      "h_stock_valid_norm": 99.73555374145508,
      "invalid_tail_samples": 21.6,
      "learning_rate": 8.734473894919564e-05,
      "loss": 0.0562,
      "positive_teacher_pairs": 27361.5,
      "query_h_norm_ratio": 0.44159904718399046,
      "relation_loss": 0.007967438222840429,
      "selected_pairs": 27361.5,
      "step": 8100,
      "student_distance_mean": 28.464018058776855,
      "teacher_distance_mean": 0.8503364622592926,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.007967438222840429,
      "weighted_relation_loss": 0.007967438222840429
    },
    {
      "action_query_batch_std": 0.4232623755931854,
      "action_query_norm": 44.015943908691405,
      "action_relation_loss": 0.00762948147021234,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.586529466791394,
      "flow_matching_loss": 0.04849672988057137,
      "grad_norm": 0.1472035050392151,
      "h_stock_valid_norm": 99.47655487060547,
      "invalid_tail_samples": 26.5,
      "learning_rate": 8.730806753684353e-05,
      "loss": 0.0561,
      "positive_teacher_pairs": 26233.7,
      "query_h_norm_ratio": 0.4424818575382233,
      "relation_loss": 0.00762948147021234,
      "selected_pairs": 26233.7,
      "step": 8110,
      "student_distance_mean": 27.975395011901856,
      "teacher_distance_mean": 0.842658269405365,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.00762948147021234,
      "weighted_relation_loss": 0.00762948147021234
    },
    {
      "action_query_batch_std": 0.4250349372625351,
      "action_query_norm": 44.021766662597656,
      "action_relation_loss": 0.007919201953336596,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.595884003741815,
      "flow_matching_loss": 0.049955977499485014,
      "grad_norm": 0.17510591447353363,
      "h_stock_valid_norm": 98.87943801879882,
      "invalid_tail_samples": 24.9,
      "learning_rate": 8.727135079167144e-05,
      "loss": 0.0579,
      "positive_teacher_pairs": 26593.6,
      "query_h_norm_ratio": 0.4452111005783081,
      "relation_loss": 0.007919201953336596,
      "selected_pairs": 26593.6,
      "step": 8120,
      "student_distance_mean": 28.189414405822752,
      "teacher_distance_mean": 0.8521535515785217,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.007919201953336596,
      "weighted_relation_loss": 0.007919201953336596
    },
    {
      "action_query_batch_std": 0.4282630175352097,
      "action_query_norm": 44.02682876586914,
      "action_relation_loss": 0.008071278268471361,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.605238540692236,
      "flow_matching_loss": 0.048348259553313254,
      "grad_norm": 0.14106886088848114,
      "h_stock_valid_norm": 98.83256759643555,
      "invalid_tail_samples": 21.8,
      "learning_rate": 8.723458875829368e-05,
      "loss": 0.0564,
      "positive_teacher_pairs": 27317.2,
      "query_h_norm_ratio": 0.44547312557697294,
      "relation_loss": 0.008071278268471361,
      "selected_pairs": 27317.2,
      "step": 8130,
      "student_distance_mean": 28.353531646728516,
      "teacher_distance_mean": 0.8486208081245422,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.008071278268471361,
      "weighted_relation_loss": 0.008071278268471361
    },
    {
      "action_query_batch_std": 0.4352621853351593,
      "action_query_norm": 44.03307952880859,
      "action_relation_loss": 0.007884502504020929,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.614593077642657,
      "flow_matching_loss": 0.04914357513189316,
      "grad_norm": 0.13511285185813904,
      "h_stock_valid_norm": 97.42145462036133,
      "invalid_tail_samples": 22.2,
      "learning_rate": 8.719778148137958e-05,
      "loss": 0.057,
      "positive_teacher_pairs": 27218.6,
      "query_h_norm_ratio": 0.4520096480846405,
      "relation_loss": 0.007884502504020929,
      "selected_pairs": 27218.6,
      "step": 8140,
      "student_distance_mean": 28.746331405639648,
      "teacher_distance_mean": 0.853823858499527,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.007884502504020929,
      "weighted_relation_loss": 0.007884502504020929
    },
    {
      "action_query_batch_std": 0.43020425736904144,
      "action_query_norm": 44.023797607421876,
      "action_relation_loss": 0.00751532856374979,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.623947614593078,
      "flow_matching_loss": 0.050651971623301505,
      "grad_norm": 0.17568674683570862,
      "h_stock_valid_norm": 96.28364562988281,
      "invalid_tail_samples": 21.7,
      "learning_rate": 8.716092900565346e-05,
      "loss": 0.0582,
      "positive_teacher_pairs": 27339.6,
      "query_h_norm_ratio": 0.4572371423244476,
      "relation_loss": 0.00751532856374979,
      "selected_pairs": 27339.6,
      "step": 8150,
      "student_distance_mean": 28.450103187561034,
      "teacher_distance_mean": 0.8594342589378356,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.00751532856374979,
      "weighted_relation_loss": 0.00751532856374979
    },
    {
      "action_query_batch_std": 0.4255453497171402,
      "action_query_norm": 44.01684379577637,
      "action_relation_loss": 0.007922840490937232,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.633302151543498,
      "flow_matching_loss": 0.0472901850938797,
      "grad_norm": 0.1906760185956955,
      "h_stock_valid_norm": 96.65470504760742,
      "invalid_tail_samples": 25.6,
      "learning_rate": 8.712403137589456e-05,
      "loss": 0.0552,
      "positive_teacher_pairs": 26443.3,
      "query_h_norm_ratio": 0.45540756583213804,
      "relation_loss": 0.007922840490937232,
      "selected_pairs": 26443.3,
      "step": 8160,
      "student_distance_mean": 28.209530448913576,
      "teacher_distance_mean": 0.8440684735774994,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.007922840490937232,
      "weighted_relation_loss": 0.007922840490937232
    },
    {
      "action_query_batch_std": 0.42777018845081327,
      "action_query_norm": 44.02592391967774,
      "action_relation_loss": 0.007562635373324156,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.642656688493919,
      "flow_matching_loss": 0.045820963382720944,
      "grad_norm": 0.12971746921539307,
      "h_stock_valid_norm": 97.02535095214844,
      "invalid_tail_samples": 24.9,
      "learning_rate": 8.708708863693697e-05,
      "loss": 0.0534,
      "positive_teacher_pairs": 26592.6,
      "query_h_norm_ratio": 0.4537610292434692,
      "relation_loss": 0.007562635373324156,
      "selected_pairs": 26592.6,
      "step": 8170,
      "student_distance_mean": 28.366193580627442,
      "teacher_distance_mean": 0.8503333389759063,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.007562635373324156,
      "weighted_relation_loss": 0.007562635373324156
    },
    {
      "action_query_batch_std": 0.43296835124492644,
      "action_query_norm": 44.03220825195312,
      "action_relation_loss": 0.007506817951798439,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.65201122544434,
      "flow_matching_loss": 0.04626196809113026,
      "grad_norm": 0.16591913998126984,
      "h_stock_valid_norm": 97.50201644897462,
      "invalid_tail_samples": 24.8,
      "learning_rate": 8.70501008336696e-05,
      "loss": 0.0538,
      "positive_teacher_pairs": 26625.7,
      "query_h_norm_ratio": 0.45161049664020536,
      "relation_loss": 0.007506817951798439,
      "selected_pairs": 26625.7,
      "step": 8180,
      "student_distance_mean": 28.821630477905273,
      "teacher_distance_mean": 0.8682703733444214,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.007506817951798439,
      "weighted_relation_loss": 0.007506817951798439
    },
    {
      "action_query_batch_std": 0.4288283109664917,
      "action_query_norm": 44.01708908081055,
      "action_relation_loss": 0.007889734301716089,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.6613657623947615,
      "flow_matching_loss": 0.04704994894564152,
      "grad_norm": 0.12006941437721252,
      "h_stock_valid_norm": 97.74892196655273,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.701306801103611e-05,
      "loss": 0.0549,
      "positive_teacher_pairs": 26687.9,
      "query_h_norm_ratio": 0.4503095746040344,
      "relation_loss": 0.007889734301716089,
      "selected_pairs": 26687.9,
      "step": 8190,
      "student_distance_mean": 28.488816833496095,
      "teacher_distance_mean": 0.8569098889827729,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.007889734301716089,
      "weighted_relation_loss": 0.007889734301716089
    },
    {
      "action_query_batch_std": 0.4285422474145889,
      "action_query_norm": 44.01868858337402,
      "action_relation_loss": 0.007792112370952964,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.670720299345183,
      "flow_matching_loss": 0.05030185766518116,
      "grad_norm": 0.17416079342365265,
      "h_stock_valid_norm": 97.23787841796874,
      "invalid_tail_samples": 24.2,
      "learning_rate": 8.697599021403489e-05,
      "loss": 0.0581,
      "positive_teacher_pairs": 26757.4,
      "query_h_norm_ratio": 0.45269259810447693,
      "relation_loss": 0.007792112370952964,
      "selected_pairs": 26757.4,
      "step": 8200,
      "student_distance_mean": 28.445238876342774,
      "teacher_distance_mean": 0.8611616313457489,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.007792112370952964,
      "weighted_relation_loss": 0.007792112370952964
    },
    {
      "action_query_batch_std": 0.4257931798696518,
      "action_query_norm": 44.012088012695315,
      "action_relation_loss": 0.007580767851322889,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.680074836295604,
      "flow_matching_loss": 0.048674411699175836,
      "grad_norm": 0.18427976965904236,
      "h_stock_valid_norm": 97.12867584228516,
      "invalid_tail_samples": 23.8,
      "learning_rate": 8.693886748771896e-05,
      "loss": 0.0563,
      "positive_teacher_pairs": 26844.6,
      "query_h_norm_ratio": 0.45313394665718076,
      "relation_loss": 0.007580767851322889,
      "selected_pairs": 26844.6,
      "step": 8210,
      "student_distance_mean": 28.33196144104004,
      "teacher_distance_mean": 0.84745472073555,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.007580767851322889,
      "weighted_relation_loss": 0.007580767851322889
    },
    {
      "action_query_batch_std": 0.4270163387060165,
      "action_query_norm": 44.00828590393066,
      "action_relation_loss": 0.0078091217204928395,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.689429373246024,
      "flow_matching_loss": 0.04552707932889462,
      "grad_norm": 0.08977701514959335,
      "h_stock_valid_norm": 97.05286254882813,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.690169987719593e-05,
      "loss": 0.0533,
      "positive_teacher_pairs": 26694.2,
      "query_h_norm_ratio": 0.4534525662660599,
      "relation_loss": 0.0078091217204928395,
      "selected_pairs": 26694.2,
      "step": 8220,
      "student_distance_mean": 28.329614639282227,
      "teacher_distance_mean": 0.8469603776931762,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0078091217204928395,
      "weighted_relation_loss": 0.0078091217204928395
    },
    {
      "action_query_batch_std": 0.43146146833896637,
      "action_query_norm": 44.01851692199707,
      "action_relation_loss": 0.007952545676380396,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.698783910196445,
      "flow_matching_loss": 0.0493313118815422,
      "grad_norm": 0.14051923155784607,
      "h_stock_valid_norm": 97.24282760620117,
      "invalid_tail_samples": 21.4,
      "learning_rate": 8.686448742762791e-05,
      "loss": 0.0573,
      "positive_teacher_pairs": 27404.6,
      "query_h_norm_ratio": 0.4526707321405411,
      "relation_loss": 0.007952545676380396,
      "selected_pairs": 27404.6,
      "step": 8230,
      "student_distance_mean": 28.552868461608888,
      "teacher_distance_mean": 0.8488977253437042,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.007952545676380396,
      "weighted_relation_loss": 0.007952545676380396
    },
    {
      "action_query_batch_std": 0.4242718040943146,
      "action_query_norm": 44.01430473327637,
      "action_relation_loss": 0.007697345549240708,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.708138447146866,
      "flow_matching_loss": 0.04589716829359532,
      "grad_norm": 0.1309722661972046,
      "h_stock_valid_norm": 96.73686447143555,
      "invalid_tail_samples": 24.9,
      "learning_rate": 8.682723018423161e-05,
      "loss": 0.0536,
      "positive_teacher_pairs": 26593.1,
      "query_h_norm_ratio": 0.4549908936023712,
      "relation_loss": 0.007697345549240708,
      "selected_pairs": 26593.1,
      "step": 8240,
      "student_distance_mean": 28.26511421203613,
      "teacher_distance_mean": 0.8615715384483338,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.007697345549240708,
      "weighted_relation_loss": 0.007697345549240708
    },
    {
      "action_query_batch_std": 0.4332836985588074,
      "action_query_norm": 44.02153091430664,
      "action_relation_loss": 0.0074513737577945,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.717492984097287,
      "flow_matching_loss": 0.045576220378279686,
      "grad_norm": 0.14622004330158234,
      "h_stock_valid_norm": 96.60058364868163,
      "invalid_tail_samples": 23.8,
      "learning_rate": 8.678992819227804e-05,
      "loss": 0.053,
      "positive_teacher_pairs": 26852.4,
      "query_h_norm_ratio": 0.45571787357330323,
      "relation_loss": 0.0074513737577945,
      "selected_pairs": 26852.4,
      "step": 8250,
      "student_distance_mean": 28.83572425842285,
      "teacher_distance_mean": 0.8652250468730927,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0074513737577945,
      "weighted_relation_loss": 0.0074513737577945
    },
    {
      "action_query_batch_std": 0.4318415492773056,
      "action_query_norm": 44.016373825073245,
      "action_relation_loss": 0.007649579597637057,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.726847521047708,
      "flow_matching_loss": 0.04766024425625801,
      "grad_norm": 0.16256998479366302,
      "h_stock_valid_norm": 96.22388763427735,
      "invalid_tail_samples": 24.6,
      "learning_rate": 8.675258149709265e-05,
      "loss": 0.0553,
      "positive_teacher_pairs": 26673.4,
      "query_h_norm_ratio": 0.4574431896209717,
      "relation_loss": 0.007649579597637057,
      "selected_pairs": 26673.4,
      "step": 8260,
      "student_distance_mean": 28.567394828796388,
      "teacher_distance_mean": 0.849784541130066,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.007649579597637057,
      "weighted_relation_loss": 0.007649579597637057
    },
    {
      "action_query_batch_std": 0.42702820897102356,
      "action_query_norm": 44.001701736450194,
      "action_relation_loss": 0.007348482217639685,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.736202057998129,
      "flow_matching_loss": 0.04590571895241737,
      "grad_norm": 0.12362806499004364,
      "h_stock_valid_norm": 97.31272506713867,
      "invalid_tail_samples": 24.8,
      "learning_rate": 8.67151901440552e-05,
      "loss": 0.0533,
      "positive_teacher_pairs": 26626.7,
      "query_h_norm_ratio": 0.4521714150905609,
      "relation_loss": 0.007348482217639685,
      "selected_pairs": 26626.7,
      "step": 8270,
      "student_distance_mean": 28.38247470855713,
      "teacher_distance_mean": 0.8546716570854187,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.007348482217639685,
      "weighted_relation_loss": 0.007348482217639685
    },
    {
      "action_query_batch_std": 0.41621835231781007,
      "action_query_norm": 43.97964172363281,
      "action_relation_loss": 0.007771969167515635,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.74555659494855,
      "flow_matching_loss": 0.04856429435312748,
      "grad_norm": 0.1529977172613144,
      "h_stock_valid_norm": 96.95318603515625,
      "invalid_tail_samples": 22.0,
      "learning_rate": 8.667775417859971e-05,
      "loss": 0.0563,
      "positive_teacher_pairs": 27274.3,
      "query_h_norm_ratio": 0.4536209672689438,
      "relation_loss": 0.007771969167515635,
      "selected_pairs": 27274.3,
      "step": 8280,
      "student_distance_mean": 27.76463680267334,
      "teacher_distance_mean": 0.8463737189769744,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.007771969167515635,
      "weighted_relation_loss": 0.007771969167515635
    },
    {
      "action_query_batch_std": 0.42885218262672425,
      "action_query_norm": 43.991603469848634,
      "action_relation_loss": 0.007503838278353214,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.754911131898971,
      "flow_matching_loss": 0.04482811763882637,
      "grad_norm": 0.13715891540050507,
      "h_stock_valid_norm": 96.63525085449218,
      "invalid_tail_samples": 22.4,
      "learning_rate": 8.664027364621441e-05,
      "loss": 0.0523,
      "positive_teacher_pairs": 27181.9,
      "query_h_norm_ratio": 0.45523732602596284,
      "relation_loss": 0.007503838278353214,
      "selected_pairs": 27181.9,
      "step": 8290,
      "student_distance_mean": 28.231509399414062,
      "teacher_distance_mean": 0.8445720732212066,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.007503838278353214,
      "weighted_relation_loss": 0.007503838278353214
    },
    {
      "action_query_batch_std": 0.4280654489994049,
      "action_query_norm": 43.98907623291016,
      "action_relation_loss": 0.007607677113264799,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.764265668849392,
      "flow_matching_loss": 0.045528825372457504,
      "grad_norm": 0.15106922388076782,
      "h_stock_valid_norm": 96.96095657348633,
      "invalid_tail_samples": 21.4,
      "learning_rate": 8.660274859244168e-05,
      "loss": 0.0531,
      "positive_teacher_pairs": 27406.4,
      "query_h_norm_ratio": 0.4536824882030487,
      "relation_loss": 0.007607677113264799,
      "selected_pairs": 27406.4,
      "step": 8300,
      "student_distance_mean": 28.39564781188965,
      "teacher_distance_mean": 0.8584744453430175,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.007607677113264799,
      "weighted_relation_loss": 0.007607677113264799
    },
    {
      "action_query_batch_std": 0.42289417088031767,
      "action_query_norm": 43.98852043151855,
      "action_relation_loss": 0.007574607618153095,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.773620205799813,
      "flow_matching_loss": 0.04673423431813717,
      "grad_norm": 0.16602493822574615,
      "h_stock_valid_norm": 98.015283203125,
      "invalid_tail_samples": 22.8,
      "learning_rate": 8.656517906287799e-05,
      "loss": 0.0543,
      "positive_teacher_pairs": 27082.1,
      "query_h_norm_ratio": 0.44880571961402893,
      "relation_loss": 0.007574607618153095,
      "selected_pairs": 27082.1,
      "step": 8310,
      "student_distance_mean": 28.073580169677733,
      "teacher_distance_mean": 0.8580892503261566,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.007574607618153095,
      "weighted_relation_loss": 0.007574607618153095
    },
    {
      "action_query_batch_std": 0.4226212531328201,
      "action_query_norm": 43.98795700073242,
      "action_relation_loss": 0.007412860216572881,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.782974742750234,
      "flow_matching_loss": 0.047965288534760474,
      "grad_norm": 0.15595078468322754,
      "h_stock_valid_norm": 98.29523468017578,
      "invalid_tail_samples": 22.4,
      "learning_rate": 8.652756510317387e-05,
      "loss": 0.0554,
      "positive_teacher_pairs": 27179.3,
      "query_h_norm_ratio": 0.4475112438201904,
      "relation_loss": 0.007412860216572881,
      "selected_pairs": 27179.3,
      "step": 8320,
      "student_distance_mean": 28.082425689697267,
      "teacher_distance_mean": 0.854671961069107,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.007412860216572881,
      "weighted_relation_loss": 0.007412860216572881
    },
    {
      "action_query_batch_std": 0.42596550583839415,
      "action_query_norm": 43.98702812194824,
      "action_relation_loss": 0.007696632901206613,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.792329279700655,
      "flow_matching_loss": 0.04512205123901367,
      "grad_norm": 0.11576660722494125,
      "h_stock_valid_norm": 97.12140579223633,
      "invalid_tail_samples": 23.1,
      "learning_rate": 8.648990675903382e-05,
      "loss": 0.0528,
      "positive_teacher_pairs": 27013.8,
      "query_h_norm_ratio": 0.4529211014509201,
      "relation_loss": 0.007696632901206613,
      "selected_pairs": 27013.8,
      "step": 8330,
      "student_distance_mean": 28.316817474365234,
      "teacher_distance_mean": 0.855246502161026,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.007696632901206613,
      "weighted_relation_loss": 0.007696632901206613
    },
    {
      "action_query_batch_std": 0.4281898856163025,
      "action_query_norm": 43.98663597106933,
      "action_relation_loss": 0.007746043149381876,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.801683816651076,
      "flow_matching_loss": 0.047424912452697754,
      "grad_norm": 0.1536085456609726,
      "h_stock_valid_norm": 95.43145751953125,
      "invalid_tail_samples": 22.4,
      "learning_rate": 8.64522040762163e-05,
      "loss": 0.0552,
      "positive_teacher_pairs": 27175.6,
      "query_h_norm_ratio": 0.4609294980764389,
      "relation_loss": 0.007746043149381876,
      "selected_pairs": 27175.6,
      "step": 8340,
      "student_distance_mean": 28.52908115386963,
      "teacher_distance_mean": 0.8550694584846497,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.007746043149381876,
      "weighted_relation_loss": 0.007746043149381876
    },
    {
      "action_query_batch_std": 0.429101699590683,
      "action_query_norm": 43.99142951965332,
      "action_relation_loss": 0.007792110089212656,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.8110383536014965,
      "flow_matching_loss": 0.04649077653884888,
      "grad_norm": 0.13191872835159302,
      "h_stock_valid_norm": 96.82771377563476,
      "invalid_tail_samples": 25.4,
      "learning_rate": 8.64144571005336e-05,
      "loss": 0.0543,
      "positive_teacher_pairs": 26490.7,
      "query_h_norm_ratio": 0.454364025592804,
      "relation_loss": 0.007792110089212656,
      "selected_pairs": 26490.7,
      "step": 8350,
      "student_distance_mean": 28.503062057495118,
      "teacher_distance_mean": 0.8580354869365692,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.007792110089212656,
      "weighted_relation_loss": 0.007792110089212656
    },
    {
      "action_query_batch_std": 0.4269571155309677,
      "action_query_norm": 43.98963241577148,
      "action_relation_loss": 0.0077474603429436685,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.8203928905519176,
      "flow_matching_loss": 0.047159531712532045,
      "grad_norm": 0.15741096436977386,
      "h_stock_valid_norm": 99.35511703491211,
      "invalid_tail_samples": 25.3,
      "learning_rate": 8.637666587785184e-05,
      "loss": 0.0549,
      "positive_teacher_pairs": 26507.8,
      "query_h_norm_ratio": 0.4427631378173828,
      "relation_loss": 0.0077474603429436685,
      "selected_pairs": 26507.8,
      "step": 8360,
      "student_distance_mean": 28.226550674438478,
      "teacher_distance_mean": 0.8529709875583649,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0077474603429436685,
      "weighted_relation_loss": 0.0077474603429436685
    },
    {
      "action_query_batch_std": 0.4302740603685379,
      "action_query_norm": 43.98711891174317,
      "action_relation_loss": 0.00730907991528511,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.829747427502339,
      "flow_matching_loss": 0.049688930809497836,
      "grad_norm": 0.15134063363075256,
      "h_stock_valid_norm": 99.44907836914062,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.633883045409096e-05,
      "loss": 0.057,
      "positive_teacher_pairs": 26873.5,
      "query_h_norm_ratio": 0.44231240153312684,
      "relation_loss": 0.00730907991528511,
      "selected_pairs": 26873.5,
      "step": 8370,
      "student_distance_mean": 28.43421287536621,
      "teacher_distance_mean": 0.856433767080307,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.00730907991528511,
      "weighted_relation_loss": 0.00730907991528511
    },
    {
      "action_query_batch_std": 0.42592826783657073,
      "action_query_norm": 43.98338317871094,
      "action_relation_loss": 0.007750063855201006,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.83910196445276,
      "flow_matching_loss": 0.047866794466972354,
      "grad_norm": 0.12580032646656036,
      "h_stock_valid_norm": 98.5145767211914,
      "invalid_tail_samples": 24.1,
      "learning_rate": 8.630095087522457e-05,
      "loss": 0.0556,
      "positive_teacher_pairs": 26778.5,
      "query_h_norm_ratio": 0.4464700520038605,
      "relation_loss": 0.007750063855201006,
      "selected_pairs": 26778.5,
      "step": 8380,
      "student_distance_mean": 28.309836006164552,
      "teacher_distance_mean": 0.8558541357517242,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.007750063855201006,
      "weighted_relation_loss": 0.007750063855201006
    },
    {
      "action_query_batch_std": 0.426469099521637,
      "action_query_norm": 43.97665939331055,
      "action_relation_loss": 0.007726712618023157,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.848456501403181,
      "flow_matching_loss": 0.050018896162509915,
      "grad_norm": 0.1435718536376953,
      "h_stock_valid_norm": 96.91215438842774,
      "invalid_tail_samples": 26.4,
      "learning_rate": 8.62630271872799e-05,
      "loss": 0.0577,
      "positive_teacher_pairs": 26247.1,
      "query_h_norm_ratio": 0.4537999629974365,
      "relation_loss": 0.007726712618023157,
      "selected_pairs": 26247.1,
      "step": 8390,
      "student_distance_mean": 28.304340744018553,
      "teacher_distance_mean": 0.8562076389789581,
      "valid_action_samples": 229.6,
      "weighted_auxiliary_loss": 0.007726712618023157,
      "weighted_relation_loss": 0.007726712618023157
    },
    {
      "action_query_batch_std": 0.4239307552576065,
      "action_query_norm": 43.974619674682614,
      "action_relation_loss": 0.007474426459521055,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.857811038353601,
      "flow_matching_loss": 0.0462461706250906,
      "grad_norm": 0.1352839320898056,
      "h_stock_valid_norm": 95.68838043212891,
      "invalid_tail_samples": 23.4,
      "learning_rate": 8.62250594363378e-05,
      "loss": 0.0537,
      "positive_teacher_pairs": 26954.1,
      "query_h_norm_ratio": 0.4595639169216156,
      "relation_loss": 0.007474426459521055,
      "selected_pairs": 26954.1,
      "step": 8400,
      "student_distance_mean": 28.124852561950682,
      "teacher_distance_mean": 0.8546788930892945,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.007474426459521055,
      "weighted_relation_loss": 0.007474426459521055
    },
    {
      "action_query_batch_std": 0.42263291478157045,
      "action_query_norm": 43.96986961364746,
      "action_relation_loss": 0.007402658276259899,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.867165575304022,
      "flow_matching_loss": 0.046580780297517776,
      "grad_norm": 0.09298589825630188,
      "h_stock_valid_norm": 95.80672607421874,
      "invalid_tail_samples": 25.6,
      "learning_rate": 8.61870476685327e-05,
      "loss": 0.054,
      "positive_teacher_pairs": 26431.7,
      "query_h_norm_ratio": 0.4589566469192505,
      "relation_loss": 0.007402658276259899,
      "selected_pairs": 26431.7,
      "step": 8410,
      "student_distance_mean": 28.128762817382814,
      "teacher_distance_mean": 0.8629074573516846,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.007402658276259899,
      "weighted_relation_loss": 0.007402658276259899
    },
    {
      "action_query_batch_std": 0.42753663659095764,
      "action_query_norm": 43.97956733703613,
      "action_relation_loss": 0.007007147278636694,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.876520112254443,
      "flow_matching_loss": 0.04596398212015629,
      "grad_norm": 0.14035481214523315,
      "h_stock_valid_norm": 97.055126953125,
      "invalid_tail_samples": 22.6,
      "learning_rate": 8.614899193005248e-05,
      "loss": 0.053,
      "positive_teacher_pairs": 27131.2,
      "query_h_norm_ratio": 0.4531461328268051,
      "relation_loss": 0.007007147278636694,
      "selected_pairs": 27131.2,
      "step": 8420,
      "student_distance_mean": 28.39459800720215,
      "teacher_distance_mean": 0.8633869886398315,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.007007147278636694,
      "weighted_relation_loss": 0.007007147278636694
    },
    {
      "action_query_batch_std": 0.4238526225090027,
      "action_query_norm": 43.96579360961914,
      "action_relation_loss": 0.0073253751732409,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.885874649204864,
      "flow_matching_loss": 0.047242185473442076,
      "grad_norm": 0.12715783715248108,
      "h_stock_valid_norm": 98.16782684326172,
      "invalid_tail_samples": 22.8,
      "learning_rate": 8.611089226713842e-05,
      "loss": 0.0546,
      "positive_teacher_pairs": 27084.8,
      "query_h_norm_ratio": 0.44786760210990906,
      "relation_loss": 0.0073253751732409,
      "selected_pairs": 27084.8,
      "step": 8430,
      "student_distance_mean": 27.97263870239258,
      "teacher_distance_mean": 0.853873860836029,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0073253751732409,
      "weighted_relation_loss": 0.0073253751732409
    },
    {
      "action_query_batch_std": 0.4258542239665985,
      "action_query_norm": 43.9783374786377,
      "action_relation_loss": 0.007625005627050996,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.8952291861552855,
      "flow_matching_loss": 0.045503328740596774,
      "grad_norm": 0.15886454284191132,
      "h_stock_valid_norm": 98.20935745239258,
      "invalid_tail_samples": 23.1,
      "learning_rate": 8.607274872608521e-05,
      "loss": 0.0531,
      "positive_teacher_pairs": 27010.7,
      "query_h_norm_ratio": 0.4478039354085922,
      "relation_loss": 0.007625005627050996,
      "selected_pairs": 27010.7,
      "step": 8440,
      "student_distance_mean": 28.226957702636717,
      "teacher_distance_mean": 0.8569356560707092,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.007625005627050996,
      "weighted_relation_loss": 0.007625005627050996
    },
    {
      "action_query_batch_std": 0.4243734300136566,
      "action_query_norm": 43.97398071289062,
      "action_relation_loss": 0.00723569500260055,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.904583723105707,
      "flow_matching_loss": 0.04600502103567124,
      "grad_norm": 0.12267281860113144,
      "h_stock_valid_norm": 97.2572410583496,
      "invalid_tail_samples": 24.0,
      "learning_rate": 8.603456135324088e-05,
      "loss": 0.0532,
      "positive_teacher_pairs": 26807.3,
      "query_h_norm_ratio": 0.45214673280715945,
      "relation_loss": 0.00723569500260055,
      "selected_pairs": 26807.3,
      "step": 8450,
      "student_distance_mean": 28.15826873779297,
      "teacher_distance_mean": 0.849487566947937,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.00723569500260055,
      "weighted_relation_loss": 0.00723569500260055
    },
    {
      "action_query_batch_std": 0.4157583326101303,
      "action_query_norm": 43.95969734191895,
      "action_relation_loss": 0.0075571052730083466,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.913938260056128,
      "flow_matching_loss": 0.04580897204577923,
      "grad_norm": 0.1267099827528,
      "h_stock_valid_norm": 96.5091552734375,
      "invalid_tail_samples": 25.1,
      "learning_rate": 8.599633019500665e-05,
      "loss": 0.0534,
      "positive_teacher_pairs": 26547.9,
      "query_h_norm_ratio": 0.45550259947776794,
      "relation_loss": 0.0075571052730083466,
      "selected_pairs": 26547.9,
      "step": 8460,
      "student_distance_mean": 27.663361930847167,
      "teacher_distance_mean": 0.8437596499919892,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0075571052730083466,
      "weighted_relation_loss": 0.0075571052730083466
    },
    {
      "action_query_batch_std": 0.41521849036216735,
      "action_query_norm": 43.947044372558594,
      "action_relation_loss": 0.007277318555861711,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.923292797006548,
      "flow_matching_loss": 0.04744480215013027,
      "grad_norm": 0.11089879274368286,
      "h_stock_valid_norm": 96.83582916259766,
      "invalid_tail_samples": 20.9,
      "learning_rate": 8.595805529783702e-05,
      "loss": 0.0547,
      "positive_teacher_pairs": 27523.1,
      "query_h_norm_ratio": 0.45383664071559904,
      "relation_loss": 0.007277318555861711,
      "selected_pairs": 27523.1,
      "step": 8470,
      "student_distance_mean": 27.561239624023436,
      "teacher_distance_mean": 0.8487771928310395,
      "valid_action_samples": 235.1,
      "weighted_auxiliary_loss": 0.007277318555861711,
      "weighted_relation_loss": 0.007277318555861711
    },
    {
      "action_query_batch_std": 0.4230006754398346,
      "action_query_norm": 43.9656681060791,
      "action_relation_loss": 0.007297933148220181,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.932647333956969,
      "flow_matching_loss": 0.048602069914340976,
      "grad_norm": 0.12216314673423767,
      "h_stock_valid_norm": 96.16336593627929,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.591973670823958e-05,
      "loss": 0.0559,
      "positive_teacher_pairs": 26875.1,
      "query_h_norm_ratio": 0.45722258985042574,
      "relation_loss": 0.007297933148220181,
      "selected_pairs": 26875.1,
      "step": 8480,
      "student_distance_mean": 27.988363838195802,
      "teacher_distance_mean": 0.8533796668052673,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.007297933148220181,
      "weighted_relation_loss": 0.007297933148220181
    },
    {
      "action_query_batch_std": 0.42354666292667387,
      "action_query_norm": 43.963528060913085,
      "action_relation_loss": 0.007542255148291588,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.94200187090739,
      "flow_matching_loss": 0.04673025272786617,
      "grad_norm": 0.10668029636144638,
      "h_stock_valid_norm": 95.08496704101563,
      "invalid_tail_samples": 22.0,
      "learning_rate": 8.588137447277503e-05,
      "loss": 0.0543,
      "positive_teacher_pairs": 27265.4,
      "query_h_norm_ratio": 0.4623686134815216,
      "relation_loss": 0.007542255148291588,
      "selected_pairs": 27265.4,
      "step": 8490,
      "student_distance_mean": 28.26146697998047,
      "teacher_distance_mean": 0.8520666539669037,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.007542255148291588,
      "weighted_relation_loss": 0.007542255148291588
    },
    {
      "action_query_batch_std": 0.41982183754444125,
      "action_query_norm": 43.9501449584961,
      "action_relation_loss": 0.007526654517278075,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.951356407857811,
      "flow_matching_loss": 0.04790439903736114,
      "grad_norm": 0.13052426278591156,
      "h_stock_valid_norm": 97.2033447265625,
      "invalid_tail_samples": 25.6,
      "learning_rate": 8.584296863805709e-05,
      "loss": 0.0554,
      "positive_teacher_pairs": 26437.1,
      "query_h_norm_ratio": 0.4521683603525162,
      "relation_loss": 0.007526654517278075,
      "selected_pairs": 26437.1,
      "step": 8500,
      "student_distance_mean": 28.018465423583983,
      "teacher_distance_mean": 0.8640937507152557,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.007526654517278075,
      "weighted_relation_loss": 0.007526654517278075
    },
    {
      "action_query_batch_std": 0.42009073197841645,
      "action_query_norm": 43.947077941894534,
      "action_relation_loss": 0.007661324040964246,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.960710944808232,
      "flow_matching_loss": 0.046523669734597206,
      "grad_norm": 0.12837959825992584,
      "h_stock_valid_norm": 98.81184997558594,
      "invalid_tail_samples": 22.3,
      "learning_rate": 8.580451925075248e-05,
      "loss": 0.0542,
      "positive_teacher_pairs": 27197.1,
      "query_h_norm_ratio": 0.44476069808006286,
      "relation_loss": 0.007661324040964246,
      "selected_pairs": 27197.1,
      "step": 8510,
      "student_distance_mean": 27.747192573547363,
      "teacher_distance_mean": 0.8637963175773621,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.007661324040964246,
      "weighted_relation_loss": 0.007661324040964246
    },
    {
      "action_query_batch_std": 0.4220236212015152,
      "action_query_norm": 43.95615463256836,
      "action_relation_loss": 0.0075988687574863436,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.9700654817586525,
      "flow_matching_loss": 0.04876883327960968,
      "grad_norm": 0.12357847392559052,
      "h_stock_valid_norm": 98.99312820434571,
      "invalid_tail_samples": 24.1,
      "learning_rate": 8.576602635758085e-05,
      "loss": 0.0564,
      "positive_teacher_pairs": 26778.9,
      "query_h_norm_ratio": 0.44403440356254575,
      "relation_loss": 0.0075988687574863436,
      "selected_pairs": 26778.9,
      "step": 8520,
      "student_distance_mean": 27.993226623535158,
      "teacher_distance_mean": 0.8484537422657012,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0075988687574863436,
      "weighted_relation_loss": 0.0075988687574863436
    },
    {
      "action_query_batch_std": 0.421994686126709,
      "action_query_norm": 43.95356369018555,
      "action_relation_loss": 0.007098163850605488,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.979420018709074,
      "flow_matching_loss": 0.046895507350564004,
      "grad_norm": 0.14316560328006744,
      "h_stock_valid_norm": 98.31989364624023,
      "invalid_tail_samples": 23.3,
      "learning_rate": 8.572749000531469e-05,
      "loss": 0.054,
      "positive_teacher_pairs": 26967.7,
      "query_h_norm_ratio": 0.4470578581094742,
      "relation_loss": 0.007098163850605488,
      "selected_pairs": 26967.7,
      "step": 8530,
      "student_distance_mean": 27.97937889099121,
      "teacher_distance_mean": 0.8490293800830842,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.007098163850605488,
      "weighted_relation_loss": 0.007098163850605488
    },
    {
      "action_query_batch_std": 0.42013837695121764,
      "action_query_norm": 43.942940139770506,
      "action_relation_loss": 0.007781144138425589,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.988774555659495,
      "flow_matching_loss": 0.04764540269970894,
      "grad_norm": 0.10698860138654709,
      "h_stock_valid_norm": 97.02477416992187,
      "invalid_tail_samples": 23.5,
      "learning_rate": 8.568891024077926e-05,
      "loss": 0.0554,
      "positive_teacher_pairs": 26921.0,
      "query_h_norm_ratio": 0.4529289036989212,
      "relation_loss": 0.007781144138425589,
      "selected_pairs": 26921.0,
      "step": 8540,
      "student_distance_mean": 27.968743896484376,
      "teacher_distance_mean": 0.8548388659954071,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.007781144138425589,
      "weighted_relation_loss": 0.007781144138425589
    },
    {
      "action_query_batch_std": 0.4238926380872726,
      "action_query_norm": 43.955112075805665,
      "action_relation_loss": 0.007192614162340761,
      "auxiliary_loss_weight": 1.0,
      "epoch": 7.998129092609916,
      "flow_matching_loss": 0.047683994844555856,
      "grad_norm": 0.13875488936901093,
      "h_stock_valid_norm": 96.26467666625976,
      "invalid_tail_samples": 21.6,
      "learning_rate": 8.565028711085265e-05,
      "loss": 0.0549,
      "positive_teacher_pairs": 27362.5,
      "query_h_norm_ratio": 0.4566106915473938,
      "relation_loss": 0.007192614162340761,
      "selected_pairs": 27362.5,
      "step": 8550,
      "student_distance_mean": 28.16718292236328,
      "teacher_distance_mean": 0.8515759885311127,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.007192614162340761,
      "weighted_relation_loss": 0.007192614162340761
    },
    {
      "action_query_batch_std": 0.42348997890949247,
      "action_query_norm": 43.94782905578613,
      "action_relation_loss": 0.006908567855134606,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.007483629560337,
      "flow_matching_loss": 0.04590841792523861,
      "grad_norm": 0.1362297534942627,
      "h_stock_valid_norm": 96.70370788574219,
      "invalid_tail_samples": 21.2,
      "learning_rate": 8.561162066246561e-05,
      "loss": 0.0528,
      "positive_teacher_pairs": 24460.0,
      "query_h_norm_ratio": 0.4544671833515167,
      "relation_loss": 0.006908567855134606,
      "selected_pairs": 24460.0,
      "step": 8560,
      "student_distance_mean": 28.088581466674803,
      "teacher_distance_mean": 0.8518855214118958,
      "valid_action_samples": 214.9,
      "weighted_auxiliary_loss": 0.006908567855134606,
      "weighted_relation_loss": 0.006908567855134606
    },
    {
      "action_query_batch_std": 0.4291206419467926,
      "action_query_norm": 43.9570255279541,
      "action_relation_loss": 0.007722982997074723,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.016838166510757,
      "flow_matching_loss": 0.047304991632699966,
      "grad_norm": 0.11695721745491028,
      "h_stock_valid_norm": 98.08285675048828,
      "invalid_tail_samples": 22.6,
      "learning_rate": 8.557291094260152e-05,
      "loss": 0.055,
      "positive_teacher_pairs": 27131.2,
      "query_h_norm_ratio": 0.44816834330558775,
      "relation_loss": 0.007722982997074723,
      "selected_pairs": 27131.2,
      "step": 8570,
      "student_distance_mean": 28.507128524780274,
      "teacher_distance_mean": 0.8629788517951965,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.007722982997074723,
      "weighted_relation_loss": 0.007722982997074723
    },
    {
      "action_query_batch_std": 0.42601925134658813,
      "action_query_norm": 43.956763076782224,
      "action_relation_loss": 0.007714546704664827,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.026192703461179,
      "flow_matching_loss": 0.04563991017639637,
      "grad_norm": 0.19080717861652374,
      "h_stock_valid_norm": 98.6210205078125,
      "invalid_tail_samples": 23.6,
      "learning_rate": 8.553415799829629e-05,
      "loss": 0.0534,
      "positive_teacher_pairs": 26905.3,
      "query_h_norm_ratio": 0.44571994841098783,
      "relation_loss": 0.007714546704664827,
      "selected_pairs": 26905.3,
      "step": 8580,
      "student_distance_mean": 28.413923454284667,
      "teacher_distance_mean": 0.8549827337265015,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.007714546704664827,
      "weighted_relation_loss": 0.007714546704664827
    },
    {
      "action_query_batch_std": 0.42336366772651673,
      "action_query_norm": 43.94032669067383,
      "action_relation_loss": 0.0071874796878546475,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.0355472404116,
      "flow_matching_loss": 0.04865385554730892,
      "grad_norm": 0.1939125806093216,
      "h_stock_valid_norm": 97.5484603881836,
      "invalid_tail_samples": 26.7,
      "learning_rate": 8.549536187663847e-05,
      "loss": 0.0558,
      "positive_teacher_pairs": 26179.7,
      "query_h_norm_ratio": 0.4504532188177109,
      "relation_loss": 0.0071874796878546475,
      "selected_pairs": 26179.7,
      "step": 8590,
      "student_distance_mean": 28.127886962890624,
      "teacher_distance_mean": 0.84943807721138,
      "valid_action_samples": 229.3,
      "weighted_auxiliary_loss": 0.0071874796878546475,
      "weighted_relation_loss": 0.0071874796878546475
    },
    {
      "action_query_batch_std": 0.41924806237220763,
      "action_query_norm": 43.93180694580078,
      "action_relation_loss": 0.007180757075548172,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.044901777362021,
      "flow_matching_loss": 0.04712548963725567,
      "grad_norm": 0.13399650156497955,
      "h_stock_valid_norm": 97.32887496948243,
      "invalid_tail_samples": 22.5,
      "learning_rate": 8.545652262476897e-05,
      "loss": 0.0543,
      "positive_teacher_pairs": 27154.8,
      "query_h_norm_ratio": 0.451384562253952,
      "relation_loss": 0.007180757075548172,
      "selected_pairs": 27154.8,
      "step": 8600,
      "student_distance_mean": 27.769563102722167,
      "teacher_distance_mean": 0.8434096276760101,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.007180757075548172,
      "weighted_relation_loss": 0.007180757075548172
    },
    {
      "action_query_batch_std": 0.42201185822486875,
      "action_query_norm": 43.94203681945801,
      "action_relation_loss": 0.007100494066253305,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.054256314312441,
      "flow_matching_loss": 0.045868707820773125,
      "grad_norm": 0.10537111014127731,
      "h_stock_valid_norm": 96.9633056640625,
      "invalid_tail_samples": 21.7,
      "learning_rate": 8.541764028988115e-05,
      "loss": 0.053,
      "positive_teacher_pairs": 27336.8,
      "query_h_norm_ratio": 0.45318941175937655,
      "relation_loss": 0.007100494066253305,
      "selected_pairs": 27336.8,
      "step": 8610,
      "student_distance_mean": 28.040625190734865,
      "teacher_distance_mean": 0.8477138936519623,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.007100494066253305,
      "weighted_relation_loss": 0.007100494066253305
    },
    {
      "action_query_batch_std": 0.4123802185058594,
      "action_query_norm": 43.916630554199216,
      "action_relation_loss": 0.007305454229936004,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.063610851262862,
      "flow_matching_loss": 0.04513353556394577,
      "grad_norm": 0.11943161487579346,
      "h_stock_valid_norm": 96.78743743896484,
      "invalid_tail_samples": 22.2,
      "learning_rate": 8.537871491922071e-05,
      "loss": 0.0524,
      "positive_teacher_pairs": 27225.2,
      "query_h_norm_ratio": 0.4537474036216736,
      "relation_loss": 0.007305454229936004,
      "selected_pairs": 27225.2,
      "step": 8620,
      "student_distance_mean": 27.401344871520998,
      "teacher_distance_mean": 0.843720281124115,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.007305454229936004,
      "weighted_relation_loss": 0.007305454229936004
    },
    {
      "action_query_batch_std": 0.4216314673423767,
      "action_query_norm": 43.93425025939941,
      "action_relation_loss": 0.007019047020003199,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.072965388213284,
      "flow_matching_loss": 0.046373509988188746,
      "grad_norm": 0.11469542235136032,
      "h_stock_valid_norm": 97.58631057739258,
      "invalid_tail_samples": 23.1,
      "learning_rate": 8.533974656008566e-05,
      "loss": 0.0534,
      "positive_teacher_pairs": 27011.4,
      "query_h_norm_ratio": 0.4502187818288803,
      "relation_loss": 0.007019047020003199,
      "selected_pairs": 27011.4,
      "step": 8630,
      "student_distance_mean": 28.091682815551756,
      "teacher_distance_mean": 0.8588936805725098,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.007019047020003199,
      "weighted_relation_loss": 0.007019047020003199
    },
    {
      "action_query_batch_std": 0.4234235048294067,
      "action_query_norm": 43.92993431091308,
      "action_relation_loss": 0.007034252537414432,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.082319925163704,
      "flow_matching_loss": 0.048586010187864306,
      "grad_norm": 0.1312694102525711,
      "h_stock_valid_norm": 98.93394088745117,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.530073525982621e-05,
      "loss": 0.0556,
      "positive_teacher_pairs": 26873.3,
      "query_h_norm_ratio": 0.44403491020202634,
      "relation_loss": 0.007034252537414432,
      "selected_pairs": 26873.3,
      "step": 8640,
      "student_distance_mean": 28.141394233703615,
      "teacher_distance_mean": 0.8544857263565063,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.007034252537414432,
      "weighted_relation_loss": 0.007034252537414432
    },
    {
      "action_query_batch_std": 0.41908576488494875,
      "action_query_norm": 43.92261848449707,
      "action_relation_loss": 0.006773356348276138,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.091674462114126,
      "flow_matching_loss": 0.0457873173058033,
      "grad_norm": 0.11499934643507004,
      "h_stock_valid_norm": 98.21416625976562,
      "invalid_tail_samples": 25.7,
      "learning_rate": 8.526168106584476e-05,
      "loss": 0.0526,
      "positive_teacher_pairs": 26410.9,
      "query_h_norm_ratio": 0.44722412526607513,
      "relation_loss": 0.006773356348276138,
      "selected_pairs": 26410.9,
      "step": 8650,
      "student_distance_mean": 27.840539741516114,
      "teacher_distance_mean": 0.8467743217945098,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.006773356348276138,
      "weighted_relation_loss": 0.006773356348276138
    },
    {
      "action_query_batch_std": 0.42278982102870943,
      "action_query_norm": 43.9349063873291,
      "action_relation_loss": 0.006929404754191637,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.101028999064546,
      "flow_matching_loss": 0.04609042592346668,
      "grad_norm": 0.15012235939502716,
      "h_stock_valid_norm": 96.43500061035157,
      "invalid_tail_samples": 26.5,
      "learning_rate": 8.522258402559586e-05,
      "loss": 0.053,
      "positive_teacher_pairs": 26229.1,
      "query_h_norm_ratio": 0.4556113541126251,
      "relation_loss": 0.006929404754191637,
      "selected_pairs": 26229.1,
      "step": 8660,
      "student_distance_mean": 28.049662017822264,
      "teacher_distance_mean": 0.8539003133773804,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.006929404754191637,
      "weighted_relation_loss": 0.006929404754191637
    },
    {
      "action_query_batch_std": 0.42506853938102723,
      "action_query_norm": 43.931037521362306,
      "action_relation_loss": 0.006968844449147582,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.110383536014968,
      "flow_matching_loss": 0.045113234966993335,
      "grad_norm": 0.10631553828716278,
      "h_stock_valid_norm": 95.60526809692382,
      "invalid_tail_samples": 24.3,
      "learning_rate": 8.518344418658613e-05,
      "loss": 0.0521,
      "positive_teacher_pairs": 26738.3,
      "query_h_norm_ratio": 0.45950846970081327,
      "relation_loss": 0.006968844449147582,
      "selected_pairs": 26738.3,
      "step": 8670,
      "student_distance_mean": 28.273645973205568,
      "teacher_distance_mean": 0.8648101568222046,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.006968844449147582,
      "weighted_relation_loss": 0.006968844449147582
    },
    {
      "action_query_batch_std": 0.4191784143447876,
      "action_query_norm": 43.9197883605957,
      "action_relation_loss": 0.007150169275701046,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.119738072965388,
      "flow_matching_loss": 0.042525756359100345,
      "grad_norm": 0.14525383710861206,
      "h_stock_valid_norm": 96.17382049560547,
      "invalid_tail_samples": 22.9,
      "learning_rate": 8.51442615963741e-05,
      "loss": 0.0497,
      "positive_teacher_pairs": 27067.9,
      "query_h_norm_ratio": 0.4566741347312927,
      "relation_loss": 0.007150169275701046,
      "selected_pairs": 27067.9,
      "step": 8680,
      "student_distance_mean": 27.769123840332032,
      "teacher_distance_mean": 0.8526050269603729,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.007150169275701046,
      "weighted_relation_loss": 0.007150169275701046
    },
    {
      "action_query_batch_std": 0.4205889642238617,
      "action_query_norm": 43.9195972442627,
      "action_relation_loss": 0.006836267141625285,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.129092609915809,
      "flow_matching_loss": 0.04451309479773045,
      "grad_norm": 0.10569857805967331,
      "h_stock_valid_norm": 97.43444213867187,
      "invalid_tail_samples": 22.0,
      "learning_rate": 8.510503630257034e-05,
      "loss": 0.0513,
      "positive_teacher_pairs": 27263.8,
      "query_h_norm_ratio": 0.45077510476112365,
      "relation_loss": 0.006836267141625285,
      "selected_pairs": 27263.8,
      "step": 8690,
      "student_distance_mean": 28.028731536865234,
      "teacher_distance_mean": 0.8489628374576569,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.006836267141625285,
      "weighted_relation_loss": 0.006836267141625285
    },
    {
      "action_query_batch_std": 0.4269349277019501,
      "action_query_norm": 43.93198089599609,
      "action_relation_loss": 0.00687260264530778,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.13844714686623,
      "flow_matching_loss": 0.045883318036794664,
      "grad_norm": 0.15078219771385193,
      "h_stock_valid_norm": 97.82791900634766,
      "invalid_tail_samples": 22.4,
      "learning_rate": 8.50657683528373e-05,
      "loss": 0.0528,
      "positive_teacher_pairs": 27174.1,
      "query_h_norm_ratio": 0.4490806609392166,
      "relation_loss": 0.00687260264530778,
      "selected_pairs": 27174.1,
      "step": 8700,
      "student_distance_mean": 28.31458511352539,
      "teacher_distance_mean": 0.8669259250164032,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.00687260264530778,
      "weighted_relation_loss": 0.00687260264530778
    },
    {
      "action_query_batch_std": 0.4277936011552811,
      "action_query_norm": 43.93000106811523,
      "action_relation_loss": 0.007253555441275239,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.14780168381665,
      "flow_matching_loss": 0.0480208620429039,
      "grad_norm": 0.1560758501291275,
      "h_stock_valid_norm": 97.65871353149414,
      "invalid_tail_samples": 22.9,
      "learning_rate": 8.502645779488922e-05,
      "loss": 0.0553,
      "positive_teacher_pairs": 27053.9,
      "query_h_norm_ratio": 0.4498364806175232,
      "relation_loss": 0.007253555441275239,
      "selected_pairs": 27053.9,
      "step": 8710,
      "student_distance_mean": 28.401187896728516,
      "teacher_distance_mean": 0.8687021851539611,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.007253555441275239,
      "weighted_relation_loss": 0.007253555441275239
    },
    {
      "action_query_batch_std": 0.4212217181921005,
      "action_query_norm": 43.91743698120117,
      "action_relation_loss": 0.007075955299660564,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.157156220767073,
      "flow_matching_loss": 0.0478284765034914,
      "grad_norm": 0.1506224125623703,
      "h_stock_valid_norm": 97.52448043823242,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.498710467649213e-05,
      "loss": 0.0549,
      "positive_teacher_pairs": 26684.5,
      "query_h_norm_ratio": 0.4503260225057602,
      "relation_loss": 0.007075955299660564,
      "selected_pairs": 26684.5,
      "step": 8720,
      "student_distance_mean": 27.958683967590332,
      "teacher_distance_mean": 0.8560765027999878,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.007075955299660564,
      "weighted_relation_loss": 0.007075955299660564
    },
    {
      "action_query_batch_std": 0.42106653451919557,
      "action_query_norm": 43.91418991088867,
      "action_relation_loss": 0.007459052745252848,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.166510757717493,
      "flow_matching_loss": 0.04665656313300133,
      "grad_norm": 0.12586797773838043,
      "h_stock_valid_norm": 97.0181983947754,
      "invalid_tail_samples": 22.4,
      "learning_rate": 8.494770904546381e-05,
      "loss": 0.0541,
      "positive_teacher_pairs": 27179.5,
      "query_h_norm_ratio": 0.45264374315738676,
      "relation_loss": 0.007459052745252848,
      "selected_pairs": 27179.5,
      "step": 8730,
      "student_distance_mean": 27.8693603515625,
      "teacher_distance_mean": 0.8550914883613586,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.007459052745252848,
      "weighted_relation_loss": 0.007459052745252848
    },
    {
      "action_query_batch_std": 0.4156854510307312,
      "action_query_norm": 43.90176086425781,
      "action_relation_loss": 0.00708808396011591,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.175865294667913,
      "flow_matching_loss": 0.04668034724891186,
      "grad_norm": 0.1318303793668747,
      "h_stock_valid_norm": 97.00935287475586,
      "invalid_tail_samples": 24.7,
      "learning_rate": 8.490827094967363e-05,
      "loss": 0.0538,
      "positive_teacher_pairs": 26640.7,
      "query_h_norm_ratio": 0.4525542020797729,
      "relation_loss": 0.00708808396011591,
      "selected_pairs": 26640.7,
      "step": 8740,
      "student_distance_mean": 27.61294231414795,
      "teacher_distance_mean": 0.8538088619709014,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.00708808396011591,
      "weighted_relation_loss": 0.00708808396011591
    },
    {
      "action_query_batch_std": 0.4351051151752472,
      "action_query_norm": 43.9469409942627,
      "action_relation_loss": 0.0067407948896288875,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.185219831618335,
      "flow_matching_loss": 0.045781269669532776,
      "grad_norm": 0.12068509310483932,
      "h_stock_valid_norm": 96.72054977416992,
      "invalid_tail_samples": 23.5,
      "learning_rate": 8.486879043704263e-05,
      "loss": 0.0525,
      "positive_teacher_pairs": 26914.1,
      "query_h_norm_ratio": 0.4543731361627579,
      "relation_loss": 0.0067407948896288875,
      "selected_pairs": 26914.1,
      "step": 8750,
      "student_distance_mean": 28.83966236114502,
      "teacher_distance_mean": 0.8800650715827942,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0067407948896288875,
      "weighted_relation_loss": 0.0067407948896288875
    },
    {
      "action_query_batch_std": 0.4252243012189865,
      "action_query_norm": 43.91037139892578,
      "action_relation_loss": 0.006938811251893639,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.194574368568755,
      "flow_matching_loss": 0.04664055109024048,
      "grad_norm": 0.13820792734622955,
      "h_stock_valid_norm": 97.61003875732422,
      "invalid_tail_samples": 24.8,
      "learning_rate": 8.482926755554332e-05,
      "loss": 0.0536,
      "positive_teacher_pairs": 26615.2,
      "query_h_norm_ratio": 0.44985887706279754,
      "relation_loss": 0.006938811251893639,
      "selected_pairs": 26615.2,
      "step": 8760,
      "student_distance_mean": 28.173834228515624,
      "teacher_distance_mean": 0.8555474162101746,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.006938811251893639,
      "weighted_relation_loss": 0.006938811251893639
    },
    {
      "action_query_batch_std": 0.411721533536911,
      "action_query_norm": 43.88929901123047,
      "action_relation_loss": 0.007159694703295827,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.203928905519177,
      "flow_matching_loss": 0.046215930953621864,
      "grad_norm": 0.12319021672010422,
      "h_stock_valid_norm": 97.93996810913086,
      "invalid_tail_samples": 24.2,
      "learning_rate": 8.478970235319975e-05,
      "loss": 0.0534,
      "positive_teacher_pairs": 26764.3,
      "query_h_norm_ratio": 0.4481289625167847,
      "relation_loss": 0.007159694703295827,
      "selected_pairs": 26764.3,
      "step": 8770,
      "student_distance_mean": 27.22636432647705,
      "teacher_distance_mean": 0.836435729265213,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.007159694703295827,
      "weighted_relation_loss": 0.007159694703295827
    },
    {
      "action_query_batch_std": 0.41872560381889345,
      "action_query_norm": 43.91015853881836,
      "action_relation_loss": 0.006952118640765548,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.213283442469598,
      "flow_matching_loss": 0.04361160509288311,
      "grad_norm": 0.118682362139225,
      "h_stock_valid_norm": 98.79817810058594,
      "invalid_tail_samples": 24.8,
      "learning_rate": 8.475009487808737e-05,
      "loss": 0.0506,
      "positive_teacher_pairs": 26619.6,
      "query_h_norm_ratio": 0.4444511830806732,
      "relation_loss": 0.006952118640765548,
      "selected_pairs": 26619.6,
      "step": 8780,
      "student_distance_mean": 27.931583595275878,
      "teacher_distance_mean": 0.8552746415138245,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.006952118640765548,
      "weighted_relation_loss": 0.006952118640765548
    },
    {
      "action_query_batch_std": 0.41660635471343993,
      "action_query_norm": 43.902377700805665,
      "action_relation_loss": 0.007239822018891573,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.22263797942002,
      "flow_matching_loss": 0.044477829709649086,
      "grad_norm": 0.1412964016199112,
      "h_stock_valid_norm": 98.89350280761718,
      "invalid_tail_samples": 23.0,
      "learning_rate": 8.471044517833298e-05,
      "loss": 0.0517,
      "positive_teacher_pairs": 27043.0,
      "query_h_norm_ratio": 0.4439403086900711,
      "relation_loss": 0.007239822018891573,
      "selected_pairs": 27043.0,
      "step": 8790,
      "student_distance_mean": 27.621064949035645,
      "teacher_distance_mean": 0.8426782727241516,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.007239822018891573,
      "weighted_relation_loss": 0.007239822018891573
    },
    {
      "action_query_batch_std": 0.4161278009414673,
      "action_query_norm": 43.889519119262694,
      "action_relation_loss": 0.007204625755548477,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.23199251637044,
      "flow_matching_loss": 0.0435048907995224,
      "grad_norm": 0.1045118123292923,
      "h_stock_valid_norm": 97.88548049926757,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.467075330211474e-05,
      "loss": 0.0507,
      "positive_teacher_pairs": 26878.7,
      "query_h_norm_ratio": 0.44838384687900545,
      "relation_loss": 0.007204625755548477,
      "selected_pairs": 26878.7,
      "step": 8800,
      "student_distance_mean": 27.57810115814209,
      "teacher_distance_mean": 0.8576909244060517,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.007204625755548477,
      "weighted_relation_loss": 0.007204625755548477
    },
    {
      "action_query_batch_std": 0.4171852707862854,
      "action_query_norm": 43.89381332397461,
      "action_relation_loss": 0.007282237196341157,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.24134705332086,
      "flow_matching_loss": 0.04532815255224705,
      "grad_norm": 0.12229806929826736,
      "h_stock_valid_norm": 97.43988494873047,
      "invalid_tail_samples": 22.7,
      "learning_rate": 8.463101929766198e-05,
      "loss": 0.0526,
      "positive_teacher_pairs": 27102.2,
      "query_h_norm_ratio": 0.4504745423793793,
      "relation_loss": 0.007282237196341157,
      "selected_pairs": 27102.2,
      "step": 8810,
      "student_distance_mean": 27.608507347106933,
      "teacher_distance_mean": 0.8528629004955292,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.007282237196341157,
      "weighted_relation_loss": 0.007282237196341157
    },
    {
      "action_query_batch_std": 0.4185037583112717,
      "action_query_norm": 43.89755706787109,
      "action_relation_loss": 0.007056148536503315,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.250701590271282,
      "flow_matching_loss": 0.04649025611579418,
      "grad_norm": 0.112484872341156,
      "h_stock_valid_norm": 96.87101440429687,
      "invalid_tail_samples": 25.6,
      "learning_rate": 8.459124321325528e-05,
      "loss": 0.0535,
      "positive_teacher_pairs": 26433.8,
      "query_h_norm_ratio": 0.4531586438417435,
      "relation_loss": 0.007056148536503315,
      "selected_pairs": 26433.8,
      "step": 8820,
      "student_distance_mean": 27.88565158843994,
      "teacher_distance_mean": 0.8579978287220001,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.007056148536503315,
      "weighted_relation_loss": 0.007056148536503315
    },
    {
      "action_query_batch_std": 0.4176113188266754,
      "action_query_norm": 43.89223670959473,
      "action_relation_loss": 0.006915962602943182,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.260056127221702,
      "flow_matching_loss": 0.046006011217832564,
      "grad_norm": 0.11399848014116287,
      "h_stock_valid_norm": 96.35373458862304,
      "invalid_tail_samples": 23.3,
      "learning_rate": 8.455142509722634e-05,
      "loss": 0.0529,
      "positive_teacher_pairs": 26970.5,
      "query_h_norm_ratio": 0.45553478598594666,
      "relation_loss": 0.006915962602943182,
      "selected_pairs": 26970.5,
      "step": 8830,
      "student_distance_mean": 27.84008159637451,
      "teacher_distance_mean": 0.8579325318336487,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.006915962602943182,
      "weighted_relation_loss": 0.006915962602943182
    },
    {
      "action_query_batch_std": 0.42243653535842896,
      "action_query_norm": 43.90356674194336,
      "action_relation_loss": 0.0070041110273450615,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.269410664172124,
      "flow_matching_loss": 0.04540826342999935,
      "grad_norm": 0.19621285796165466,
      "h_stock_valid_norm": 96.78793106079101,
      "invalid_tail_samples": 23.9,
      "learning_rate": 8.451156499795792e-05,
      "loss": 0.0524,
      "positive_teacher_pairs": 26829.3,
      "query_h_norm_ratio": 0.45360914766788485,
      "relation_loss": 0.0070041110273450615,
      "selected_pairs": 26829.3,
      "step": 8840,
      "student_distance_mean": 27.94754066467285,
      "teacher_distance_mean": 0.8635756433010101,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0070041110273450615,
      "weighted_relation_loss": 0.0070041110273450615
    },
    {
      "action_query_batch_std": 0.4160458117723465,
      "action_query_norm": 43.89172897338867,
      "action_relation_loss": 0.007231745636090636,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.278765201122544,
      "flow_matching_loss": 0.04638017863035202,
      "grad_norm": 0.1424795538187027,
      "h_stock_valid_norm": 97.82511672973632,
      "invalid_tail_samples": 26.0,
      "learning_rate": 8.447166296388381e-05,
      "loss": 0.0536,
      "positive_teacher_pairs": 26342.9,
      "query_h_norm_ratio": 0.4486831396818161,
      "relation_loss": 0.007231745636090636,
      "selected_pairs": 26342.9,
      "step": 8850,
      "student_distance_mean": 27.780608558654784,
      "teacher_distance_mean": 0.8567080676555634,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.007231745636090636,
      "weighted_relation_loss": 0.007231745636090636
    },
    {
      "action_query_batch_std": 0.4183528870344162,
      "action_query_norm": 43.88261833190918,
      "action_relation_loss": 0.007374684046953917,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.288119738072965,
      "flow_matching_loss": 0.04792109467089176,
      "grad_norm": 0.12616151571273804,
      "h_stock_valid_norm": 97.2877296447754,
      "invalid_tail_samples": 21.6,
      "learning_rate": 8.443171904348872e-05,
      "loss": 0.0553,
      "positive_teacher_pairs": 27356.5,
      "query_h_norm_ratio": 0.4510636508464813,
      "relation_loss": 0.007374684046953917,
      "selected_pairs": 27356.5,
      "step": 8860,
      "student_distance_mean": 27.775785827636717,
      "teacher_distance_mean": 0.8495571970939636,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.007374684046953917,
      "weighted_relation_loss": 0.007374684046953917
    },
    {
      "action_query_batch_std": 0.42049930691719056,
      "action_query_norm": 43.8871883392334,
      "action_relation_loss": 0.006957215908914805,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.297474275023387,
      "flow_matching_loss": 0.04594536423683167,
      "grad_norm": 0.1448177844285965,
      "h_stock_valid_norm": 97.08489379882812,
      "invalid_tail_samples": 22.3,
      "learning_rate": 8.43917332853083e-05,
      "loss": 0.0529,
      "positive_teacher_pairs": 27200.2,
      "query_h_norm_ratio": 0.45206101834774015,
      "relation_loss": 0.006957215908914805,
      "selected_pairs": 27200.2,
      "step": 8870,
      "student_distance_mean": 27.84167308807373,
      "teacher_distance_mean": 0.8558946967124939,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.006957215908914805,
      "weighted_relation_loss": 0.006957215908914805
    },
    {
      "action_query_batch_std": 0.41972306072711946,
      "action_query_norm": 43.89923095703125,
      "action_relation_loss": 0.006930726300925017,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.306828811973807,
      "flow_matching_loss": 0.04373471811413765,
      "grad_norm": 0.13485394418239594,
      "h_stock_valid_norm": 97.88751678466797,
      "invalid_tail_samples": 24.0,
      "learning_rate": 8.435170573792901e-05,
      "loss": 0.0507,
      "positive_teacher_pairs": 26805.0,
      "query_h_norm_ratio": 0.44846988916397096,
      "relation_loss": 0.006930726300925017,
      "selected_pairs": 26805.0,
      "step": 8880,
      "student_distance_mean": 27.830104255676268,
      "teacher_distance_mean": 0.8562002539634704,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.006930726300925017,
      "weighted_relation_loss": 0.006930726300925017
    },
    {
      "action_query_batch_std": 0.41267602145671844,
      "action_query_norm": 43.8775390625,
      "action_relation_loss": 0.007104457914829254,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.316183348924229,
      "flow_matching_loss": 0.04631451815366745,
      "grad_norm": 0.12582489848136902,
      "h_stock_valid_norm": 98.24256057739258,
      "invalid_tail_samples": 22.6,
      "learning_rate": 8.431163644998808e-05,
      "loss": 0.0534,
      "positive_teacher_pairs": 27124.4,
      "query_h_norm_ratio": 0.4466305375099182,
      "relation_loss": 0.007104457914829254,
      "selected_pairs": 27124.4,
      "step": 8890,
      "student_distance_mean": 27.429419708251952,
      "teacher_distance_mean": 0.8497036218643188,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.007104457914829254,
      "weighted_relation_loss": 0.007104457914829254
    },
    {
      "action_query_batch_std": 0.4178241968154907,
      "action_query_norm": 43.88898277282715,
      "action_relation_loss": 0.0071935686748474835,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.325537885874649,
      "flow_matching_loss": 0.04569524228572845,
      "grad_norm": 0.13658227026462555,
      "h_stock_valid_norm": 97.28395233154296,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.42715254701735e-05,
      "loss": 0.0529,
      "positive_teacher_pairs": 26869.0,
      "query_h_norm_ratio": 0.45115831196308137,
      "relation_loss": 0.0071935686748474835,
      "selected_pairs": 26869.0,
      "step": 8900,
      "student_distance_mean": 27.560445022583007,
      "teacher_distance_mean": 0.85209139585495,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0071935686748474835,
      "weighted_relation_loss": 0.0071935686748474835
    },
    {
      "action_query_batch_std": 0.42320042848587036,
      "action_query_norm": 43.89022026062012,
      "action_relation_loss": 0.007080347836017608,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.334892422825071,
      "flow_matching_loss": 0.04749860353767872,
      "grad_norm": 0.14671286940574646,
      "h_stock_valid_norm": 96.55368347167969,
      "invalid_tail_samples": 21.8,
      "learning_rate": 8.42313728472239e-05,
      "loss": 0.0546,
      "positive_teacher_pairs": 27316.6,
      "query_h_norm_ratio": 0.45456973314285276,
      "relation_loss": 0.007080347836017608,
      "selected_pairs": 27316.6,
      "step": 8910,
      "student_distance_mean": 28.056501579284667,
      "teacher_distance_mean": 0.8652068138122558,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.007080347836017608,
      "weighted_relation_loss": 0.007080347836017608
    },
    {
      "action_query_batch_std": 0.4159750998020172,
      "action_query_norm": 43.87538070678711,
      "action_relation_loss": 0.0073629400692880155,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.344246959775491,
      "flow_matching_loss": 0.04629280008375645,
      "grad_norm": 0.15828070044517517,
      "h_stock_valid_norm": 96.8961166381836,
      "invalid_tail_samples": 23.6,
      "learning_rate": 8.419117862992844e-05,
      "loss": 0.0537,
      "positive_teacher_pairs": 26898.3,
      "query_h_norm_ratio": 0.45281524658203126,
      "relation_loss": 0.0073629400692880155,
      "selected_pairs": 26898.3,
      "step": 8920,
      "student_distance_mean": 27.54950885772705,
      "teacher_distance_mean": 0.8532455027103424,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0073629400692880155,
      "weighted_relation_loss": 0.0073629400692880155
    },
    {
      "action_query_batch_std": 0.41954638063907623,
      "action_query_norm": 43.88972969055176,
      "action_relation_loss": 0.007116616144776345,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.353601496725911,
      "flow_matching_loss": 0.04525823928415775,
      "grad_norm": 0.16367053985595703,
      "h_stock_valid_norm": 97.36239395141601,
      "invalid_tail_samples": 25.4,
      "learning_rate": 8.415094286712694e-05,
      "loss": 0.0524,
      "positive_teacher_pairs": 26485.4,
      "query_h_norm_ratio": 0.45079152286052704,
      "relation_loss": 0.007116616144776345,
      "selected_pairs": 26485.4,
      "step": 8930,
      "student_distance_mean": 27.905140495300294,
      "teacher_distance_mean": 0.8579631745815277,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.007116616144776345,
      "weighted_relation_loss": 0.007116616144776345
    },
    {
      "action_query_batch_std": 0.42058715522289275,
      "action_query_norm": 43.89367027282715,
      "action_relation_loss": 0.0069723965134471655,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.362956033676333,
      "flow_matching_loss": 0.043621654808521274,
      "grad_norm": 0.13437819480895996,
      "h_stock_valid_norm": 97.65973205566407,
      "invalid_tail_samples": 24.2,
      "learning_rate": 8.411066560770965e-05,
      "loss": 0.0506,
      "positive_teacher_pairs": 26767.9,
      "query_h_norm_ratio": 0.4494582086801529,
      "relation_loss": 0.0069723965134471655,
      "selected_pairs": 26767.9,
      "step": 8940,
      "student_distance_mean": 28.044933128356934,
      "teacher_distance_mean": 0.8607928276062011,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0069723965134471655,
      "weighted_relation_loss": 0.0069723965134471655
    },
    {
      "action_query_batch_std": 0.4180634945631027,
      "action_query_norm": 43.882733535766604,
      "action_relation_loss": 0.006872535916045308,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.372310570626754,
      "flow_matching_loss": 0.04461361020803452,
      "grad_norm": 0.12797822058200836,
      "h_stock_valid_norm": 97.87904891967773,
      "invalid_tail_samples": 25.4,
      "learning_rate": 8.407034690061721e-05,
      "loss": 0.0515,
      "positive_teacher_pairs": 26492.7,
      "query_h_norm_ratio": 0.4483386754989624,
      "relation_loss": 0.006872535916045308,
      "selected_pairs": 26492.7,
      "step": 8950,
      "student_distance_mean": 27.793812561035157,
      "teacher_distance_mean": 0.858334481716156,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.006872535916045308,
      "weighted_relation_loss": 0.006872535916045308
    },
    {
      "action_query_batch_std": 0.4104821801185608,
      "action_query_norm": 43.86969566345215,
      "action_relation_loss": 0.007369797211140394,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.381665107577176,
      "flow_matching_loss": 0.04501472003757954,
      "grad_norm": 0.09943342208862305,
      "h_stock_valid_norm": 97.32910995483398,
      "invalid_tail_samples": 20.6,
      "learning_rate": 8.402998679484067e-05,
      "loss": 0.0524,
      "positive_teacher_pairs": 27612.4,
      "query_h_norm_ratio": 0.4507427304983139,
      "relation_loss": 0.007369797211140394,
      "selected_pairs": 27612.4,
      "step": 8960,
      "student_distance_mean": 27.25452709197998,
      "teacher_distance_mean": 0.8451368391513825,
      "valid_action_samples": 235.4,
      "weighted_auxiliary_loss": 0.007369797211140394,
      "weighted_relation_loss": 0.007369797211140394
    },
    {
      "action_query_batch_std": 0.412656182050705,
      "action_query_norm": 43.865051651000975,
      "action_relation_loss": 0.007018161844462157,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.391019644527596,
      "flow_matching_loss": 0.043654027953743936,
      "grad_norm": 0.11020374298095703,
      "h_stock_valid_norm": 97.37557830810547,
      "invalid_tail_samples": 23.8,
      "learning_rate": 8.398958533942135e-05,
      "loss": 0.0507,
      "positive_teacher_pairs": 26857.1,
      "query_h_norm_ratio": 0.4504767656326294,
      "relation_loss": 0.007018161844462157,
      "selected_pairs": 26857.1,
      "step": 8970,
      "student_distance_mean": 27.384510612487794,
      "teacher_distance_mean": 0.8501461029052735,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.007018161844462157,
      "weighted_relation_loss": 0.007018161844462157
    },
    {
      "action_query_batch_std": 0.41720463037490846,
      "action_query_norm": 43.878013610839844,
      "action_relation_loss": 0.006840241327881813,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.400374181478018,
      "flow_matching_loss": 0.04345819056034088,
      "grad_norm": 0.10741280764341354,
      "h_stock_valid_norm": 97.58636245727538,
      "invalid_tail_samples": 26.6,
      "learning_rate": 8.394914258345083e-05,
      "loss": 0.0503,
      "positive_teacher_pairs": 26212.4,
      "query_h_norm_ratio": 0.44963842928409575,
      "relation_loss": 0.006840241327881813,
      "selected_pairs": 26212.4,
      "step": 8980,
      "student_distance_mean": 27.724877548217773,
      "teacher_distance_mean": 0.8493151783943176,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.006840241327881813,
      "weighted_relation_loss": 0.006840241327881813
    },
    {
      "action_query_batch_std": 0.4181339621543884,
      "action_query_norm": 43.88251647949219,
      "action_relation_loss": 0.007040696404874325,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.409728718428438,
      "flow_matching_loss": 0.04480618052184582,
      "grad_norm": 0.11430953443050385,
      "h_stock_valid_norm": 97.61528625488282,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.390865857607089e-05,
      "loss": 0.0518,
      "positive_teacher_pairs": 26691.9,
      "query_h_norm_ratio": 0.4495487570762634,
      "relation_loss": 0.007040696404874325,
      "selected_pairs": 26691.9,
      "step": 8990,
      "student_distance_mean": 27.917364501953124,
      "teacher_distance_mean": 0.8573441565036773,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.007040696404874325,
      "weighted_relation_loss": 0.007040696404874325
    },
    {
      "action_query_batch_std": 0.4160723388195038,
      "action_query_norm": 43.871364974975585,
      "action_relation_loss": 0.006977333361282945,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.419083255378858,
      "flow_matching_loss": 0.04735618755221367,
      "grad_norm": 0.11871238797903061,
      "h_stock_valid_norm": 97.51914672851562,
      "invalid_tail_samples": 24.9,
      "learning_rate": 8.386813336647339e-05,
      "loss": 0.0543,
      "positive_teacher_pairs": 26603.5,
      "query_h_norm_ratio": 0.44987890124320984,
      "relation_loss": 0.006977333361282945,
      "selected_pairs": 26603.5,
      "step": 9000,
      "student_distance_mean": 27.56922836303711,
      "teacher_distance_mean": 0.8536613285541534,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.006977333361282945,
      "weighted_relation_loss": 0.006977333361282945
    },
    {
      "action_query_batch_std": 0.4125144243240356,
      "action_query_norm": 43.8609561920166,
      "action_relation_loss": 0.0067899477202445265,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.42843779232928,
      "flow_matching_loss": 0.04534907564520836,
      "grad_norm": 0.1360165774822235,
      "h_stock_valid_norm": 97.2429443359375,
      "invalid_tail_samples": 24.7,
      "learning_rate": 8.38275670039003e-05,
      "loss": 0.0521,
      "positive_teacher_pairs": 26645.4,
      "query_h_norm_ratio": 0.4510488897562027,
      "relation_loss": 0.0067899477202445265,
      "selected_pairs": 26645.4,
      "step": 9010,
      "student_distance_mean": 27.351578521728516,
      "teacher_distance_mean": 0.8508257865905762,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0067899477202445265,
      "weighted_relation_loss": 0.0067899477202445265
    },
    {
      "action_query_batch_std": 0.41386868953704836,
      "action_query_norm": 43.85934181213379,
      "action_relation_loss": 0.00671639977954328,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.4377923292797,
      "flow_matching_loss": 0.044404400140047075,
      "grad_norm": 0.1295713186264038,
      "h_stock_valid_norm": 96.68522262573242,
      "invalid_tail_samples": 24.1,
      "learning_rate": 8.378695953764356e-05,
      "loss": 0.0511,
      "positive_teacher_pairs": 26778.5,
      "query_h_norm_ratio": 0.453633588552475,
      "relation_loss": 0.00671639977954328,
      "selected_pairs": 26778.5,
      "step": 9020,
      "student_distance_mean": 27.480048370361327,
      "teacher_distance_mean": 0.8536748230457306,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.00671639977954328,
      "weighted_relation_loss": 0.00671639977954328
    },
    {
      "action_query_batch_std": 0.4178889811038971,
      "action_query_norm": 43.866091918945315,
      "action_relation_loss": 0.0068694828543812035,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.447146866230122,
      "flow_matching_loss": 0.0446570910513401,
      "grad_norm": 0.14106936752796173,
      "h_stock_valid_norm": 97.29557342529297,
      "invalid_tail_samples": 27.2,
      "learning_rate": 8.374631101704509e-05,
      "loss": 0.0515,
      "positive_teacher_pairs": 26071.9,
      "query_h_norm_ratio": 0.4508617162704468,
      "relation_loss": 0.0068694828543812035,
      "selected_pairs": 26071.9,
      "step": 9030,
      "student_distance_mean": 27.804838180541992,
      "teacher_distance_mean": 0.8642367660999298,
      "valid_action_samples": 228.8,
      "weighted_auxiliary_loss": 0.0068694828543812035,
      "weighted_relation_loss": 0.0068694828543812035
    },
    {
      "action_query_batch_std": 0.41159883737564085,
      "action_query_norm": 43.847465133666994,
      "action_relation_loss": 0.007153515564277768,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.456501403180543,
      "flow_matching_loss": 0.044596743583679196,
      "grad_norm": 0.2078017145395279,
      "h_stock_valid_norm": 97.74654769897461,
      "invalid_tail_samples": 24.6,
      "learning_rate": 8.370562149149665e-05,
      "loss": 0.0518,
      "positive_teacher_pairs": 26664.1,
      "query_h_norm_ratio": 0.4485902369022369,
      "relation_loss": 0.007153515564277768,
      "selected_pairs": 26664.1,
      "step": 9040,
      "student_distance_mean": 27.3637638092041,
      "teacher_distance_mean": 0.8568119704723358,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.007153515564277768,
      "weighted_relation_loss": 0.007153515564277768
    },
    {
      "action_query_batch_std": 0.41113002598285675,
      "action_query_norm": 43.846181106567386,
      "action_relation_loss": 0.0069810108747333285,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.465855940130963,
      "flow_matching_loss": 0.04489595629274845,
      "grad_norm": 0.16730694472789764,
      "h_stock_valid_norm": 98.39652862548829,
      "invalid_tail_samples": 23.5,
      "learning_rate": 8.366489101043989e-05,
      "loss": 0.0519,
      "positive_teacher_pairs": 26919.4,
      "query_h_norm_ratio": 0.4456131845712662,
      "relation_loss": 0.0069810108747333285,
      "selected_pairs": 26919.4,
      "step": 9050,
      "student_distance_mean": 27.39779052734375,
      "teacher_distance_mean": 0.8533543348312378,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0069810108747333285,
      "weighted_relation_loss": 0.0069810108747333285
    },
    {
      "action_query_batch_std": 0.4210167437791824,
      "action_query_norm": 43.86803283691406,
      "action_relation_loss": 0.006769807264208794,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.475210477081385,
      "flow_matching_loss": 0.04479452706873417,
      "grad_norm": 0.14452001452445984,
      "h_stock_valid_norm": 98.48345184326172,
      "invalid_tail_samples": 25.0,
      "learning_rate": 8.362411962336614e-05,
      "loss": 0.0516,
      "positive_teacher_pairs": 26577.9,
      "query_h_norm_ratio": 0.4454420506954193,
      "relation_loss": 0.006769807264208794,
      "selected_pairs": 26577.9,
      "step": 9060,
      "student_distance_mean": 28.090895080566405,
      "teacher_distance_mean": 0.8750076413154602,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.006769807264208794,
      "weighted_relation_loss": 0.006769807264208794
    },
    {
      "action_query_batch_std": 0.40855253040790557,
      "action_query_norm": 43.836158752441406,
      "action_relation_loss": 0.007205093884840608,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.484565014031805,
      "flow_matching_loss": 0.04884595237672329,
      "grad_norm": 0.1536255031824112,
      "h_stock_valid_norm": 97.79092941284179,
      "invalid_tail_samples": 25.0,
      "learning_rate": 8.358330737981651e-05,
      "loss": 0.0561,
      "positive_teacher_pairs": 26577.8,
      "query_h_norm_ratio": 0.44826812744140626,
      "relation_loss": 0.007205093884840608,
      "selected_pairs": 26577.8,
      "step": 9070,
      "student_distance_mean": 27.223252487182616,
      "teacher_distance_mean": 0.8519914329051972,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.007205093884840608,
      "weighted_relation_loss": 0.007205093884840608
    },
    {
      "action_query_batch_std": 0.41144677996635437,
      "action_query_norm": 43.84132347106934,
      "action_relation_loss": 0.006975344708189368,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.493919550982227,
      "flow_matching_loss": 0.04906584620475769,
      "grad_norm": 0.13465608656406403,
      "h_stock_valid_norm": 98.30850067138672,
      "invalid_tail_samples": 22.6,
      "learning_rate": 8.35424543293817e-05,
      "loss": 0.056,
      "positive_teacher_pairs": 27126.9,
      "query_h_norm_ratio": 0.44596792161464693,
      "relation_loss": 0.006975344708189368,
      "selected_pairs": 27126.9,
      "step": 9080,
      "student_distance_mean": 27.345273780822755,
      "teacher_distance_mean": 0.8450584411621094,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.006975344708189368,
      "weighted_relation_loss": 0.006975344708189368
    },
    {
      "action_query_batch_std": 0.4151453673839569,
      "action_query_norm": 43.857187271118164,
      "action_relation_loss": 0.0066034491173923016,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.503274087932647,
      "flow_matching_loss": 0.04550344683229923,
      "grad_norm": 0.18870443105697632,
      "h_stock_valid_norm": 99.08850326538087,
      "invalid_tail_samples": 23.9,
      "learning_rate": 8.350156052170206e-05,
      "loss": 0.0521,
      "positive_teacher_pairs": 26829.3,
      "query_h_norm_ratio": 0.4426100134849548,
      "relation_loss": 0.0066034491173923016,
      "selected_pairs": 26829.3,
      "step": 9090,
      "student_distance_mean": 27.785671615600585,
      "teacher_distance_mean": 0.8609326958656311,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0066034491173923016,
      "weighted_relation_loss": 0.0066034491173923016
    },
    {
      "action_query_batch_std": 0.41796556413173674,
      "action_query_norm": 43.852918243408205,
      "action_relation_loss": 0.006918365322053433,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.512628624883067,
      "flow_matching_loss": 0.04666164182126522,
      "grad_norm": 0.14117293059825897,
      "h_stock_valid_norm": 99.06907806396484,
      "invalid_tail_samples": 21.9,
      "learning_rate": 8.346062600646738e-05,
      "loss": 0.0536,
      "positive_teacher_pairs": 27296.8,
      "query_h_norm_ratio": 0.44265501499176024,
      "relation_loss": 0.006918365322053433,
      "selected_pairs": 27296.8,
      "step": 9100,
      "student_distance_mean": 27.6788631439209,
      "teacher_distance_mean": 0.8626000165939331,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.006918365322053433,
      "weighted_relation_loss": 0.006918365322053433
    },
    {
      "action_query_batch_std": 0.41860414743423463,
      "action_query_norm": 43.84898338317871,
      "action_relation_loss": 0.007036965200677514,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.52198316183349,
      "flow_matching_loss": 0.04951459541916847,
      "grad_norm": 0.1792784035205841,
      "h_stock_valid_norm": 97.19214324951172,
      "invalid_tail_samples": 23.6,
      "learning_rate": 8.341965083341697e-05,
      "loss": 0.0566,
      "positive_teacher_pairs": 26893.0,
      "query_h_norm_ratio": 0.4511714369058609,
      "relation_loss": 0.007036965200677514,
      "selected_pairs": 26893.0,
      "step": 9110,
      "student_distance_mean": 27.696491241455078,
      "teacher_distance_mean": 0.8504085063934326,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.007036965200677514,
      "weighted_relation_loss": 0.007036965200677514
    },
    {
      "action_query_batch_std": 0.4167448878288269,
      "action_query_norm": 43.84844284057617,
      "action_relation_loss": 0.0068429382517933846,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.53133769878391,
      "flow_matching_loss": 0.04431203417479992,
      "grad_norm": 0.1407782882452011,
      "h_stock_valid_norm": 96.87025604248046,
      "invalid_tail_samples": 25.4,
      "learning_rate": 8.337863505233953e-05,
      "loss": 0.0512,
      "positive_teacher_pairs": 26481.3,
      "query_h_norm_ratio": 0.4526563614606857,
      "relation_loss": 0.0068429382517933846,
      "selected_pairs": 26481.3,
      "step": 9120,
      "student_distance_mean": 27.610541152954102,
      "teacher_distance_mean": 0.8544484734535217,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0068429382517933846,
      "weighted_relation_loss": 0.0068429382517933846
    },
    {
      "action_query_batch_std": 0.41493788063526155,
      "action_query_norm": 43.83956069946289,
      "action_relation_loss": 0.006740539008751512,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.540692235734332,
      "flow_matching_loss": 0.044400861114263536,
      "grad_norm": 0.15022794902324677,
      "h_stock_valid_norm": 97.23373947143554,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.33375787130731e-05,
      "loss": 0.0511,
      "positive_teacher_pairs": 26686.3,
      "query_h_norm_ratio": 0.450872528553009,
      "relation_loss": 0.006740539008751512,
      "selected_pairs": 26686.3,
      "step": 9130,
      "student_distance_mean": 27.570371246337892,
      "teacher_distance_mean": 0.8587462544441223,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.006740539008751512,
      "weighted_relation_loss": 0.006740539008751512
    },
    {
      "action_query_batch_std": 0.4201876163482666,
      "action_query_norm": 43.856594467163085,
      "action_relation_loss": 0.007263009017333388,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.550046772684752,
      "flow_matching_loss": 0.046661775931715964,
      "grad_norm": 0.1233576163649559,
      "h_stock_valid_norm": 98.05047836303712,
      "invalid_tail_samples": 24.8,
      "learning_rate": 8.3296481865505e-05,
      "loss": 0.0539,
      "positive_teacher_pairs": 26624.9,
      "query_h_norm_ratio": 0.4472931981086731,
      "relation_loss": 0.007263009017333388,
      "selected_pairs": 26624.9,
      "step": 9140,
      "student_distance_mean": 27.967891502380372,
      "teacher_distance_mean": 0.8522750616073609,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.007263009017333388,
      "weighted_relation_loss": 0.007263009017333388
    },
    {
      "action_query_batch_std": 0.41631239354610444,
      "action_query_norm": 43.85129318237305,
      "action_relation_loss": 0.0066599954385310415,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.559401309635174,
      "flow_matching_loss": 0.0432076096534729,
      "grad_norm": 0.1662677824497223,
      "h_stock_valid_norm": 98.65189590454102,
      "invalid_tail_samples": 22.7,
      "learning_rate": 8.325534455957181e-05,
      "loss": 0.0499,
      "positive_teacher_pairs": 27106.1,
      "query_h_norm_ratio": 0.4445102632045746,
      "relation_loss": 0.0066599954385310415,
      "selected_pairs": 27106.1,
      "step": 9150,
      "student_distance_mean": 27.582911872863768,
      "teacher_distance_mean": 0.8575829267501831,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0066599954385310415,
      "weighted_relation_loss": 0.0066599954385310415
    },
    {
      "action_query_batch_std": 0.4125862181186676,
      "action_query_norm": 43.834646987915036,
      "action_relation_loss": 0.006879009446129203,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.568755846585594,
      "flow_matching_loss": 0.04423500746488571,
      "grad_norm": 0.15915873646736145,
      "h_stock_valid_norm": 97.78634262084961,
      "invalid_tail_samples": 22.8,
      "learning_rate": 8.321416684525917e-05,
      "loss": 0.0511,
      "positive_teacher_pairs": 27087.5,
      "query_h_norm_ratio": 0.44827722012996674,
      "relation_loss": 0.006879009446129203,
      "selected_pairs": 27087.5,
      "step": 9160,
      "student_distance_mean": 27.457914543151855,
      "teacher_distance_mean": 0.8577404677867889,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.006879009446129203,
      "weighted_relation_loss": 0.006879009446129203
    },
    {
      "action_query_batch_std": 0.41636512279510496,
      "action_query_norm": 43.8410758972168,
      "action_relation_loss": 0.0064380333293229345,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.578110383536014,
      "flow_matching_loss": 0.047319105640053746,
      "grad_norm": 0.1731821596622467,
      "h_stock_valid_norm": 97.0224624633789,
      "invalid_tail_samples": 22.3,
      "learning_rate": 8.317294877260193e-05,
      "loss": 0.0538,
      "positive_teacher_pairs": 27199.0,
      "query_h_norm_ratio": 0.45187492966651915,
      "relation_loss": 0.0064380333293229345,
      "selected_pairs": 27199.0,
      "step": 9170,
      "student_distance_mean": 27.653591918945313,
      "teacher_distance_mean": 0.8592587888240815,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0064380333293229345,
      "weighted_relation_loss": 0.0064380333293229345
    },
    {
      "action_query_batch_std": 0.4158812344074249,
      "action_query_norm": 43.84077949523926,
      "action_relation_loss": 0.006582329142838716,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.587464920486436,
      "flow_matching_loss": 0.042994480580091476,
      "grad_norm": 0.12305797636508942,
      "h_stock_valid_norm": 97.31711959838867,
      "invalid_tail_samples": 21.1,
      "learning_rate": 8.313169039168394e-05,
      "loss": 0.0496,
      "positive_teacher_pairs": 27482.4,
      "query_h_norm_ratio": 0.45050026178359986,
      "relation_loss": 0.006582329142838716,
      "selected_pairs": 27482.4,
      "step": 9180,
      "student_distance_mean": 27.680141639709472,
      "teacher_distance_mean": 0.8564399898052215,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.006582329142838716,
      "weighted_relation_loss": 0.006582329142838716
    },
    {
      "action_query_batch_std": 0.4151363879442215,
      "action_query_norm": 43.83142738342285,
      "action_relation_loss": 0.006491822330281139,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.596819457436856,
      "flow_matching_loss": 0.047266003862023356,
      "grad_norm": 0.13320258259773254,
      "h_stock_valid_norm": 97.0239860534668,
      "invalid_tail_samples": 24.3,
      "learning_rate": 8.3090391752638e-05,
      "loss": 0.0538,
      "positive_teacher_pairs": 26741.5,
      "query_h_norm_ratio": 0.45176618695259096,
      "relation_loss": 0.006491822330281139,
      "selected_pairs": 26741.5,
      "step": 9190,
      "student_distance_mean": 27.648482704162596,
      "teacher_distance_mean": 0.8654028534889221,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.006491822330281139,
      "weighted_relation_loss": 0.006491822330281139
    },
    {
      "action_query_batch_std": 0.411237844824791,
      "action_query_norm": 43.82454605102539,
      "action_relation_loss": 0.007067151879891753,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.606173994387278,
      "flow_matching_loss": 0.04406767450273037,
      "grad_norm": 0.1460668444633484,
      "h_stock_valid_norm": 96.747314453125,
      "invalid_tail_samples": 22.0,
      "learning_rate": 8.304905290564586e-05,
      "loss": 0.0511,
      "positive_teacher_pairs": 27266.2,
      "query_h_norm_ratio": 0.4529818415641785,
      "relation_loss": 0.007067151879891753,
      "selected_pairs": 27266.2,
      "step": 9200,
      "student_distance_mean": 27.35224304199219,
      "teacher_distance_mean": 0.8390914857387543,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.007067151879891753,
      "weighted_relation_loss": 0.007067151879891753
    },
    {
      "action_query_batch_std": 0.41597189009189606,
      "action_query_norm": 43.83089027404785,
      "action_relation_loss": 0.006809269031509757,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.615528531337699,
      "flow_matching_loss": 0.04325110837817192,
      "grad_norm": 0.23685766756534576,
      "h_stock_valid_norm": 97.25578308105469,
      "invalid_tail_samples": 19.1,
      "learning_rate": 8.300767390093813e-05,
      "loss": 0.0501,
      "positive_teacher_pairs": 27952.5,
      "query_h_norm_ratio": 0.45068033039569855,
      "relation_loss": 0.006809269031509757,
      "selected_pairs": 27952.5,
      "step": 9210,
      "student_distance_mean": 27.557299423217774,
      "teacher_distance_mean": 0.8545582175254822,
      "valid_action_samples": 236.9,
      "weighted_auxiliary_loss": 0.006809269031509757,
      "weighted_relation_loss": 0.006809269031509757
    },
    {
      "action_query_batch_std": 0.41007161140441895,
      "action_query_norm": 43.811102294921874,
      "action_relation_loss": 0.0069769237656146284,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.62488306828812,
      "flow_matching_loss": 0.044563673064112665,
      "grad_norm": 0.19252564013004303,
      "h_stock_valid_norm": 96.73966827392579,
      "invalid_tail_samples": 24.4,
      "learning_rate": 8.296625478879417e-05,
      "loss": 0.0515,
      "positive_teacher_pairs": 26711.9,
      "query_h_norm_ratio": 0.4528827786445618,
      "relation_loss": 0.0069769237656146284,
      "selected_pairs": 26711.9,
      "step": 9220,
      "student_distance_mean": 27.153449058532715,
      "teacher_distance_mean": 0.846791785955429,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0069769237656146284,
      "weighted_relation_loss": 0.0069769237656146284
    },
    {
      "action_query_batch_std": 0.41910321116447447,
      "action_query_norm": 43.84667739868164,
      "action_relation_loss": 0.006982256891205907,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.63423760523854,
      "flow_matching_loss": 0.04482609145343304,
      "grad_norm": 0.16216018795967102,
      "h_stock_valid_norm": 96.6762924194336,
      "invalid_tail_samples": 22.3,
      "learning_rate": 8.292479561954214e-05,
      "loss": 0.0518,
      "positive_teacher_pairs": 27202.2,
      "query_h_norm_ratio": 0.4535456568002701,
      "relation_loss": 0.006982256891205907,
      "selected_pairs": 27202.2,
      "step": 9230,
      "student_distance_mean": 27.7441125869751,
      "teacher_distance_mean": 0.85308056473732,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.006982256891205907,
      "weighted_relation_loss": 0.006982256891205907
    },
    {
      "action_query_batch_std": 0.41708724796772,
      "action_query_norm": 43.840559005737305,
      "action_relation_loss": 0.006823012232780456,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.643592142188961,
      "flow_matching_loss": 0.04406600221991539,
      "grad_norm": 0.10459553450345993,
      "h_stock_valid_norm": 97.16410293579102,
      "invalid_tail_samples": 23.8,
      "learning_rate": 8.288329644355883e-05,
      "loss": 0.0509,
      "positive_teacher_pairs": 26848.0,
      "query_h_norm_ratio": 0.451206237077713,
      "relation_loss": 0.006823012232780456,
      "selected_pairs": 26848.0,
      "step": 9240,
      "student_distance_mean": 27.597332572937013,
      "teacher_distance_mean": 0.8500030636787415,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.006823012232780456,
      "weighted_relation_loss": 0.006823012232780456
    },
    {
      "action_query_batch_std": 0.41466244459152224,
      "action_query_norm": 43.831145477294925,
      "action_relation_loss": 0.006747239222750068,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.652946679139383,
      "flow_matching_loss": 0.0450581356883049,
      "grad_norm": 0.11290202289819717,
      "h_stock_valid_norm": 96.88598098754883,
      "invalid_tail_samples": 22.5,
      "learning_rate": 8.284175731126965e-05,
      "loss": 0.0518,
      "positive_teacher_pairs": 27155.9,
      "query_h_norm_ratio": 0.4524026393890381,
      "relation_loss": 0.006747239222750068,
      "selected_pairs": 27155.9,
      "step": 9250,
      "student_distance_mean": 27.641706466674805,
      "teacher_distance_mean": 0.8554554581642151,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.006747239222750068,
      "weighted_relation_loss": 0.006747239222750068
    },
    {
      "action_query_batch_std": 0.414641398191452,
      "action_query_norm": 43.82602653503418,
      "action_relation_loss": 0.006538986740633845,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.662301216089803,
      "flow_matching_loss": 0.04415214322507381,
      "grad_norm": 0.151033416390419,
      "h_stock_valid_norm": 96.92360458374023,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.280017827314854e-05,
      "loss": 0.0507,
      "positive_teacher_pairs": 26880.4,
      "query_h_norm_ratio": 0.4521759212017059,
      "relation_loss": 0.006538986740633845,
      "selected_pairs": 26880.4,
      "step": 9260,
      "student_distance_mean": 27.634297370910645,
      "teacher_distance_mean": 0.8544388234615325,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.006538986740633845,
      "weighted_relation_loss": 0.006538986740633845
    },
    {
      "action_query_batch_std": 0.4148677259683609,
      "action_query_norm": 43.82186088562012,
      "action_relation_loss": 0.006756503973156214,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.671655753040225,
      "flow_matching_loss": 0.04796503074467182,
      "grad_norm": 0.1747182160615921,
      "h_stock_valid_norm": 97.18084564208985,
      "invalid_tail_samples": 23.2,
      "learning_rate": 8.2758559379718e-05,
      "loss": 0.0547,
      "positive_teacher_pairs": 26986.5,
      "query_h_norm_ratio": 0.45093564689159393,
      "relation_loss": 0.006756503973156214,
      "selected_pairs": 26986.5,
      "step": 9270,
      "student_distance_mean": 27.453397750854492,
      "teacher_distance_mean": 0.8508757412433624,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.006756503973156214,
      "weighted_relation_loss": 0.006756503973156214
    },
    {
      "action_query_batch_std": 0.4183506786823273,
      "action_query_norm": 43.821261978149415,
      "action_relation_loss": 0.006720964377745986,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.681010289990645,
      "flow_matching_loss": 0.04569237120449543,
      "grad_norm": 0.14029067754745483,
      "h_stock_valid_norm": 97.81346130371094,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.271690068154887e-05,
      "loss": 0.0524,
      "positive_teacher_pairs": 26872.2,
      "query_h_norm_ratio": 0.44801182150840757,
      "relation_loss": 0.006720964377745986,
      "selected_pairs": 26872.2,
      "step": 9280,
      "student_distance_mean": 27.78600196838379,
      "teacher_distance_mean": 0.8620408058166504,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.006720964377745986,
      "weighted_relation_loss": 0.006720964377745986
    },
    {
      "action_query_batch_std": 0.41611326634883883,
      "action_query_norm": 43.81821136474609,
      "action_relation_loss": 0.006858883379027248,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.690364826941066,
      "flow_matching_loss": 0.04377685859799385,
      "grad_norm": 0.15403077006340027,
      "h_stock_valid_norm": 97.45025024414062,
      "invalid_tail_samples": 21.9,
      "learning_rate": 8.26752022292604e-05,
      "loss": 0.0506,
      "positive_teacher_pairs": 27291.2,
      "query_h_norm_ratio": 0.44965237081050874,
      "relation_loss": 0.006858883379027248,
      "selected_pairs": 27291.2,
      "step": 9290,
      "student_distance_mean": 27.624691009521484,
      "teacher_distance_mean": 0.8552644968032836,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.006858883379027248,
      "weighted_relation_loss": 0.006858883379027248
    },
    {
      "action_query_batch_std": 0.4181284993886948,
      "action_query_norm": 43.82995491027832,
      "action_relation_loss": 0.006847690138965845,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.699719363891488,
      "flow_matching_loss": 0.04424487538635731,
      "grad_norm": 0.11801118403673172,
      "h_stock_valid_norm": 98.02811126708984,
      "invalid_tail_samples": 25.1,
      "learning_rate": 8.263346407352018e-05,
      "loss": 0.0511,
      "positive_teacher_pairs": 26551.2,
      "query_h_norm_ratio": 0.4471242755651474,
      "relation_loss": 0.006847690138965845,
      "selected_pairs": 26551.2,
      "step": 9300,
      "student_distance_mean": 27.74499340057373,
      "teacher_distance_mean": 0.8556943356990814,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.006847690138965845,
      "weighted_relation_loss": 0.006847690138965845
    },
    {
      "action_query_batch_std": 0.41333244144916537,
      "action_query_norm": 43.81872634887695,
      "action_relation_loss": 0.006751716183498502,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.709073900841908,
      "flow_matching_loss": 0.04601556994020939,
      "grad_norm": 0.13163794577121735,
      "h_stock_valid_norm": 97.98049468994141,
      "invalid_tail_samples": 21.9,
      "learning_rate": 8.259168626504395e-05,
      "loss": 0.0528,
      "positive_teacher_pairs": 27293.3,
      "query_h_norm_ratio": 0.44722504913806915,
      "relation_loss": 0.006751716183498502,
      "selected_pairs": 27293.3,
      "step": 9310,
      "student_distance_mean": 27.455596733093262,
      "teacher_distance_mean": 0.8483156025409698,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.006751716183498502,
      "weighted_relation_loss": 0.006751716183498502
    },
    {
      "action_query_batch_std": 0.41665681898593904,
      "action_query_norm": 43.81883506774902,
      "action_relation_loss": 0.006280075293034315,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.71842843779233,
      "flow_matching_loss": 0.0448486540466547,
      "grad_norm": 0.12028935551643372,
      "h_stock_valid_norm": 96.97459869384765,
      "invalid_tail_samples": 24.6,
      "learning_rate": 8.254986885459572e-05,
      "loss": 0.0511,
      "positive_teacher_pairs": 26667.1,
      "query_h_norm_ratio": 0.45186265408992765,
      "relation_loss": 0.006280075293034315,
      "selected_pairs": 26667.1,
      "step": 9320,
      "student_distance_mean": 27.613594436645506,
      "teacher_distance_mean": 0.8559191942214965,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.006280075293034315,
      "weighted_relation_loss": 0.006280075293034315
    },
    {
      "action_query_batch_std": 0.41540138721466063,
      "action_query_norm": 43.82491836547852,
      "action_relation_loss": 0.006693276995792985,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.72778297474275,
      "flow_matching_loss": 0.04491581097245216,
      "grad_norm": 0.12810955941677094,
      "h_stock_valid_norm": 97.36254577636718,
      "invalid_tail_samples": 24.1,
      "learning_rate": 8.250801189298757e-05,
      "loss": 0.0516,
      "positive_teacher_pairs": 26792.7,
      "query_h_norm_ratio": 0.45012386739253996,
      "relation_loss": 0.006693276995792985,
      "selected_pairs": 26792.7,
      "step": 9330,
      "student_distance_mean": 27.570647621154784,
      "teacher_distance_mean": 0.8536652147769928,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.006693276995792985,
      "weighted_relation_loss": 0.006693276995792985
    },
    {
      "action_query_batch_std": 0.4185864984989166,
      "action_query_norm": 43.82639999389649,
      "action_relation_loss": 0.006821421626955271,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.737137511693172,
      "flow_matching_loss": 0.04409266859292984,
      "grad_norm": 0.17507033050060272,
      "h_stock_valid_norm": 97.3416732788086,
      "invalid_tail_samples": 26.2,
      "learning_rate": 8.246611543107967e-05,
      "loss": 0.0509,
      "positive_teacher_pairs": 26315.6,
      "query_h_norm_ratio": 0.45023883879184723,
      "relation_loss": 0.006821421626955271,
      "selected_pairs": 26315.6,
      "step": 9340,
      "student_distance_mean": 27.94123954772949,
      "teacher_distance_mean": 0.8641900658607483,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.006821421626955271,
      "weighted_relation_loss": 0.006821421626955271
    },
    {
      "action_query_batch_std": 0.4121039718389511,
      "action_query_norm": 43.80749588012695,
      "action_relation_loss": 0.006661097845062613,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.746492048643592,
      "flow_matching_loss": 0.04422585852444172,
      "grad_norm": 0.1504031866788864,
      "h_stock_valid_norm": 97.30049362182618,
      "invalid_tail_samples": 23.8,
      "learning_rate": 8.242417951978018e-05,
      "loss": 0.0509,
      "positive_teacher_pairs": 26851.8,
      "query_h_norm_ratio": 0.45023624300956727,
      "relation_loss": 0.006661097845062613,
      "selected_pairs": 26851.8,
      "step": 9350,
      "student_distance_mean": 27.408666610717773,
      "teacher_distance_mean": 0.847886073589325,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.006661097845062613,
      "weighted_relation_loss": 0.006661097845062613
    },
    {
      "action_query_batch_std": 0.41747669577598573,
      "action_query_norm": 43.81907958984375,
      "action_relation_loss": 0.006550814770162105,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.755846585594012,
      "flow_matching_loss": 0.04478264674544334,
      "grad_norm": 0.15198449790477753,
      "h_stock_valid_norm": 98.26199951171876,
      "invalid_tail_samples": 25.7,
      "learning_rate": 8.238220421004517e-05,
      "loss": 0.0513,
      "positive_teacher_pairs": 26414.5,
      "query_h_norm_ratio": 0.44594680666923525,
      "relation_loss": 0.006550814770162105,
      "selected_pairs": 26414.5,
      "step": 9360,
      "student_distance_mean": 27.825495529174805,
      "teacher_distance_mean": 0.8575779855251312,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.006550814770162105,
      "weighted_relation_loss": 0.006550814770162105
    },
    {
      "action_query_batch_std": 0.4105788469314575,
      "action_query_norm": 43.807141876220705,
      "action_relation_loss": 0.006855682609602809,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.765201122544434,
      "flow_matching_loss": 0.04648066125810146,
      "grad_norm": 0.15347732603549957,
      "h_stock_valid_norm": 98.1555046081543,
      "invalid_tail_samples": 22.8,
      "learning_rate": 8.234018955287859e-05,
      "loss": 0.0533,
      "positive_teacher_pairs": 27082.6,
      "query_h_norm_ratio": 0.44630596935749056,
      "relation_loss": 0.006855682609602809,
      "selected_pairs": 27082.6,
      "step": 9370,
      "student_distance_mean": 27.349742126464843,
      "teacher_distance_mean": 0.845087069272995,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.006855682609602809,
      "weighted_relation_loss": 0.006855682609602809
    },
    {
      "action_query_batch_std": 0.41116792857646944,
      "action_query_norm": 43.79433479309082,
      "action_relation_loss": 0.006841523991897702,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.774555659494855,
      "flow_matching_loss": 0.044623110070824624,
      "grad_norm": 0.19438262283802032,
      "h_stock_valid_norm": 97.84526977539062,
      "invalid_tail_samples": 22.2,
      "learning_rate": 8.229813559933224e-05,
      "loss": 0.0515,
      "positive_teacher_pairs": 27225.4,
      "query_h_norm_ratio": 0.4475896120071411,
      "relation_loss": 0.006841523991897702,
      "selected_pairs": 27225.4,
      "step": 9380,
      "student_distance_mean": 27.477862930297853,
      "teacher_distance_mean": 0.8554318130016327,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.006841523991897702,
      "weighted_relation_loss": 0.006841523991897702
    },
    {
      "action_query_batch_std": 0.4164917588233948,
      "action_query_norm": 43.80765724182129,
      "action_relation_loss": 0.00668094758875668,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.783910196445277,
      "flow_matching_loss": 0.04858676008880138,
      "grad_norm": 0.18695087730884552,
      "h_stock_valid_norm": 97.7941650390625,
      "invalid_tail_samples": 25.7,
      "learning_rate": 8.22560424005056e-05,
      "loss": 0.0553,
      "positive_teacher_pairs": 26422.0,
      "query_h_norm_ratio": 0.44796126782894136,
      "relation_loss": 0.00668094758875668,
      "selected_pairs": 26422.0,
      "step": 9390,
      "student_distance_mean": 27.703461837768554,
      "teacher_distance_mean": 0.858987295627594,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.00668094758875668,
      "weighted_relation_loss": 0.00668094758875668
    },
    {
      "action_query_batch_std": 0.4165519028902054,
      "action_query_norm": 43.808590698242185,
      "action_relation_loss": 0.006932881474494934,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.793264733395697,
      "flow_matching_loss": 0.043631714582443235,
      "grad_norm": 0.18778102099895477,
      "h_stock_valid_norm": 97.90099868774413,
      "invalid_tail_samples": 22.1,
      "learning_rate": 8.22139100075459e-05,
      "loss": 0.0506,
      "positive_teacher_pairs": 27243.0,
      "query_h_norm_ratio": 0.44748042821884154,
      "relation_loss": 0.006932881474494934,
      "selected_pairs": 27243.0,
      "step": 9400,
      "student_distance_mean": 27.788149452209474,
      "teacher_distance_mean": 0.8569692969322205,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.006932881474494934,
      "weighted_relation_loss": 0.006932881474494934
    },
    {
      "action_query_batch_std": 0.4099120706319809,
      "action_query_norm": 43.792636489868165,
      "action_relation_loss": 0.006908598402515054,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.802619270346117,
      "flow_matching_loss": 0.04650494903326034,
      "grad_norm": 0.15757015347480774,
      "h_stock_valid_norm": 97.68261413574218,
      "invalid_tail_samples": 25.1,
      "learning_rate": 8.217173847164799e-05,
      "loss": 0.0534,
      "positive_teacher_pairs": 26548.0,
      "query_h_norm_ratio": 0.44832201302051544,
      "relation_loss": 0.006908598402515054,
      "selected_pairs": 26548.0,
      "step": 9410,
      "student_distance_mean": 27.184478569030762,
      "teacher_distance_mean": 0.8524632453918457,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.006908598402515054,
      "weighted_relation_loss": 0.006908598402515054
    },
    {
      "action_query_batch_std": 0.4113775283098221,
      "action_query_norm": 43.78576583862305,
      "action_relation_loss": 0.006478962674736977,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.811973807296539,
      "flow_matching_loss": 0.043730982765555385,
      "grad_norm": 0.13848133385181427,
      "h_stock_valid_norm": 97.76626129150391,
      "invalid_tail_samples": 26.7,
      "learning_rate": 8.212952784405422e-05,
      "loss": 0.0502,
      "positive_teacher_pairs": 26180.6,
      "query_h_norm_ratio": 0.44786505997180937,
      "relation_loss": 0.006478962674736977,
      "selected_pairs": 26180.6,
      "step": 9420,
      "student_distance_mean": 27.392044067382812,
      "teacher_distance_mean": 0.8604609131813049,
      "valid_action_samples": 229.3,
      "weighted_auxiliary_loss": 0.006478962674736977,
      "weighted_relation_loss": 0.006478962674736977
    },
    {
      "action_query_batch_std": 0.4100773990154266,
      "action_query_norm": 43.7808162689209,
      "action_relation_loss": 0.006941298069432378,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.82132834424696,
      "flow_matching_loss": 0.044346295297145844,
      "grad_norm": 0.1564919352531433,
      "h_stock_valid_norm": 97.25473327636719,
      "invalid_tail_samples": 25.4,
      "learning_rate": 8.208727817605451e-05,
      "loss": 0.0513,
      "positive_teacher_pairs": 26478.8,
      "query_h_norm_ratio": 0.450169038772583,
      "relation_loss": 0.006941298069432378,
      "selected_pairs": 26478.8,
      "step": 9430,
      "student_distance_mean": 27.31537780761719,
      "teacher_distance_mean": 0.8485219478607178,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.006941298069432378,
      "weighted_relation_loss": 0.006941298069432378
    },
    {
      "action_query_batch_std": 0.41328610181808473,
      "action_query_norm": 43.79443016052246,
      "action_relation_loss": 0.006617309479042888,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.830682881197381,
      "flow_matching_loss": 0.045112811774015424,
      "grad_norm": 0.14159202575683594,
      "h_stock_valid_norm": 97.07307586669921,
      "invalid_tail_samples": 25.7,
      "learning_rate": 8.204498951898618e-05,
      "loss": 0.0517,
      "positive_teacher_pairs": 26415.1,
      "query_h_norm_ratio": 0.45115049183368683,
      "relation_loss": 0.006617309479042888,
      "selected_pairs": 26415.1,
      "step": 9440,
      "student_distance_mean": 27.515417289733886,
      "teacher_distance_mean": 0.8543251752853394,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.006617309479042888,
      "weighted_relation_loss": 0.006617309479042888
    },
    {
      "action_query_batch_std": 0.41167467534542085,
      "action_query_norm": 43.790283584594725,
      "action_relation_loss": 0.006878915382549167,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.840037418147801,
      "flow_matching_loss": 0.044040121138095856,
      "grad_norm": 0.12264656275510788,
      "h_stock_valid_norm": 97.14928512573242,
      "invalid_tail_samples": 25.7,
      "learning_rate": 8.200266192423395e-05,
      "loss": 0.0509,
      "positive_teacher_pairs": 26411.2,
      "query_h_norm_ratio": 0.45075660943984985,
      "relation_loss": 0.006878915382549167,
      "selected_pairs": 26411.2,
      "step": 9450,
      "student_distance_mean": 27.370631980895997,
      "teacher_distance_mean": 0.8563586354255677,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.006878915382549167,
      "weighted_relation_loss": 0.006878915382549167
    },
    {
      "action_query_batch_std": 0.4096493452787399,
      "action_query_norm": 43.78660011291504,
      "action_relation_loss": 0.006928741838783026,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.849391955098223,
      "flow_matching_loss": 0.045308321714401245,
      "grad_norm": 0.14592139422893524,
      "h_stock_valid_norm": 97.0854377746582,
      "invalid_tail_samples": 22.8,
      "learning_rate": 8.196029544322981e-05,
      "loss": 0.0522,
      "positive_teacher_pairs": 27088.9,
      "query_h_norm_ratio": 0.45101449489593504,
      "relation_loss": 0.006928741838783026,
      "selected_pairs": 27088.9,
      "step": 9460,
      "student_distance_mean": 27.245824241638182,
      "teacher_distance_mean": 0.860787284374237,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.006928741838783026,
      "weighted_relation_loss": 0.006928741838783026
    },
    {
      "action_query_batch_std": 0.4097634255886078,
      "action_query_norm": 43.77159576416015,
      "action_relation_loss": 0.007048167986795306,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.858746492048644,
      "flow_matching_loss": 0.04558665342628956,
      "grad_norm": 0.14291943609714508,
      "h_stock_valid_norm": 96.98802490234375,
      "invalid_tail_samples": 23.1,
      "learning_rate": 8.191789012745305e-05,
      "loss": 0.0526,
      "positive_teacher_pairs": 27009.4,
      "query_h_norm_ratio": 0.4513133347034454,
      "relation_loss": 0.007048167986795306,
      "selected_pairs": 27009.4,
      "step": 9470,
      "student_distance_mean": 27.21296787261963,
      "teacher_distance_mean": 0.849185562133789,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.007048167986795306,
      "weighted_relation_loss": 0.007048167986795306
    },
    {
      "action_query_batch_std": 0.4058157175779343,
      "action_query_norm": 43.77057991027832,
      "action_relation_loss": 0.006538587622344494,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.868101028999064,
      "flow_matching_loss": 0.04382181316614151,
      "grad_norm": 0.16799664497375488,
      "h_stock_valid_norm": 97.3169662475586,
      "invalid_tail_samples": 21.4,
      "learning_rate": 8.187544602843014e-05,
      "loss": 0.0504,
      "positive_teacher_pairs": 27410.5,
      "query_h_norm_ratio": 0.44977677464485166,
      "relation_loss": 0.006538587622344494,
      "selected_pairs": 27410.5,
      "step": 9480,
      "student_distance_mean": 26.977783012390137,
      "teacher_distance_mean": 0.8500980257987976,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.006538587622344494,
      "weighted_relation_loss": 0.006538587622344494
    },
    {
      "action_query_batch_std": 0.4074021100997925,
      "action_query_norm": 43.767416381835936,
      "action_relation_loss": 0.006628699554130435,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.877455565949486,
      "flow_matching_loss": 0.04498414658010006,
      "grad_norm": 0.16952621936798096,
      "h_stock_valid_norm": 98.6097511291504,
      "invalid_tail_samples": 23.8,
      "learning_rate": 8.183296319773466e-05,
      "loss": 0.0516,
      "positive_teacher_pairs": 26856.2,
      "query_h_norm_ratio": 0.4438697785139084,
      "relation_loss": 0.006628699554130435,
      "selected_pairs": 26856.2,
      "step": 9490,
      "student_distance_mean": 26.954374694824217,
      "teacher_distance_mean": 0.8592406749725342,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.006628699554130435,
      "weighted_relation_loss": 0.006628699554130435
    },
    {
      "action_query_batch_std": 0.41354718506336213,
      "action_query_norm": 43.78543243408203,
      "action_relation_loss": 0.006798798404633999,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.886810102899906,
      "flow_matching_loss": 0.04331250824034214,
      "grad_norm": 0.13245652616024017,
      "h_stock_valid_norm": 99.76839065551758,
      "invalid_tail_samples": 21.2,
      "learning_rate": 8.179044168698721e-05,
      "loss": 0.0501,
      "positive_teacher_pairs": 27456.7,
      "query_h_norm_ratio": 0.43887331485748293,
      "relation_loss": 0.006798798404633999,
      "selected_pairs": 27456.7,
      "step": 9500,
      "student_distance_mean": 27.513676261901857,
      "teacher_distance_mean": 0.8601311206817627,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.006798798404633999,
      "weighted_relation_loss": 0.006798798404633999
    },
    {
      "action_query_batch_std": 0.4128148347139359,
      "action_query_norm": 43.782100677490234,
      "action_relation_loss": 0.0068293468095362185,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.896164639850328,
      "flow_matching_loss": 0.04120820760726929,
      "grad_norm": 0.12905842065811157,
      "h_stock_valid_norm": 99.49588394165039,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.174788154785548e-05,
      "loss": 0.048,
      "positive_teacher_pairs": 26688.8,
      "query_h_norm_ratio": 0.44004566967487335,
      "relation_loss": 0.0068293468095362185,
      "selected_pairs": 26688.8,
      "step": 9510,
      "student_distance_mean": 27.502738761901856,
      "teacher_distance_mean": 0.847318708896637,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0068293468095362185,
      "weighted_relation_loss": 0.0068293468095362185
    },
    {
      "action_query_batch_std": 0.4123550742864609,
      "action_query_norm": 43.77639198303223,
      "action_relation_loss": 0.006342108361423016,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.905519176800748,
      "flow_matching_loss": 0.042892501875758174,
      "grad_norm": 0.10089806467294693,
      "h_stock_valid_norm": 98.40437393188476,
      "invalid_tail_samples": 24.2,
      "learning_rate": 8.170528283205403e-05,
      "loss": 0.0492,
      "positive_teacher_pairs": 26759.9,
      "query_h_norm_ratio": 0.4448688864707947,
      "relation_loss": 0.006342108361423016,
      "selected_pairs": 26759.9,
      "step": 9520,
      "student_distance_mean": 27.31769428253174,
      "teacher_distance_mean": 0.8535420715808868,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.006342108361423016,
      "weighted_relation_loss": 0.006342108361423016
    },
    {
      "action_query_batch_std": 0.4138423502445221,
      "action_query_norm": 43.78129959106445,
      "action_relation_loss": 0.006609952123835683,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.91487371375117,
      "flow_matching_loss": 0.04351613074541092,
      "grad_norm": 0.12091429531574249,
      "h_stock_valid_norm": 98.25045166015624,
      "invalid_tail_samples": 24.4,
      "learning_rate": 8.166264559134435e-05,
      "loss": 0.0501,
      "positive_teacher_pairs": 26708.5,
      "query_h_norm_ratio": 0.4456106603145599,
      "relation_loss": 0.006609952123835683,
      "selected_pairs": 26708.5,
      "step": 9530,
      "student_distance_mean": 27.458945274353027,
      "teacher_distance_mean": 0.8594070732593536,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.006609952123835683,
      "weighted_relation_loss": 0.006609952123835683
    },
    {
      "action_query_batch_std": 0.4140898138284683,
      "action_query_norm": 43.77146415710449,
      "action_relation_loss": 0.006243875063955784,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.92422825070159,
      "flow_matching_loss": 0.043038759008049966,
      "grad_norm": 0.10619425028562546,
      "h_stock_valid_norm": 99.09958190917969,
      "invalid_tail_samples": 23.3,
      "learning_rate": 8.161996987753467e-05,
      "loss": 0.0493,
      "positive_teacher_pairs": 26968.2,
      "query_h_norm_ratio": 0.44170214533805846,
      "relation_loss": 0.006243875063955784,
      "selected_pairs": 26968.2,
      "step": 9540,
      "student_distance_mean": 27.481653785705568,
      "teacher_distance_mean": 0.8582861006259919,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.006243875063955784,
      "weighted_relation_loss": 0.006243875063955784
    },
    {
      "action_query_batch_std": 0.4107528567314148,
      "action_query_norm": 43.75379257202148,
      "action_relation_loss": 0.006739122839644551,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.93358278765201,
      "flow_matching_loss": 0.04472208209335804,
      "grad_norm": 0.13241559267044067,
      "h_stock_valid_norm": 99.24439086914063,
      "invalid_tail_samples": 25.9,
      "learning_rate": 8.157725574248e-05,
      "loss": 0.0515,
      "positive_teacher_pairs": 26371.3,
      "query_h_norm_ratio": 0.4408714234828949,
      "relation_loss": 0.006739122839644551,
      "selected_pairs": 26371.3,
      "step": 9550,
      "student_distance_mean": 27.29177188873291,
      "teacher_distance_mean": 0.8556404113769531,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.006739122839644551,
      "weighted_relation_loss": 0.006739122839644551
    },
    {
      "action_query_batch_std": 0.41416856050491335,
      "action_query_norm": 43.76611404418945,
      "action_relation_loss": 0.006820881040766835,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.942937324602433,
      "flow_matching_loss": 0.0437344241887331,
      "grad_norm": 0.1468942016363144,
      "h_stock_valid_norm": 98.48986206054687,
      "invalid_tail_samples": 25.1,
      "learning_rate": 8.153450323808209e-05,
      "loss": 0.0506,
      "positive_teacher_pairs": 26551.3,
      "query_h_norm_ratio": 0.4443775027990341,
      "relation_loss": 0.006820881040766835,
      "selected_pairs": 26551.3,
      "step": 9560,
      "student_distance_mean": 27.56168727874756,
      "teacher_distance_mean": 0.8579295158386231,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.006820881040766835,
      "weighted_relation_loss": 0.006820881040766835
    },
    {
      "action_query_batch_std": 0.4113119810819626,
      "action_query_norm": 43.76968231201172,
      "action_relation_loss": 0.006737554771825671,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.952291861552853,
      "flow_matching_loss": 0.04232647605240345,
      "grad_norm": 0.1605735868215561,
      "h_stock_valid_norm": 98.27714691162109,
      "invalid_tail_samples": 23.9,
      "learning_rate": 8.149171241628923e-05,
      "loss": 0.0491,
      "positive_teacher_pairs": 26832.9,
      "query_h_norm_ratio": 0.4453745990991592,
      "relation_loss": 0.006737554771825671,
      "selected_pairs": 26832.9,
      "step": 9570,
      "student_distance_mean": 27.41331615447998,
      "teacher_distance_mean": 0.8480618178844452,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.006737554771825671,
      "weighted_relation_loss": 0.006737554771825671
    },
    {
      "action_query_batch_std": 0.4166644632816315,
      "action_query_norm": 43.7689395904541,
      "action_relation_loss": 0.006436594342812896,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.961646398503275,
      "flow_matching_loss": 0.04490119554102421,
      "grad_norm": 0.15984036028385162,
      "h_stock_valid_norm": 97.73498306274413,
      "invalid_tail_samples": 24.4,
      "learning_rate": 8.144888332909631e-05,
      "loss": 0.0513,
      "positive_teacher_pairs": 26706.4,
      "query_h_norm_ratio": 0.4478389769792557,
      "relation_loss": 0.006436594342812896,
      "selected_pairs": 26706.4,
      "step": 9580,
      "student_distance_mean": 27.633431434631348,
      "teacher_distance_mean": 0.8566411852836608,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.006436594342812896,
      "weighted_relation_loss": 0.006436594342812896
    },
    {
      "action_query_batch_std": 0.41210649311542513,
      "action_query_norm": 43.76444511413574,
      "action_relation_loss": 0.006709659984335303,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.971000935453695,
      "flow_matching_loss": 0.043578508868813516,
      "grad_norm": 0.17893129587173462,
      "h_stock_valid_norm": 97.92195205688476,
      "invalid_tail_samples": 21.9,
      "learning_rate": 8.14060160285447e-05,
      "loss": 0.0503,
      "positive_teacher_pairs": 27293.5,
      "query_h_norm_ratio": 0.446934112906456,
      "relation_loss": 0.006709659984335303,
      "selected_pairs": 27293.5,
      "step": 9590,
      "student_distance_mean": 27.323781776428223,
      "teacher_distance_mean": 0.8483127593994141,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.006709659984335303,
      "weighted_relation_loss": 0.006709659984335303
    },
    {
      "action_query_batch_std": 0.40600988268852234,
      "action_query_norm": 43.73651313781738,
      "action_relation_loss": 0.006791098788380623,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.980355472404115,
      "flow_matching_loss": 0.04383592568337917,
      "grad_norm": 0.14267383515834808,
      "h_stock_valid_norm": 98.04434432983399,
      "invalid_tail_samples": 23.0,
      "learning_rate": 8.136311056672223e-05,
      "loss": 0.0506,
      "positive_teacher_pairs": 27040.8,
      "query_h_norm_ratio": 0.4460992753505707,
      "relation_loss": 0.006791098788380623,
      "selected_pairs": 27040.8,
      "step": 9600,
      "student_distance_mean": 26.982030296325682,
      "teacher_distance_mean": 0.8440395653247833,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.006791098788380623,
      "weighted_relation_loss": 0.006791098788380623
    },
    {
      "action_query_batch_std": 0.41445266604423525,
      "action_query_norm": 43.76182327270508,
      "action_relation_loss": 0.007037349138408899,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.989710009354537,
      "flow_matching_loss": 0.04229952208697796,
      "grad_norm": 0.08761968463659286,
      "h_stock_valid_norm": 98.58602447509766,
      "invalid_tail_samples": 23.4,
      "learning_rate": 8.132016699576308e-05,
      "loss": 0.0493,
      "positive_teacher_pairs": 26944.2,
      "query_h_norm_ratio": 0.44390375912189484,
      "relation_loss": 0.007037349138408899,
      "selected_pairs": 26944.2,
      "step": 9610,
      "student_distance_mean": 27.482684326171874,
      "teacher_distance_mean": 0.8556291282176971,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.007037349138408899,
      "weighted_relation_loss": 0.007037349138408899
    },
    {
      "action_query_batch_std": 0.41234889924526213,
      "action_query_norm": 43.762142944335935,
      "action_relation_loss": 0.006562242656946183,
      "auxiliary_loss_weight": 1.0,
      "epoch": 8.999064546304957,
      "flow_matching_loss": 0.04595545306801796,
      "grad_norm": 0.1267533153295517,
      "h_stock_valid_norm": 97.53206634521484,
      "invalid_tail_samples": 25.7,
      "learning_rate": 8.127718536784771e-05,
      "loss": 0.0525,
      "positive_teacher_pairs": 26407.7,
      "query_h_norm_ratio": 0.44870067238807676,
      "relation_loss": 0.006562242656946183,
      "selected_pairs": 26407.7,
      "step": 9620,
      "student_distance_mean": 27.414274978637696,
      "teacher_distance_mean": 0.8562273681163788,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.006562242656946183,
      "weighted_relation_loss": 0.006562242656946183
    },
    {
      "action_query_batch_std": 0.40733048915863035,
      "action_query_norm": 43.7471565246582,
      "action_relation_loss": 0.006142041506245732,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.00841908325538,
      "flow_matching_loss": 0.04358980134129524,
      "grad_norm": 0.11327292025089264,
      "h_stock_valid_norm": 97.6576530456543,
      "invalid_tail_samples": 20.0,
      "learning_rate": 8.12341657352029e-05,
      "loss": 0.0497,
      "positive_teacher_pairs": 24759.8,
      "query_h_norm_ratio": 0.4479673981666565,
      "relation_loss": 0.006142041506245732,
      "selected_pairs": 24759.8,
      "step": 9630,
      "student_distance_mean": 27.14452362060547,
      "teacher_distance_mean": 0.8503988027572632,
      "valid_action_samples": 216.1,
      "weighted_auxiliary_loss": 0.006142041506245732,
      "weighted_relation_loss": 0.006142041506245732
    },
    {
      "action_query_batch_std": 0.4122011184692383,
      "action_query_norm": 43.75107116699219,
      "action_relation_loss": 0.006276788143441081,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.0177736202058,
      "flow_matching_loss": 0.042181844264268874,
      "grad_norm": 0.1041979044675827,
      "h_stock_valid_norm": 98.56256637573242,
      "invalid_tail_samples": 24.5,
      "learning_rate": 8.119110815010151e-05,
      "loss": 0.0485,
      "positive_teacher_pairs": 26692.2,
      "query_h_norm_ratio": 0.44389387369155886,
      "relation_loss": 0.006276788143441081,
      "selected_pairs": 26692.2,
      "step": 9640,
      "student_distance_mean": 27.444321060180663,
      "teacher_distance_mean": 0.853518408536911,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.006276788143441081,
      "weighted_relation_loss": 0.006276788143441081
    },
    {
      "action_query_batch_std": 0.41720038950443267,
      "action_query_norm": 43.76147232055664,
      "action_relation_loss": 0.006098431115970015,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.027128157156222,
      "flow_matching_loss": 0.04332579560577869,
      "grad_norm": 0.13577616214752197,
      "h_stock_valid_norm": 98.01544799804688,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.114801266486261e-05,
      "loss": 0.0494,
      "positive_teacher_pairs": 26873.4,
      "query_h_norm_ratio": 0.4464817613363266,
      "relation_loss": 0.006098431115970015,
      "selected_pairs": 26873.4,
      "step": 9650,
      "student_distance_mean": 27.655187606811523,
      "teacher_distance_mean": 0.8578909158706665,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.006098431115970015,
      "weighted_relation_loss": 0.006098431115970015
    },
    {
      "action_query_batch_std": 0.41082937121391294,
      "action_query_norm": 43.75075759887695,
      "action_relation_loss": 0.006728369556367397,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.036482694106642,
      "flow_matching_loss": 0.04398267231881618,
      "grad_norm": 0.11549156904220581,
      "h_stock_valid_norm": 97.11425323486328,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.110487933185122e-05,
      "loss": 0.0507,
      "positive_teacher_pairs": 26883.5,
      "query_h_norm_ratio": 0.45051189661026003,
      "relation_loss": 0.006728369556367397,
      "selected_pairs": 26883.5,
      "step": 9660,
      "student_distance_mean": 27.279931259155273,
      "teacher_distance_mean": 0.8559792697429657,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.006728369556367397,
      "weighted_relation_loss": 0.006728369556367397
    },
    {
      "action_query_batch_std": 0.4097182422876358,
      "action_query_norm": 43.750572204589844,
      "action_relation_loss": 0.006391653046011925,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.045837231057062,
      "flow_matching_loss": 0.04177981801331043,
      "grad_norm": 0.12010468542575836,
      "h_stock_valid_norm": 97.38479919433594,
      "invalid_tail_samples": 23.8,
      "learning_rate": 8.106170820347847e-05,
      "loss": 0.0482,
      "positive_teacher_pairs": 26850.6,
      "query_h_norm_ratio": 0.4492573320865631,
      "relation_loss": 0.006391653046011925,
      "selected_pairs": 26850.6,
      "step": 9670,
      "student_distance_mean": 27.12587547302246,
      "teacher_distance_mean": 0.849178946018219,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.006391653046011925,
      "weighted_relation_loss": 0.006391653046011925
    },
    {
      "action_query_batch_std": 0.4158382713794708,
      "action_query_norm": 43.75578269958496,
      "action_relation_loss": 0.006022986536845565,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.055191768007484,
      "flow_matching_loss": 0.0410142756998539,
      "grad_norm": 0.14762525260448456,
      "h_stock_valid_norm": 98.39460220336915,
      "invalid_tail_samples": 23.0,
      "learning_rate": 8.101849933220132e-05,
      "loss": 0.047,
      "positive_teacher_pairs": 27031.8,
      "query_h_norm_ratio": 0.44469998478889466,
      "relation_loss": 0.006022986536845565,
      "selected_pairs": 27031.8,
      "step": 9680,
      "student_distance_mean": 27.658311462402345,
      "teacher_distance_mean": 0.8557874321937561,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.006022986536845565,
      "weighted_relation_loss": 0.006022986536845565
    },
    {
      "action_query_batch_std": 0.4166691780090332,
      "action_query_norm": 43.75702247619629,
      "action_relation_loss": 0.0063837046269327406,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.064546304957904,
      "flow_matching_loss": 0.04288597479462623,
      "grad_norm": 0.16528411209583282,
      "h_stock_valid_norm": 98.8978157043457,
      "invalid_tail_samples": 21.3,
      "learning_rate": 8.097525277052264e-05,
      "loss": 0.0493,
      "positive_teacher_pairs": 27428.1,
      "query_h_norm_ratio": 0.4424491286277771,
      "relation_loss": 0.0063837046269327406,
      "selected_pairs": 27428.1,
      "step": 9690,
      "student_distance_mean": 27.684058952331544,
      "teacher_distance_mean": 0.8606919765472412,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.0063837046269327406,
      "weighted_relation_loss": 0.0063837046269327406
    },
    {
      "action_query_batch_std": 0.41154694855213164,
      "action_query_norm": 43.74074516296387,
      "action_relation_loss": 0.006492310017347336,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.073900841908326,
      "flow_matching_loss": 0.04383802078664303,
      "grad_norm": 0.12457481771707535,
      "h_stock_valid_norm": 99.68286819458008,
      "invalid_tail_samples": 23.0,
      "learning_rate": 8.093196857099106e-05,
      "loss": 0.0503,
      "positive_teacher_pairs": 27039.1,
      "query_h_norm_ratio": 0.4388004750013351,
      "relation_loss": 0.006492310017347336,
      "selected_pairs": 27039.1,
      "step": 9700,
      "student_distance_mean": 27.388664627075194,
      "teacher_distance_mean": 0.8559840500354767,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.006492310017347336,
      "weighted_relation_loss": 0.006492310017347336
    },
    {
      "action_query_batch_std": 0.41126593947410583,
      "action_query_norm": 43.74372177124023,
      "action_relation_loss": 0.0064406702760607,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.083255378858746,
      "flow_matching_loss": 0.04107342436909676,
      "grad_norm": 0.10632678121328354,
      "h_stock_valid_norm": 99.29809265136718,
      "invalid_tail_samples": 22.8,
      "learning_rate": 8.088864678620096e-05,
      "loss": 0.0475,
      "positive_teacher_pairs": 27078.4,
      "query_h_norm_ratio": 0.44053504765033724,
      "relation_loss": 0.0064406702760607,
      "selected_pairs": 27078.4,
      "step": 9710,
      "student_distance_mean": 27.35912094116211,
      "teacher_distance_mean": 0.8510213673114777,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0064406702760607,
      "weighted_relation_loss": 0.0064406702760607
    },
    {
      "action_query_batch_std": 0.4129501819610596,
      "action_query_norm": 43.73633728027344,
      "action_relation_loss": 0.006346471887081861,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.092609915809167,
      "flow_matching_loss": 0.04348317794501781,
      "grad_norm": 0.1165349930524826,
      "h_stock_valid_norm": 98.57222747802734,
      "invalid_tail_samples": 24.6,
      "learning_rate": 8.084528746879243e-05,
      "loss": 0.0498,
      "positive_teacher_pairs": 26664.5,
      "query_h_norm_ratio": 0.44370352029800414,
      "relation_loss": 0.006346471887081861,
      "selected_pairs": 26664.5,
      "step": 9720,
      "student_distance_mean": 27.405942916870117,
      "teacher_distance_mean": 0.8535687625408173,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.006346471887081861,
      "weighted_relation_loss": 0.006346471887081861
    },
    {
      "action_query_batch_std": 0.416101348400116,
      "action_query_norm": 43.75420036315918,
      "action_relation_loss": 0.006178586930036545,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.101964452759589,
      "flow_matching_loss": 0.04157295897603035,
      "grad_norm": 0.10997242480516434,
      "h_stock_valid_norm": 98.54168167114258,
      "invalid_tail_samples": 22.1,
      "learning_rate": 8.080189067145108e-05,
      "loss": 0.0478,
      "positive_teacher_pairs": 27248.0,
      "query_h_norm_ratio": 0.4440201550722122,
      "relation_loss": 0.006178586930036545,
      "selected_pairs": 27248.0,
      "step": 9730,
      "student_distance_mean": 27.574499320983886,
      "teacher_distance_mean": 0.8639096915721893,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.006178586930036545,
      "weighted_relation_loss": 0.006178586930036545
    },
    {
      "action_query_batch_std": 0.40805347859859464,
      "action_query_norm": 43.730393981933595,
      "action_relation_loss": 0.006477950233966112,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.111318989710009,
      "flow_matching_loss": 0.04366462789475918,
      "grad_norm": 0.14132611453533173,
      "h_stock_valid_norm": 98.04456634521485,
      "invalid_tail_samples": 23.2,
      "learning_rate": 8.075845644690813e-05,
      "loss": 0.0501,
      "positive_teacher_pairs": 26988.2,
      "query_h_norm_ratio": 0.44603207111358645,
      "relation_loss": 0.006477950233966112,
      "selected_pairs": 26988.2,
      "step": 9740,
      "student_distance_mean": 27.107929611206053,
      "teacher_distance_mean": 0.8465313851833344,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.006477950233966112,
      "weighted_relation_loss": 0.006477950233966112
    },
    {
      "action_query_batch_std": 0.4103537768125534,
      "action_query_norm": 43.7315860748291,
      "action_relation_loss": 0.00599568672478199,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.12067352666043,
      "flow_matching_loss": 0.04131508991122246,
      "grad_norm": 0.11268521845340729,
      "h_stock_valid_norm": 98.119384765625,
      "invalid_tail_samples": 25.5,
      "learning_rate": 8.07149848479403e-05,
      "loss": 0.0473,
      "positive_teacher_pairs": 26454.8,
      "query_h_norm_ratio": 0.4457051634788513,
      "relation_loss": 0.00599568672478199,
      "selected_pairs": 26454.8,
      "step": 9750,
      "student_distance_mean": 27.40929584503174,
      "teacher_distance_mean": 0.8530336201190949,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.00599568672478199,
      "weighted_relation_loss": 0.00599568672478199
    },
    {
      "action_query_batch_std": 0.41066834032535554,
      "action_query_norm": 43.72464294433594,
      "action_relation_loss": 0.006412297207862139,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.130028063610851,
      "flow_matching_loss": 0.04354771301150322,
      "grad_norm": 0.15014176070690155,
      "h_stock_valid_norm": 97.59890823364258,
      "invalid_tail_samples": 22.9,
      "learning_rate": 8.067147592736962e-05,
      "loss": 0.05,
      "positive_teacher_pairs": 27063.0,
      "query_h_norm_ratio": 0.44801058769226076,
      "relation_loss": 0.006412297207862139,
      "selected_pairs": 27063.0,
      "step": 9760,
      "student_distance_mean": 27.32064437866211,
      "teacher_distance_mean": 0.8509238004684448,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.006412297207862139,
      "weighted_relation_loss": 0.006412297207862139
    },
    {
      "action_query_batch_std": 0.4139649301767349,
      "action_query_norm": 43.73290328979492,
      "action_relation_loss": 0.006315688556060195,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.139382600561273,
      "flow_matching_loss": 0.041209737211465834,
      "grad_norm": 0.11289586126804352,
      "h_stock_valid_norm": 97.72986068725587,
      "invalid_tail_samples": 25.5,
      "learning_rate": 8.062792973806357e-05,
      "loss": 0.0475,
      "positive_teacher_pairs": 26458.3,
      "query_h_norm_ratio": 0.4474932670593262,
      "relation_loss": 0.006315688556060195,
      "selected_pairs": 26458.3,
      "step": 9770,
      "student_distance_mean": 27.544468879699707,
      "teacher_distance_mean": 0.8586795508861542,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.006315688556060195,
      "weighted_relation_loss": 0.006315688556060195
    },
    {
      "action_query_batch_std": 0.41432434916496275,
      "action_query_norm": 43.73242034912109,
      "action_relation_loss": 0.006333761010318994,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.148737137511693,
      "flow_matching_loss": 0.0412579745054245,
      "grad_norm": 0.1355791836977005,
      "h_stock_valid_norm": 98.02211837768554,
      "invalid_tail_samples": 24.3,
      "learning_rate": 8.058434633293485e-05,
      "loss": 0.0476,
      "positive_teacher_pairs": 26735.5,
      "query_h_norm_ratio": 0.4461503952741623,
      "relation_loss": 0.006333761010318994,
      "selected_pairs": 26735.5,
      "step": 9780,
      "student_distance_mean": 27.63404941558838,
      "teacher_distance_mean": 0.8601351320743561,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.006333761010318994,
      "weighted_relation_loss": 0.006333761010318994
    },
    {
      "action_query_batch_std": 0.4115309208631516,
      "action_query_norm": 43.7244197845459,
      "action_relation_loss": 0.0064100987743586305,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.158091674462113,
      "flow_matching_loss": 0.041729750111699104,
      "grad_norm": 0.12764276564121246,
      "h_stock_valid_norm": 98.04760360717773,
      "invalid_tail_samples": 25.3,
      "learning_rate": 8.054072576494143e-05,
      "loss": 0.0481,
      "positive_teacher_pairs": 26499.6,
      "query_h_norm_ratio": 0.4459558963775635,
      "relation_loss": 0.0064100987743586305,
      "selected_pairs": 26499.6,
      "step": 9790,
      "student_distance_mean": 27.275227546691895,
      "teacher_distance_mean": 0.8533718585968018,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0064100987743586305,
      "weighted_relation_loss": 0.0064100987743586305
    },
    {
      "action_query_batch_std": 0.4108523279428482,
      "action_query_norm": 43.725023651123045,
      "action_relation_loss": 0.006261580809950829,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.167446211412535,
      "flow_matching_loss": 0.04332497529685497,
      "grad_norm": 0.1309632509946823,
      "h_stock_valid_norm": 98.2116928100586,
      "invalid_tail_samples": 23.1,
      "learning_rate": 8.04970680870864e-05,
      "loss": 0.0496,
      "positive_teacher_pairs": 27030.0,
      "query_h_norm_ratio": 0.4452157974243164,
      "relation_loss": 0.006261580809950829,
      "selected_pairs": 27030.0,
      "step": 9800,
      "student_distance_mean": 27.32003059387207,
      "teacher_distance_mean": 0.8572708010673523,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.006261580809950829,
      "weighted_relation_loss": 0.006261580809950829
    },
    {
      "action_query_batch_std": 0.405502462387085,
      "action_query_norm": 43.71147117614746,
      "action_relation_loss": 0.006575066084042192,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.176800748362956,
      "flow_matching_loss": 0.04262648038566112,
      "grad_norm": 0.13128620386123657,
      "h_stock_valid_norm": 97.73867797851562,
      "invalid_tail_samples": 24.0,
      "learning_rate": 8.045337335241793e-05,
      "loss": 0.0492,
      "positive_teacher_pairs": 26800.4,
      "query_h_norm_ratio": 0.4472298800945282,
      "relation_loss": 0.006575066084042192,
      "selected_pairs": 26800.4,
      "step": 9810,
      "student_distance_mean": 26.914121437072755,
      "teacher_distance_mean": 0.8466021120548248,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.006575066084042192,
      "weighted_relation_loss": 0.006575066084042192
    },
    {
      "action_query_batch_std": 0.4109055757522583,
      "action_query_norm": 43.71658706665039,
      "action_relation_loss": 0.006576800998300314,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.186155285313378,
      "flow_matching_loss": 0.04276511557400227,
      "grad_norm": 0.1467253714799881,
      "h_stock_valid_norm": 97.26030578613282,
      "invalid_tail_samples": 23.2,
      "learning_rate": 8.040964161402932e-05,
      "loss": 0.0493,
      "positive_teacher_pairs": 26989.0,
      "query_h_norm_ratio": 0.44948320388793944,
      "relation_loss": 0.006576800998300314,
      "selected_pairs": 26989.0,
      "step": 9820,
      "student_distance_mean": 27.283828353881837,
      "teacher_distance_mean": 0.8527725577354431,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.006576800998300314,
      "weighted_relation_loss": 0.006576800998300314
    },
    {
      "action_query_batch_std": 0.4093040615320206,
      "action_query_norm": 43.714605712890624,
      "action_relation_loss": 0.006382745131850243,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.195509822263798,
      "flow_matching_loss": 0.04235457070171833,
      "grad_norm": 0.15338991582393646,
      "h_stock_valid_norm": 97.3637321472168,
      "invalid_tail_samples": 22.4,
      "learning_rate": 8.036587292505869e-05,
      "loss": 0.0487,
      "positive_teacher_pairs": 27173.4,
      "query_h_norm_ratio": 0.4489850282669067,
      "relation_loss": 0.006382745131850243,
      "selected_pairs": 27173.4,
      "step": 9830,
      "student_distance_mean": 27.149531745910643,
      "teacher_distance_mean": 0.846016275882721,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.006382745131850243,
      "weighted_relation_loss": 0.006382745131850243
    },
    {
      "action_query_batch_std": 0.40857090055942535,
      "action_query_norm": 43.71787796020508,
      "action_relation_loss": 0.006154344882816076,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.204864359214218,
      "flow_matching_loss": 0.043176133558154105,
      "grad_norm": 0.1336180418729782,
      "h_stock_valid_norm": 97.5422264099121,
      "invalid_tail_samples": 23.1,
      "learning_rate": 8.032206733868912e-05,
      "loss": 0.0493,
      "positive_teacher_pairs": 27010.4,
      "query_h_norm_ratio": 0.4481989413499832,
      "relation_loss": 0.006154344882816076,
      "selected_pairs": 27010.4,
      "step": 9840,
      "student_distance_mean": 27.148822402954103,
      "teacher_distance_mean": 0.846471506357193,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.006154344882816076,
      "weighted_relation_loss": 0.006154344882816076
    },
    {
      "action_query_batch_std": 0.40583436787128446,
      "action_query_norm": 43.707952880859374,
      "action_relation_loss": 0.00645698462612927,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.21421889616464,
      "flow_matching_loss": 0.04279773272573948,
      "grad_norm": 0.146414116024971,
      "h_stock_valid_norm": 97.82526473999023,
      "invalid_tail_samples": 24.9,
      "learning_rate": 8.027822490814858e-05,
      "loss": 0.0493,
      "positive_teacher_pairs": 26600.0,
      "query_h_norm_ratio": 0.4467981219291687,
      "relation_loss": 0.00645698462612927,
      "selected_pairs": 26600.0,
      "step": 9850,
      "student_distance_mean": 27.068870735168456,
      "teacher_distance_mean": 0.8485131978988647,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.00645698462612927,
      "weighted_relation_loss": 0.00645698462612927
    },
    {
      "action_query_batch_std": 0.40875016152858734,
      "action_query_norm": 43.71453437805176,
      "action_relation_loss": 0.006380832195281983,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.22357343311506,
      "flow_matching_loss": 0.04430273175239563,
      "grad_norm": 0.1611575335264206,
      "h_stock_valid_norm": 98.6059700012207,
      "invalid_tail_samples": 23.0,
      "learning_rate": 8.023434568670972e-05,
      "loss": 0.0507,
      "positive_teacher_pairs": 27033.0,
      "query_h_norm_ratio": 0.44333437383174895,
      "relation_loss": 0.006380832195281983,
      "selected_pairs": 27033.0,
      "step": 9860,
      "student_distance_mean": 27.175897789001464,
      "teacher_distance_mean": 0.8528063893318176,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.006380832195281983,
      "weighted_relation_loss": 0.006380832195281983
    },
    {
      "action_query_batch_std": 0.4135339707136154,
      "action_query_norm": 43.71826248168945,
      "action_relation_loss": 0.006509368307888508,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.232927970065482,
      "flow_matching_loss": 0.04421707913279534,
      "grad_norm": 0.14389668405056,
      "h_stock_valid_norm": 98.2126564025879,
      "invalid_tail_samples": 22.9,
      "learning_rate": 8.019042972768992e-05,
      "loss": 0.0507,
      "positive_teacher_pairs": 27059.4,
      "query_h_norm_ratio": 0.4451466351747513,
      "relation_loss": 0.006509368307888508,
      "selected_pairs": 27059.4,
      "step": 9870,
      "student_distance_mean": 27.643983459472658,
      "teacher_distance_mean": 0.866274493932724,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.006509368307888508,
      "weighted_relation_loss": 0.006509368307888508
    },
    {
      "action_query_batch_std": 0.40696955621242525,
      "action_query_norm": 43.70193367004394,
      "action_relation_loss": 0.006509123276919127,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.242282507015902,
      "flow_matching_loss": 0.04083598181605339,
      "grad_norm": 0.15560869872570038,
      "h_stock_valid_norm": 97.64670028686524,
      "invalid_tail_samples": 22.6,
      "learning_rate": 8.014647708445124e-05,
      "loss": 0.0473,
      "positive_teacher_pairs": 27130.3,
      "query_h_norm_ratio": 0.4475625425577164,
      "relation_loss": 0.006509123276919127,
      "selected_pairs": 27130.3,
      "step": 9880,
      "student_distance_mean": 26.907972717285155,
      "teacher_distance_mean": 0.8426767706871032,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.006509123276919127,
      "weighted_relation_loss": 0.006509123276919127
    },
    {
      "action_query_batch_std": 0.41536334455013274,
      "action_query_norm": 43.71986198425293,
      "action_relation_loss": 0.006266840826719999,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.251637043966324,
      "flow_matching_loss": 0.04085164070129395,
      "grad_norm": 0.13200393319129944,
      "h_stock_valid_norm": 98.18675231933594,
      "invalid_tail_samples": 24.3,
      "learning_rate": 8.010248781040027e-05,
      "loss": 0.0471,
      "positive_teacher_pairs": 26738.0,
      "query_h_norm_ratio": 0.44527918100357056,
      "relation_loss": 0.006266840826719999,
      "selected_pairs": 26738.0,
      "step": 9890,
      "student_distance_mean": 27.486834907531737,
      "teacher_distance_mean": 0.8584299087524414,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.006266840826719999,
      "weighted_relation_loss": 0.006266840826719999
    },
    {
      "action_query_batch_std": 0.40481250286102294,
      "action_query_norm": 43.68788146972656,
      "action_relation_loss": 0.006418243050575256,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.260991580916745,
      "flow_matching_loss": 0.042689231038093564,
      "grad_norm": 0.09652238339185715,
      "h_stock_valid_norm": 98.20542068481446,
      "invalid_tail_samples": 25.8,
      "learning_rate": 8.005846195898814e-05,
      "loss": 0.0491,
      "positive_teacher_pairs": 26387.5,
      "query_h_norm_ratio": 0.4448646128177643,
      "relation_loss": 0.006418243050575256,
      "selected_pairs": 26387.5,
      "step": 9900,
      "student_distance_mean": 27.01586856842041,
      "teacher_distance_mean": 0.8450894773006439,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.006418243050575256,
      "weighted_relation_loss": 0.006418243050575256
    },
    {
      "action_query_batch_std": 0.40444457828998565,
      "action_query_norm": 43.69422378540039,
      "action_relation_loss": 0.006176332384347916,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.270346117867165,
      "flow_matching_loss": 0.04005963802337646,
      "grad_norm": 0.12390632927417755,
      "h_stock_valid_norm": 97.14479751586914,
      "invalid_tail_samples": 21.6,
      "learning_rate": 8.00143995837104e-05,
      "loss": 0.0462,
      "positive_teacher_pairs": 27361.5,
      "query_h_norm_ratio": 0.44979654252529144,
      "relation_loss": 0.006176332384347916,
      "selected_pairs": 27361.5,
      "step": 9910,
      "student_distance_mean": 26.9236047744751,
      "teacher_distance_mean": 0.851405656337738,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.006176332384347916,
      "weighted_relation_loss": 0.006176332384347916
    },
    {
      "action_query_batch_std": 0.407366207242012,
      "action_query_norm": 43.70577392578125,
      "action_relation_loss": 0.006299178162589669,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.279700654817587,
      "flow_matching_loss": 0.04217899888753891,
      "grad_norm": 0.15887132287025452,
      "h_stock_valid_norm": 96.70309982299804,
      "invalid_tail_samples": 21.5,
      "learning_rate": 7.997030073810698e-05,
      "loss": 0.0485,
      "positive_teacher_pairs": 27387.0,
      "query_h_norm_ratio": 0.4519603461027145,
      "relation_loss": 0.006299178162589669,
      "selected_pairs": 27387.0,
      "step": 9920,
      "student_distance_mean": 27.135367774963377,
      "teacher_distance_mean": 0.8527666747570037,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.006299178162589669,
      "weighted_relation_loss": 0.006299178162589669
    },
    {
      "action_query_batch_std": 0.4103169798851013,
      "action_query_norm": 43.69969787597656,
      "action_relation_loss": 0.006111278664320707,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.289055191768007,
      "flow_matching_loss": 0.043930934742093086,
      "grad_norm": 0.15520355105400085,
      "h_stock_valid_norm": 97.38507919311523,
      "invalid_tail_samples": 25.7,
      "learning_rate": 7.992616547576217e-05,
      "loss": 0.05,
      "positive_teacher_pairs": 26413.0,
      "query_h_norm_ratio": 0.4487401843070984,
      "relation_loss": 0.006111278664320707,
      "selected_pairs": 26413.0,
      "step": 9930,
      "student_distance_mean": 27.42834339141846,
      "teacher_distance_mean": 0.8601338505744934,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.006111278664320707,
      "weighted_relation_loss": 0.006111278664320707
    },
    {
      "action_query_batch_std": 0.41694411635398865,
      "action_query_norm": 43.72177581787109,
      "action_relation_loss": 0.006012868322432041,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.298409728718429,
      "flow_matching_loss": 0.0417089868336916,
      "grad_norm": 0.08605649322271347,
      "h_stock_valid_norm": 98.45235443115234,
      "invalid_tail_samples": 23.3,
      "learning_rate": 7.988199385030445e-05,
      "loss": 0.0477,
      "positive_teacher_pairs": 26960.3,
      "query_h_norm_ratio": 0.44410121738910674,
      "relation_loss": 0.006012868322432041,
      "selected_pairs": 26960.3,
      "step": 9940,
      "student_distance_mean": 27.700617027282714,
      "teacher_distance_mean": 0.8676073133945466,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.006012868322432041,
      "weighted_relation_loss": 0.006012868322432041
    },
    {
      "action_query_batch_std": 0.4058087825775146,
      "action_query_norm": 43.6860710144043,
      "action_relation_loss": 0.006158540537580848,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.30776426566885,
      "flow_matching_loss": 0.04086344353854656,
      "grad_norm": 0.1336379200220108,
      "h_stock_valid_norm": 99.16725006103516,
      "invalid_tail_samples": 24.0,
      "learning_rate": 7.98377859154065e-05,
      "loss": 0.047,
      "positive_teacher_pairs": 26821.6,
      "query_h_norm_ratio": 0.44053137898445127,
      "relation_loss": 0.006158540537580848,
      "selected_pairs": 26821.6,
      "step": 9950,
      "student_distance_mean": 27.008912086486816,
      "teacher_distance_mean": 0.8541478514671326,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.006158540537580848,
      "weighted_relation_loss": 0.006158540537580848
    },
    {
      "action_query_batch_std": 0.40849547684192655,
      "action_query_norm": 43.69305000305176,
      "action_relation_loss": 0.006232244987040758,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.31711880261927,
      "flow_matching_loss": 0.04293464384973049,
      "grad_norm": 0.11866407096385956,
      "h_stock_valid_norm": 99.47249298095703,
      "invalid_tail_samples": 23.5,
      "learning_rate": 7.979354172478516e-05,
      "loss": 0.0492,
      "positive_teacher_pairs": 26922.8,
      "query_h_norm_ratio": 0.4392498642206192,
      "relation_loss": 0.006232244987040758,
      "selected_pairs": 26922.8,
      "step": 9960,
      "student_distance_mean": 27.215798950195314,
      "teacher_distance_mean": 0.8559262037277222,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.006232244987040758,
      "weighted_relation_loss": 0.006232244987040758
    },
    {
      "action_query_batch_std": 0.4039095401763916,
      "action_query_norm": 43.680443572998044,
      "action_relation_loss": 0.006250527175143361,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.326473339569691,
      "flow_matching_loss": 0.04357996098697185,
      "grad_norm": 0.1431552618741989,
      "h_stock_valid_norm": 98.97580413818359,
      "invalid_tail_samples": 24.0,
      "learning_rate": 7.974926133220127e-05,
      "loss": 0.0498,
      "positive_teacher_pairs": 26801.1,
      "query_h_norm_ratio": 0.4413307338953018,
      "relation_loss": 0.006250527175143361,
      "selected_pairs": 26801.1,
      "step": 9970,
      "student_distance_mean": 26.97277069091797,
      "teacher_distance_mean": 0.850598955154419,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.006250527175143361,
      "weighted_relation_loss": 0.006250527175143361
    },
    {
      "action_query_batch_std": 0.407455375790596,
      "action_query_norm": 43.6724349975586,
      "action_relation_loss": 0.006416732771322131,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.335827876520112,
      "flow_matching_loss": 0.04404903464019298,
      "grad_norm": 0.10961129516363144,
      "h_stock_valid_norm": 98.41303100585938,
      "invalid_tail_samples": 23.2,
      "learning_rate": 7.970494479145967e-05,
      "loss": 0.0505,
      "positive_teacher_pairs": 26986.7,
      "query_h_norm_ratio": 0.44377354681491854,
      "relation_loss": 0.006416732771322131,
      "selected_pairs": 26986.7,
      "step": 9980,
      "student_distance_mean": 26.951102256774902,
      "teacher_distance_mean": 0.8517721593379974,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.006416732771322131,
      "weighted_relation_loss": 0.006416732771322131
    },
    {
      "action_query_batch_std": 0.4104610443115234,
      "action_query_norm": 43.695263671875,
      "action_relation_loss": 0.006397949531674385,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.345182413470534,
      "flow_matching_loss": 0.04253679364919662,
      "grad_norm": 0.1478220820426941,
      "h_stock_valid_norm": 97.68032608032226,
      "invalid_tail_samples": 23.1,
      "learning_rate": 7.966059215640919e-05,
      "loss": 0.0489,
      "positive_teacher_pairs": 27013.6,
      "query_h_norm_ratio": 0.4473386287689209,
      "relation_loss": 0.006397949531674385,
      "selected_pairs": 27013.6,
      "step": 9990,
      "student_distance_mean": 27.24705638885498,
      "teacher_distance_mean": 0.856253319978714,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.006397949531674385,
      "weighted_relation_loss": 0.006397949531674385
    },
    {
      "action_query_batch_std": 0.41009676158428193,
      "action_query_norm": 43.691419219970705,
      "action_relation_loss": 0.006494758976623416,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.354536950420954,
      "flow_matching_loss": 0.040817943215370175,
      "grad_norm": 0.14481067657470703,
      "h_stock_valid_norm": 96.81069488525391,
      "invalid_tail_samples": 25.6,
      "learning_rate": 7.961620348094241e-05,
      "loss": 0.0473,
      "positive_teacher_pairs": 26443.5,
      "query_h_norm_ratio": 0.45131114721298216,
      "relation_loss": 0.006494758976623416,
      "selected_pairs": 26443.5,
      "step": 10000,
      "student_distance_mean": 27.298096656799316,
      "teacher_distance_mean": 0.8542515516281128,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.006494758976623416,
      "weighted_relation_loss": 0.006494758976623416
    },
    {
      "action_query_batch_std": 0.4052854150533676,
      "action_query_norm": 43.676760864257815,
      "action_relation_loss": 0.006526212161406875,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.363891487371376,
      "flow_matching_loss": 0.044513695687055585,
      "grad_norm": 0.1350618153810501,
      "h_stock_valid_norm": 97.2825553894043,
      "invalid_tail_samples": 27.7,
      "learning_rate": 7.957177881899577e-05,
      "loss": 0.051,
      "positive_teacher_pairs": 25962.7,
      "query_h_norm_ratio": 0.4489875465631485,
      "relation_loss": 0.006526212161406875,
      "selected_pairs": 25962.7,
      "step": 10010,
      "student_distance_mean": 27.009145736694336,
      "teacher_distance_mean": 0.8496267616748809,
      "valid_action_samples": 228.3,
      "weighted_auxiliary_loss": 0.006526212161406875,
      "weighted_relation_loss": 0.006526212161406875
    },
    {
      "action_query_batch_std": 0.4060157358646393,
      "action_query_norm": 43.67468109130859,
      "action_relation_loss": 0.006079970905557275,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.373246024321796,
      "flow_matching_loss": 0.042182520404458045,
      "grad_norm": 0.134678915143013,
      "h_stock_valid_norm": 97.95097732543945,
      "invalid_tail_samples": 22.9,
      "learning_rate": 7.952731822454944e-05,
      "loss": 0.0483,
      "positive_teacher_pairs": 27070.1,
      "query_h_norm_ratio": 0.4458881616592407,
      "relation_loss": 0.006079970905557275,
      "selected_pairs": 27070.1,
      "step": 10020,
      "student_distance_mean": 27.035466384887695,
      "teacher_distance_mean": 0.8517366468906402,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.006079970905557275,
      "weighted_relation_loss": 0.006079970905557275
    },
    {
      "action_query_batch_std": 0.4043109863996506,
      "action_query_norm": 43.67064170837402,
      "action_relation_loss": 0.006525780726224184,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.382600561272216,
      "flow_matching_loss": 0.043661387637257576,
      "grad_norm": 0.10446137934923172,
      "h_stock_valid_norm": 97.21977844238282,
      "invalid_tail_samples": 23.5,
      "learning_rate": 7.948282175162722e-05,
      "loss": 0.0502,
      "positive_teacher_pairs": 26922.1,
      "query_h_norm_ratio": 0.4492014229297638,
      "relation_loss": 0.006525780726224184,
      "selected_pairs": 26922.1,
      "step": 10030,
      "student_distance_mean": 26.875993156433104,
      "teacher_distance_mean": 0.8492736339569091,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.006525780726224184,
      "weighted_relation_loss": 0.006525780726224184
    },
    {
      "action_query_batch_std": 0.4128132492303848,
      "action_query_norm": 43.68736228942871,
      "action_relation_loss": 0.006219198927283287,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.391955098222638,
      "flow_matching_loss": 0.04170876294374466,
      "grad_norm": 0.15058112144470215,
      "h_stock_valid_norm": 96.52272262573243,
      "invalid_tail_samples": 24.1,
      "learning_rate": 7.943828945429652e-05,
      "loss": 0.0479,
      "positive_teacher_pairs": 26778.7,
      "query_h_norm_ratio": 0.4526148498058319,
      "relation_loss": 0.006219198927283287,
      "selected_pairs": 26778.7,
      "step": 10040,
      "student_distance_mean": 27.4289400100708,
      "teacher_distance_mean": 0.8608789682388306,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.006219198927283287,
      "weighted_relation_loss": 0.006219198927283287
    },
    {
      "action_query_batch_std": 0.4169128119945526,
      "action_query_norm": 43.696533584594725,
      "action_relation_loss": 0.006021379539743066,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.401309635173059,
      "flow_matching_loss": 0.042602111026644705,
      "grad_norm": 0.14518161118030548,
      "h_stock_valid_norm": 96.56991653442383,
      "invalid_tail_samples": 24.9,
      "learning_rate": 7.939372138666827e-05,
      "loss": 0.0486,
      "positive_teacher_pairs": 26604.3,
      "query_h_norm_ratio": 0.45249042510986326,
      "relation_loss": 0.006021379539743066,
      "selected_pairs": 26604.3,
      "step": 10050,
      "student_distance_mean": 27.637470245361328,
      "teacher_distance_mean": 0.8692049324512482,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.006021379539743066,
      "weighted_relation_loss": 0.006021379539743066
    },
    {
      "action_query_batch_std": 0.402793949842453,
      "action_query_norm": 43.66643867492676,
      "action_relation_loss": 0.0060554014518857,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.41066417212348,
      "flow_matching_loss": 0.039789159595966336,
      "grad_norm": 0.15544581413269043,
      "h_stock_valid_norm": 96.33700027465821,
      "invalid_tail_samples": 24.2,
      "learning_rate": 7.934911760289692e-05,
      "loss": 0.0458,
      "positive_teacher_pairs": 26763.5,
      "query_h_norm_ratio": 0.4532698571681976,
      "relation_loss": 0.0060554014518857,
      "selected_pairs": 26763.5,
      "step": 10060,
      "student_distance_mean": 26.84954967498779,
      "teacher_distance_mean": 0.8455536544322968,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0060554014518857,
      "weighted_relation_loss": 0.0060554014518857
    },
    {
      "action_query_batch_std": 0.399492210149765,
      "action_query_norm": 43.65553283691406,
      "action_relation_loss": 0.006514738220721484,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.4200187090739,
      "flow_matching_loss": 0.044028590247035024,
      "grad_norm": 0.09480752795934677,
      "h_stock_valid_norm": 95.67248992919922,
      "invalid_tail_samples": 23.2,
      "learning_rate": 7.930447815718022e-05,
      "loss": 0.0505,
      "positive_teacher_pairs": 26992.9,
      "query_h_norm_ratio": 0.4563076883554459,
      "relation_loss": 0.006514738220721484,
      "selected_pairs": 26992.9,
      "step": 10070,
      "student_distance_mean": 26.5881290435791,
      "teacher_distance_mean": 0.8410319685935974,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.006514738220721484,
      "weighted_relation_loss": 0.006514738220721484
    },
    {
      "action_query_batch_std": 0.40071686506271365,
      "action_query_norm": 43.662527847290036,
      "action_relation_loss": 0.006218844512477517,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.429373246024323,
      "flow_matching_loss": 0.04015232101082802,
      "grad_norm": 0.10785079747438431,
      "h_stock_valid_norm": 95.59054031372071,
      "invalid_tail_samples": 23.2,
      "learning_rate": 7.925980310375934e-05,
      "loss": 0.0464,
      "positive_teacher_pairs": 26988.3,
      "query_h_norm_ratio": 0.4567705005407333,
      "relation_loss": 0.006218844512477517,
      "selected_pairs": 26988.3,
      "step": 10080,
      "student_distance_mean": 26.646912002563475,
      "teacher_distance_mean": 0.8425539791584015,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.006218844512477517,
      "weighted_relation_loss": 0.006218844512477517
    },
    {
      "action_query_batch_std": 0.4077584624290466,
      "action_query_norm": 43.66832275390625,
      "action_relation_loss": 0.00591970207169652,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.438727782974743,
      "flow_matching_loss": 0.042431625351309774,
      "grad_norm": 0.125077024102211,
      "h_stock_valid_norm": 96.30759353637696,
      "invalid_tail_samples": 23.5,
      "learning_rate": 7.921509249691865e-05,
      "loss": 0.0484,
      "positive_teacher_pairs": 26924.7,
      "query_h_norm_ratio": 0.4534309506416321,
      "relation_loss": 0.00591970207169652,
      "selected_pairs": 26924.7,
      "step": 10090,
      "student_distance_mean": 27.13316535949707,
      "teacher_distance_mean": 0.853213119506836,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.00591970207169652,
      "weighted_relation_loss": 0.00591970207169652
    },
    {
      "action_query_batch_std": 0.4129357486963272,
      "action_query_norm": 43.681298828125,
      "action_relation_loss": 0.006226462218910456,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.448082319925163,
      "flow_matching_loss": 0.0419250912964344,
      "grad_norm": 0.17564772069454193,
      "h_stock_valid_norm": 97.28717956542968,
      "invalid_tail_samples": 24.2,
      "learning_rate": 7.917034639098579e-05,
      "loss": 0.0482,
      "positive_teacher_pairs": 26757.3,
      "query_h_norm_ratio": 0.448996490240097,
      "relation_loss": 0.006226462218910456,
      "selected_pairs": 26757.3,
      "step": 10100,
      "student_distance_mean": 27.528849601745605,
      "teacher_distance_mean": 0.8606637954711914,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.006226462218910456,
      "weighted_relation_loss": 0.006226462218910456
    },
    {
      "action_query_batch_std": 0.40429978370666503,
      "action_query_norm": 43.66260261535645,
      "action_relation_loss": 0.006251838430762291,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.457436856875585,
      "flow_matching_loss": 0.042455215379595755,
      "grad_norm": 0.15483014285564423,
      "h_stock_valid_norm": 96.97556762695312,
      "invalid_tail_samples": 23.4,
      "learning_rate": 7.912556484033146e-05,
      "loss": 0.0487,
      "positive_teacher_pairs": 26947.2,
      "query_h_norm_ratio": 0.45024805665016177,
      "relation_loss": 0.006251838430762291,
      "selected_pairs": 26947.2,
      "step": 10110,
      "student_distance_mean": 26.950588417053222,
      "teacher_distance_mean": 0.8518133878707885,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.006251838430762291,
      "weighted_relation_loss": 0.006251838430762291
    },
    {
      "action_query_batch_std": 0.40935524702072146,
      "action_query_norm": 43.6655689239502,
      "action_relation_loss": 0.006358454143628478,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.466791393826005,
      "flow_matching_loss": 0.042232298851013185,
      "grad_norm": 0.16139987111091614,
      "h_stock_valid_norm": 97.09113540649415,
      "invalid_tail_samples": 25.6,
      "learning_rate": 7.908074789936952e-05,
      "loss": 0.0486,
      "positive_teacher_pairs": 26439.1,
      "query_h_norm_ratio": 0.44974193871021273,
      "relation_loss": 0.006358454143628478,
      "selected_pairs": 26439.1,
      "step": 10120,
      "student_distance_mean": 27.3036958694458,
      "teacher_distance_mean": 0.8648438572883606,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.006358454143628478,
      "weighted_relation_loss": 0.006358454143628478
    },
    {
      "action_query_batch_std": 0.41209739446640015,
      "action_query_norm": 43.68334197998047,
      "action_relation_loss": 0.006141552748158574,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.476145930776427,
      "flow_matching_loss": 0.041717828065156934,
      "grad_norm": 0.12135615944862366,
      "h_stock_valid_norm": 96.99209976196289,
      "invalid_tail_samples": 22.7,
      "learning_rate": 7.903589562255673e-05,
      "loss": 0.0479,
      "positive_teacher_pairs": 27100.8,
      "query_h_norm_ratio": 0.45038955509662626,
      "relation_loss": 0.006141552748158574,
      "selected_pairs": 27100.8,
      "step": 10130,
      "student_distance_mean": 27.49800052642822,
      "teacher_distance_mean": 0.8635322093963623,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.006141552748158574,
      "weighted_relation_loss": 0.006141552748158574
    },
    {
      "action_query_batch_std": 0.402214840054512,
      "action_query_norm": 43.64986724853516,
      "action_relation_loss": 0.006171608064323664,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.485500467726848,
      "flow_matching_loss": 0.040922819450497624,
      "grad_norm": 0.11088274419307709,
      "h_stock_valid_norm": 96.53931121826172,
      "invalid_tail_samples": 24.9,
      "learning_rate": 7.899100806439285e-05,
      "loss": 0.0471,
      "positive_teacher_pairs": 26595.2,
      "query_h_norm_ratio": 0.4521485060453415,
      "relation_loss": 0.006171608064323664,
      "selected_pairs": 26595.2,
      "step": 10140,
      "student_distance_mean": 26.7326078414917,
      "teacher_distance_mean": 0.855384373664856,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.006171608064323664,
      "weighted_relation_loss": 0.006171608064323664
    },
    {
      "action_query_batch_std": 0.40332110822200773,
      "action_query_norm": 43.64330520629883,
      "action_relation_loss": 0.006232768436893821,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.494855004677268,
      "flow_matching_loss": 0.04160154424607754,
      "grad_norm": 0.12870635092258453,
      "h_stock_valid_norm": 97.32907409667969,
      "invalid_tail_samples": 25.1,
      "learning_rate": 7.894608527942049e-05,
      "loss": 0.0478,
      "positive_teacher_pairs": 26553.8,
      "query_h_norm_ratio": 0.4484186828136444,
      "relation_loss": 0.006232768436893821,
      "selected_pairs": 26553.8,
      "step": 10150,
      "student_distance_mean": 26.830391502380373,
      "teacher_distance_mean": 0.8540072202682495,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.006232768436893821,
      "weighted_relation_loss": 0.006232768436893821
    },
    {
      "action_query_batch_std": 0.40767752528190615,
      "action_query_norm": 43.662986755371094,
      "action_relation_loss": 0.0061228168196976185,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.50420954162769,
      "flow_matching_loss": 0.04030931517481804,
      "grad_norm": 0.11600583046674728,
      "h_stock_valid_norm": 97.26862182617188,
      "invalid_tail_samples": 23.2,
      "learning_rate": 7.89011273222251e-05,
      "loss": 0.0464,
      "positive_teacher_pairs": 26991.3,
      "query_h_norm_ratio": 0.4488956779241562,
      "relation_loss": 0.0061228168196976185,
      "selected_pairs": 26991.3,
      "step": 10160,
      "student_distance_mean": 27.185954093933105,
      "teacher_distance_mean": 0.8537363827228546,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0061228168196976185,
      "weighted_relation_loss": 0.0061228168196976185
    },
    {
      "action_query_batch_std": 0.404280886054039,
      "action_query_norm": 43.646952056884764,
      "action_relation_loss": 0.00641453037969768,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.51356407857811,
      "flow_matching_loss": 0.04278205074369908,
      "grad_norm": 0.13379453122615814,
      "h_stock_valid_norm": 96.92952499389648,
      "invalid_tail_samples": 25.1,
      "learning_rate": 7.885613424743482e-05,
      "loss": 0.0492,
      "positive_teacher_pairs": 26551.2,
      "query_h_norm_ratio": 0.45029974579811094,
      "relation_loss": 0.00641453037969768,
      "selected_pairs": 26551.2,
      "step": 10170,
      "student_distance_mean": 27.017692947387694,
      "teacher_distance_mean": 0.8575649619102478,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.00641453037969768,
      "weighted_relation_loss": 0.00641453037969768
    },
    {
      "action_query_batch_std": 0.40744548141956327,
      "action_query_norm": 43.64334907531738,
      "action_relation_loss": 0.006099441647529602,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.522918615528532,
      "flow_matching_loss": 0.04427533484995365,
      "grad_norm": 0.14026996493339539,
      "h_stock_valid_norm": 96.59446487426757,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.881110610972044e-05,
      "loss": 0.0504,
      "positive_teacher_pairs": 27037.7,
      "query_h_norm_ratio": 0.4518245130777359,
      "relation_loss": 0.006099441647529602,
      "selected_pairs": 27037.7,
      "step": 10180,
      "student_distance_mean": 27.099408531188963,
      "teacher_distance_mean": 0.8598715007305145,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.006099441647529602,
      "weighted_relation_loss": 0.006099441647529602
    },
    {
      "action_query_batch_std": 0.4064873993396759,
      "action_query_norm": 43.640255737304685,
      "action_relation_loss": 0.0060572423040866855,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.532273152478952,
      "flow_matching_loss": 0.04233303554356098,
      "grad_norm": 0.11848988384008408,
      "h_stock_valid_norm": 96.87214050292968,
      "invalid_tail_samples": 23.8,
      "learning_rate": 7.876604296379544e-05,
      "loss": 0.0484,
      "positive_teacher_pairs": 26849.8,
      "query_h_norm_ratio": 0.4504995375871658,
      "relation_loss": 0.0060572423040866855,
      "selected_pairs": 26849.8,
      "step": 10190,
      "student_distance_mean": 27.12912311553955,
      "teacher_distance_mean": 0.854699432849884,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0060572423040866855,
      "weighted_relation_loss": 0.0060572423040866855
    },
    {
      "action_query_batch_std": 0.40660094022750853,
      "action_query_norm": 43.64720840454102,
      "action_relation_loss": 0.005849885568022728,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.541627689429372,
      "flow_matching_loss": 0.042503968626260755,
      "grad_norm": 0.11751335859298706,
      "h_stock_valid_norm": 97.83625259399415,
      "invalid_tail_samples": 24.3,
      "learning_rate": 7.872094486441579e-05,
      "loss": 0.0484,
      "positive_teacher_pairs": 26728.3,
      "query_h_norm_ratio": 0.44613032042980194,
      "relation_loss": 0.005849885568022728,
      "selected_pairs": 26728.3,
      "step": 10200,
      "student_distance_mean": 27.00868148803711,
      "teacher_distance_mean": 0.8531465291976928,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.005849885568022728,
      "weighted_relation_loss": 0.005849885568022728
    },
    {
      "action_query_batch_std": 0.41184745728969574,
      "action_query_norm": 43.66269226074219,
      "action_relation_loss": 0.005877809459343552,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.550982226379794,
      "flow_matching_loss": 0.04101603850722313,
      "grad_norm": 0.12978872656822205,
      "h_stock_valid_norm": 98.46839599609375,
      "invalid_tail_samples": 21.8,
      "learning_rate": 7.867581186637991e-05,
      "loss": 0.0469,
      "positive_teacher_pairs": 27325.5,
      "query_h_norm_ratio": 0.4434224456548691,
      "relation_loss": 0.005877809459343552,
      "selected_pairs": 27325.5,
      "step": 10210,
      "student_distance_mean": 27.45389766693115,
      "teacher_distance_mean": 0.8588705003261566,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.005877809459343552,
      "weighted_relation_loss": 0.005877809459343552
    },
    {
      "action_query_batch_std": 0.40926340520381926,
      "action_query_norm": 43.6490665435791,
      "action_relation_loss": 0.005991742759943008,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.560336763330215,
      "flow_matching_loss": 0.041297518834471705,
      "grad_norm": 0.16060204803943634,
      "h_stock_valid_norm": 98.17326507568359,
      "invalid_tail_samples": 23.7,
      "learning_rate": 7.863064402452866e-05,
      "loss": 0.0473,
      "positive_teacher_pairs": 26875.3,
      "query_h_norm_ratio": 0.4446286976337433,
      "relation_loss": 0.005991742759943008,
      "selected_pairs": 26875.3,
      "step": 10220,
      "student_distance_mean": 27.301057052612304,
      "teacher_distance_mean": 0.8585340321063996,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.005991742759943008,
      "weighted_relation_loss": 0.005991742759943008
    },
    {
      "action_query_batch_std": 0.4028813183307648,
      "action_query_norm": 43.626433563232425,
      "action_relation_loss": 0.006278185453265905,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.569691300280637,
      "flow_matching_loss": 0.04173264428973198,
      "grad_norm": 0.10867665708065033,
      "h_stock_valid_norm": 97.22257843017579,
      "invalid_tail_samples": 25.4,
      "learning_rate": 7.858544139374524e-05,
      "loss": 0.048,
      "positive_teacher_pairs": 26479.4,
      "query_h_norm_ratio": 0.4487322449684143,
      "relation_loss": 0.006278185453265905,
      "selected_pairs": 26479.4,
      "step": 10230,
      "student_distance_mean": 26.832011795043947,
      "teacher_distance_mean": 0.8458316802978516,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.006278185453265905,
      "weighted_relation_loss": 0.006278185453265905
    },
    {
      "action_query_batch_std": 0.4074635297060013,
      "action_query_norm": 43.640692138671874,
      "action_relation_loss": 0.006022618012502789,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.579045837231057,
      "flow_matching_loss": 0.041691244766116145,
      "grad_norm": 0.14004337787628174,
      "h_stock_valid_norm": 97.4255615234375,
      "invalid_tail_samples": 21.7,
      "learning_rate": 7.854020402895508e-05,
      "loss": 0.0477,
      "positive_teacher_pairs": 27347.9,
      "query_h_norm_ratio": 0.44794151186943054,
      "relation_loss": 0.006022618012502789,
      "selected_pairs": 27347.9,
      "step": 10240,
      "student_distance_mean": 27.234357643127442,
      "teacher_distance_mean": 0.8548470497131347,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.006022618012502789,
      "weighted_relation_loss": 0.006022618012502789
    },
    {
      "action_query_batch_std": 0.40805950462818147,
      "action_query_norm": 43.633527755737305,
      "action_relation_loss": 0.005981814162805676,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.588400374181479,
      "flow_matching_loss": 0.04198864139616489,
      "grad_norm": 0.10819756984710693,
      "h_stock_valid_norm": 97.7919677734375,
      "invalid_tail_samples": 26.5,
      "learning_rate": 7.849493198512587e-05,
      "loss": 0.048,
      "positive_teacher_pairs": 26225.9,
      "query_h_norm_ratio": 0.4461918145418167,
      "relation_loss": 0.005981814162805676,
      "selected_pairs": 26225.9,
      "step": 10250,
      "student_distance_mean": 27.18052520751953,
      "teacher_distance_mean": 0.8583498656749725,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.005981814162805676,
      "weighted_relation_loss": 0.005981814162805676
    },
    {
      "action_query_batch_std": 0.4089296877384186,
      "action_query_norm": 43.63932800292969,
      "action_relation_loss": 0.006043322617188096,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.597754911131899,
      "flow_matching_loss": 0.041297723352909085,
      "grad_norm": 0.11906194686889648,
      "h_stock_valid_norm": 97.77762298583984,
      "invalid_tail_samples": 25.7,
      "learning_rate": 7.84496253172674e-05,
      "loss": 0.0473,
      "positive_teacher_pairs": 26416.1,
      "query_h_norm_ratio": 0.446318057179451,
      "relation_loss": 0.006043322617188096,
      "selected_pairs": 26416.1,
      "step": 10260,
      "student_distance_mean": 27.18561725616455,
      "teacher_distance_mean": 0.8655443251132965,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.006043322617188096,
      "weighted_relation_loss": 0.006043322617188096
    },
    {
      "action_query_batch_std": 0.4098807811737061,
      "action_query_norm": 43.64414978027344,
      "action_relation_loss": 0.006145062856376171,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.60710944808232,
      "flow_matching_loss": 0.041114726290106776,
      "grad_norm": 0.143958181142807,
      "h_stock_valid_norm": 98.220556640625,
      "invalid_tail_samples": 23.5,
      "learning_rate": 7.840428408043155e-05,
      "loss": 0.0473,
      "positive_teacher_pairs": 26923.1,
      "query_h_norm_ratio": 0.44435121715068815,
      "relation_loss": 0.006145062856376171,
      "selected_pairs": 26923.1,
      "step": 10270,
      "student_distance_mean": 27.309267616271974,
      "teacher_distance_mean": 0.8572563052177429,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.006145062856376171,
      "weighted_relation_loss": 0.006145062856376171
    },
    {
      "action_query_batch_std": 0.40465123355388644,
      "action_query_norm": 43.62499847412109,
      "action_relation_loss": 0.006242505041882396,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.616463985032741,
      "flow_matching_loss": 0.04105767793953419,
      "grad_norm": 0.1314546912908554,
      "h_stock_valid_norm": 97.6490577697754,
      "invalid_tail_samples": 24.4,
      "learning_rate": 7.835890832971218e-05,
      "loss": 0.0473,
      "positive_teacher_pairs": 26710.8,
      "query_h_norm_ratio": 0.44675748646259306,
      "relation_loss": 0.006242505041882396,
      "selected_pairs": 26710.8,
      "step": 10280,
      "student_distance_mean": 27.031550025939943,
      "teacher_distance_mean": 0.8528891623020172,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.006242505041882396,
      "weighted_relation_loss": 0.006242505041882396
    },
    {
      "action_query_batch_std": 0.402260622382164,
      "action_query_norm": 43.61493339538574,
      "action_relation_loss": 0.005807118210941553,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.625818521983161,
      "flow_matching_loss": 0.04176720902323723,
      "grad_norm": 0.10742603242397308,
      "h_stock_valid_norm": 97.1992416381836,
      "invalid_tail_samples": 22.2,
      "learning_rate": 7.831349812024512e-05,
      "loss": 0.0476,
      "positive_teacher_pairs": 27224.6,
      "query_h_norm_ratio": 0.4487204343080521,
      "relation_loss": 0.005807118210941553,
      "selected_pairs": 27224.6,
      "step": 10290,
      "student_distance_mean": 26.755244636535643,
      "teacher_distance_mean": 0.8492472290992736,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.005807118210941553,
      "weighted_relation_loss": 0.005807118210941553
    },
    {
      "action_query_batch_std": 0.401741036772728,
      "action_query_norm": 43.61916084289551,
      "action_relation_loss": 0.006027586525306106,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.635173058933583,
      "flow_matching_loss": 0.040382211282849315,
      "grad_norm": 0.13724474608898163,
      "h_stock_valid_norm": 96.77870864868164,
      "invalid_tail_samples": 22.6,
      "learning_rate": 7.826805350720807e-05,
      "loss": 0.0464,
      "positive_teacher_pairs": 27132.2,
      "query_h_norm_ratio": 0.4507111251354218,
      "relation_loss": 0.006027586525306106,
      "selected_pairs": 27132.2,
      "step": 10300,
      "student_distance_mean": 26.830511856079102,
      "teacher_distance_mean": 0.8519443392753601,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.006027586525306106,
      "weighted_relation_loss": 0.006027586525306106
    },
    {
      "action_query_batch_std": 0.40342182815074923,
      "action_query_norm": 43.62202339172363,
      "action_relation_loss": 0.00597200496122241,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.644527595884004,
      "flow_matching_loss": 0.04459145329892635,
      "grad_norm": 0.16516269743442535,
      "h_stock_valid_norm": 96.6453125,
      "invalid_tail_samples": 22.0,
      "learning_rate": 7.822257454582049e-05,
      "loss": 0.0506,
      "positive_teacher_pairs": 27266.0,
      "query_h_norm_ratio": 0.45136556327342986,
      "relation_loss": 0.00597200496122241,
      "selected_pairs": 27266.0,
      "step": 10310,
      "student_distance_mean": 26.86818199157715,
      "teacher_distance_mean": 0.8531257927417755,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.00597200496122241,
      "weighted_relation_loss": 0.00597200496122241
    },
    {
      "action_query_batch_std": 0.4078166902065277,
      "action_query_norm": 43.63020133972168,
      "action_relation_loss": 0.006266231229528785,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.653882132834426,
      "flow_matching_loss": 0.04081722311675549,
      "grad_norm": 0.1651764214038849,
      "h_stock_valid_norm": 97.18257446289063,
      "invalid_tail_samples": 26.2,
      "learning_rate": 7.817706129134363e-05,
      "loss": 0.0471,
      "positive_teacher_pairs": 26299.1,
      "query_h_norm_ratio": 0.4489539176225662,
      "relation_loss": 0.006266231229528785,
      "selected_pairs": 26299.1,
      "step": 10320,
      "student_distance_mean": 27.240763854980468,
      "teacher_distance_mean": 0.8612894892692566,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.006266231229528785,
      "weighted_relation_loss": 0.006266231229528785
    },
    {
      "action_query_batch_std": 0.4006460815668106,
      "action_query_norm": 43.59994621276856,
      "action_relation_loss": 0.006284525245428085,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.663236669784846,
      "flow_matching_loss": 0.0430476538836956,
      "grad_norm": 0.15216976404190063,
      "h_stock_valid_norm": 96.79775848388672,
      "invalid_tail_samples": 22.1,
      "learning_rate": 7.813151379908036e-05,
      "loss": 0.0493,
      "positive_teacher_pairs": 27251.1,
      "query_h_norm_ratio": 0.4504265129566193,
      "relation_loss": 0.006284525245428085,
      "selected_pairs": 27251.1,
      "step": 10330,
      "student_distance_mean": 26.721153450012206,
      "teacher_distance_mean": 0.8490405976772308,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.006284525245428085,
      "weighted_relation_loss": 0.006284525245428085
    },
    {
      "action_query_batch_std": 0.41125038266181946,
      "action_query_norm": 43.63215103149414,
      "action_relation_loss": 0.006107264617457986,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.672591206735266,
      "flow_matching_loss": 0.04177122339606285,
      "grad_norm": 0.14901907742023468,
      "h_stock_valid_norm": 97.1705207824707,
      "invalid_tail_samples": 21.6,
      "learning_rate": 7.808593212437523e-05,
      "loss": 0.0479,
      "positive_teacher_pairs": 27365.9,
      "query_h_norm_ratio": 0.44903269708156585,
      "relation_loss": 0.006107264617457986,
      "selected_pairs": 27365.9,
      "step": 10340,
      "student_distance_mean": 27.15101013183594,
      "teacher_distance_mean": 0.8576920211315155,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.006107264617457986,
      "weighted_relation_loss": 0.006107264617457986
    },
    {
      "action_query_batch_std": 0.4090162605047226,
      "action_query_norm": 43.61574516296387,
      "action_relation_loss": 0.006039553368464112,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.681945743685688,
      "flow_matching_loss": 0.04021826758980751,
      "grad_norm": 0.1186525970697403,
      "h_stock_valid_norm": 97.28572387695313,
      "invalid_tail_samples": 28.4,
      "learning_rate": 7.804031632261421e-05,
      "loss": 0.0463,
      "positive_teacher_pairs": 25797.4,
      "query_h_norm_ratio": 0.44832966327667234,
      "relation_loss": 0.006039553368464112,
      "selected_pairs": 25797.4,
      "step": 10350,
      "student_distance_mean": 27.2938138961792,
      "teacher_distance_mean": 0.8592258214950561,
      "valid_action_samples": 227.6,
      "weighted_auxiliary_loss": 0.006039553368464112,
      "weighted_relation_loss": 0.006039553368464112
    },
    {
      "action_query_batch_std": 0.41325363516807556,
      "action_query_norm": 43.640276718139646,
      "action_relation_loss": 0.0060120907612144945,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.691300280636108,
      "flow_matching_loss": 0.04404298290610313,
      "grad_norm": 0.10856545716524124,
      "h_stock_valid_norm": 96.37141799926758,
      "invalid_tail_samples": 23.2,
      "learning_rate": 7.799466644922484e-05,
      "loss": 0.0501,
      "positive_teacher_pairs": 26994.8,
      "query_h_norm_ratio": 0.45283800959587095,
      "relation_loss": 0.0060120907612144945,
      "selected_pairs": 26994.8,
      "step": 10360,
      "student_distance_mean": 27.64930419921875,
      "teacher_distance_mean": 0.8661458551883697,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0060120907612144945,
      "weighted_relation_loss": 0.0060120907612144945
    },
    {
      "action_query_batch_std": 0.40836712419986726,
      "action_query_norm": 43.62252311706543,
      "action_relation_loss": 0.00591668370179832,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.70065481758653,
      "flow_matching_loss": 0.04206013567745685,
      "grad_norm": 0.14127011597156525,
      "h_stock_valid_norm": 96.53309860229493,
      "invalid_tail_samples": 24.8,
      "learning_rate": 7.794898255967602e-05,
      "loss": 0.048,
      "positive_teacher_pairs": 26622.7,
      "query_h_norm_ratio": 0.45190538465976715,
      "relation_loss": 0.00591668370179832,
      "selected_pairs": 26622.7,
      "step": 10370,
      "student_distance_mean": 27.407375717163085,
      "teacher_distance_mean": 0.8588145256042481,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.00591668370179832,
      "weighted_relation_loss": 0.00591668370179832
    },
    {
      "action_query_batch_std": 0.4052063703536987,
      "action_query_norm": 43.60356101989746,
      "action_relation_loss": 0.006149513088166714,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.71000935453695,
      "flow_matching_loss": 0.043121831864118575,
      "grad_norm": 0.12516701221466064,
      "h_stock_valid_norm": 97.64632263183594,
      "invalid_tail_samples": 22.4,
      "learning_rate": 7.790326470947795e-05,
      "loss": 0.0493,
      "positive_teacher_pairs": 27182.3,
      "query_h_norm_ratio": 0.44655503928661344,
      "relation_loss": 0.006149513088166714,
      "selected_pairs": 27182.3,
      "step": 10380,
      "student_distance_mean": 26.932056999206544,
      "teacher_distance_mean": 0.856687593460083,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.006149513088166714,
      "weighted_relation_loss": 0.006149513088166714
    },
    {
      "action_query_batch_std": 0.40281087160110474,
      "action_query_norm": 43.60473899841308,
      "action_relation_loss": 0.00610327860340476,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.71936389148737,
      "flow_matching_loss": 0.04222394526004791,
      "grad_norm": 0.12573754787445068,
      "h_stock_valid_norm": 98.27823333740234,
      "invalid_tail_samples": 20.2,
      "learning_rate": 7.785751295418217e-05,
      "loss": 0.0483,
      "positive_teacher_pairs": 27687.7,
      "query_h_norm_ratio": 0.4436940997838974,
      "relation_loss": 0.00610327860340476,
      "selected_pairs": 27687.7,
      "step": 10390,
      "student_distance_mean": 26.771534729003907,
      "teacher_distance_mean": 0.8506497025489808,
      "valid_action_samples": 235.8,
      "weighted_auxiliary_loss": 0.00610327860340476,
      "weighted_relation_loss": 0.00610327860340476
    },
    {
      "action_query_batch_std": 0.40840153992176054,
      "action_query_norm": 43.61158714294434,
      "action_relation_loss": 0.00645540482364595,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.728718428437793,
      "flow_matching_loss": 0.041976680234074595,
      "grad_norm": 0.15583497285842896,
      "h_stock_valid_norm": 97.51755752563477,
      "invalid_tail_samples": 24.4,
      "learning_rate": 7.781172734938136e-05,
      "loss": 0.0484,
      "positive_teacher_pairs": 26716.3,
      "query_h_norm_ratio": 0.44722375869750974,
      "relation_loss": 0.00645540482364595,
      "selected_pairs": 26716.3,
      "step": 10400,
      "student_distance_mean": 27.063683128356935,
      "teacher_distance_mean": 0.8530598104000091,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.00645540482364595,
      "weighted_relation_loss": 0.00645540482364595
    },
    {
      "action_query_batch_std": 0.4116141557693481,
      "action_query_norm": 43.62570037841797,
      "action_relation_loss": 0.005830793315544725,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.738072965388213,
      "flow_matching_loss": 0.04028042182326317,
      "grad_norm": 0.13864043354988098,
      "h_stock_valid_norm": 97.32256393432617,
      "invalid_tail_samples": 23.4,
      "learning_rate": 7.776590795070933e-05,
      "loss": 0.0461,
      "positive_teacher_pairs": 26946.6,
      "query_h_norm_ratio": 0.44826107919216157,
      "relation_loss": 0.005830793315544725,
      "selected_pairs": 26946.6,
      "step": 10410,
      "student_distance_mean": 27.36889934539795,
      "teacher_distance_mean": 0.8541316211223602,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.005830793315544725,
      "weighted_relation_loss": 0.005830793315544725
    },
    {
      "action_query_batch_std": 0.4103162348270416,
      "action_query_norm": 43.61433906555176,
      "action_relation_loss": 0.005963114276528359,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.747427502338635,
      "flow_matching_loss": 0.04113717600703239,
      "grad_norm": 0.14145392179489136,
      "h_stock_valid_norm": 97.5094596862793,
      "invalid_tail_samples": 23.7,
      "learning_rate": 7.772005481384099e-05,
      "loss": 0.0471,
      "positive_teacher_pairs": 26877.7,
      "query_h_norm_ratio": 0.4472850918769836,
      "relation_loss": 0.005963114276528359,
      "selected_pairs": 26877.7,
      "step": 10420,
      "student_distance_mean": 27.19495849609375,
      "teacher_distance_mean": 0.8629486858844757,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.005963114276528359,
      "weighted_relation_loss": 0.005963114276528359
    },
    {
      "action_query_batch_std": 0.40429469347000124,
      "action_query_norm": 43.60053901672363,
      "action_relation_loss": 0.006330866692587733,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.756782039289055,
      "flow_matching_loss": 0.04114146195352077,
      "grad_norm": 0.14195866882801056,
      "h_stock_valid_norm": 97.25129623413086,
      "invalid_tail_samples": 24.1,
      "learning_rate": 7.767416799449223e-05,
      "loss": 0.0475,
      "positive_teacher_pairs": 26782.8,
      "query_h_norm_ratio": 0.4483297556638718,
      "relation_loss": 0.006330866692587733,
      "selected_pairs": 26782.8,
      "step": 10430,
      "student_distance_mean": 26.84813537597656,
      "teacher_distance_mean": 0.8598274290561676,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.006330866692587733,
      "weighted_relation_loss": 0.006330866692587733
    },
    {
      "action_query_batch_std": 0.40210804641246795,
      "action_query_norm": 43.58887557983398,
      "action_relation_loss": 0.006249895738437772,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.766136576239477,
      "flow_matching_loss": 0.041943779960274694,
      "grad_norm": 0.14782114326953888,
      "h_stock_valid_norm": 97.29592742919922,
      "invalid_tail_samples": 22.3,
      "learning_rate": 7.762824754841984e-05,
      "loss": 0.0482,
      "positive_teacher_pairs": 27200.9,
      "query_h_norm_ratio": 0.4480051636695862,
      "relation_loss": 0.006249895738437772,
      "selected_pairs": 27200.9,
      "step": 10440,
      "student_distance_mean": 26.65776824951172,
      "teacher_distance_mean": 0.8484436392784118,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.006249895738437772,
      "weighted_relation_loss": 0.006249895738437772
    },
    {
      "action_query_batch_std": 0.4050355076789856,
      "action_query_norm": 43.57986564636231,
      "action_relation_loss": 0.005875431001186371,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.775491113189897,
      "flow_matching_loss": 0.04237595573067665,
      "grad_norm": 0.1507841944694519,
      "h_stock_valid_norm": 97.52685317993163,
      "invalid_tail_samples": 22.5,
      "learning_rate": 7.758229353142152e-05,
      "loss": 0.0483,
      "positive_teacher_pairs": 27150.5,
      "query_h_norm_ratio": 0.4468515545129776,
      "relation_loss": 0.005875431001186371,
      "selected_pairs": 27150.5,
      "step": 10450,
      "student_distance_mean": 26.82289352416992,
      "teacher_distance_mean": 0.8582114398479461,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.005875431001186371,
      "weighted_relation_loss": 0.005875431001186371
    },
    {
      "action_query_batch_std": 0.4072532594203949,
      "action_query_norm": 43.5973201751709,
      "action_relation_loss": 0.006304466258734464,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.784845650140317,
      "flow_matching_loss": 0.04085662141442299,
      "grad_norm": 0.17225268483161926,
      "h_stock_valid_norm": 98.18717880249024,
      "invalid_tail_samples": 24.6,
      "learning_rate": 7.753630599933572e-05,
      "loss": 0.0472,
      "positive_teacher_pairs": 26671.7,
      "query_h_norm_ratio": 0.44403534233570097,
      "relation_loss": 0.006304466258734464,
      "selected_pairs": 26671.7,
      "step": 10460,
      "student_distance_mean": 27.103122520446778,
      "teacher_distance_mean": 0.8551178574562073,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.006304466258734464,
      "weighted_relation_loss": 0.006304466258734464
    },
    {
      "action_query_batch_std": 0.40650850236415864,
      "action_query_norm": 43.60128479003906,
      "action_relation_loss": 0.0060158440377563235,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.79420018709074,
      "flow_matching_loss": 0.04238296151161194,
      "grad_norm": 0.11567968875169754,
      "h_stock_valid_norm": 97.89195175170899,
      "invalid_tail_samples": 21.5,
      "learning_rate": 7.749028500804161e-05,
      "loss": 0.0484,
      "positive_teacher_pairs": 27389.4,
      "query_h_norm_ratio": 0.445408096909523,
      "relation_loss": 0.0060158440377563235,
      "selected_pairs": 27389.4,
      "step": 10470,
      "student_distance_mean": 27.090055847167967,
      "teacher_distance_mean": 0.8557397902011872,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.0060158440377563235,
      "weighted_relation_loss": 0.0060158440377563235
    },
    {
      "action_query_batch_std": 0.4103729784488678,
      "action_query_norm": 43.5937198638916,
      "action_relation_loss": 0.005745603516697883,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.80355472404116,
      "flow_matching_loss": 0.042312169820070265,
      "grad_norm": 0.1332351267337799,
      "h_stock_valid_norm": 97.1809066772461,
      "invalid_tail_samples": 22.4,
      "learning_rate": 7.744423061345906e-05,
      "loss": 0.0481,
      "positive_teacher_pairs": 27174.5,
      "query_h_norm_ratio": 0.4485906660556793,
      "relation_loss": 0.005745603516697883,
      "selected_pairs": 27174.5,
      "step": 10480,
      "student_distance_mean": 27.255622100830077,
      "teacher_distance_mean": 0.8634011030197144,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.005745603516697883,
      "weighted_relation_loss": 0.005745603516697883
    },
    {
      "action_query_batch_std": 0.40598202049732207,
      "action_query_norm": 43.58270492553711,
      "action_relation_loss": 0.006077315332368017,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.812909260991582,
      "flow_matching_loss": 0.04114428162574768,
      "grad_norm": 0.1277613490819931,
      "h_stock_valid_norm": 96.75662078857422,
      "invalid_tail_samples": 23.7,
      "learning_rate": 7.73981428715485e-05,
      "loss": 0.0472,
      "positive_teacher_pairs": 26873.1,
      "query_h_norm_ratio": 0.45044015645980834,
      "relation_loss": 0.006077315332368017,
      "selected_pairs": 26873.1,
      "step": 10490,
      "student_distance_mean": 26.983981323242187,
      "teacher_distance_mean": 0.8587723076343536,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.006077315332368017,
      "weighted_relation_loss": 0.006077315332368017
    },
    {
      "action_query_batch_std": 0.4077274203300476,
      "action_query_norm": 43.593279647827146,
      "action_relation_loss": 0.006006378028541803,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.822263797942002,
      "flow_matching_loss": 0.0424052145332098,
      "grad_norm": 0.15006639063358307,
      "h_stock_valid_norm": 97.08090896606446,
      "invalid_tail_samples": 21.6,
      "learning_rate": 7.735202183831085e-05,
      "loss": 0.0484,
      "positive_teacher_pairs": 27359.2,
      "query_h_norm_ratio": 0.44904405176639556,
      "relation_loss": 0.006006378028541803,
      "selected_pairs": 27359.2,
      "step": 10500,
      "student_distance_mean": 27.194240951538085,
      "teacher_distance_mean": 0.8621573328971863,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.006006378028541803,
      "weighted_relation_loss": 0.006006378028541803
    },
    {
      "action_query_batch_std": 0.40439220070838927,
      "action_query_norm": 43.57762489318848,
      "action_relation_loss": 0.0060967457946389915,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.831618334892422,
      "flow_matching_loss": 0.04092314653098583,
      "grad_norm": 0.16384568810462952,
      "h_stock_valid_norm": 97.10746231079102,
      "invalid_tail_samples": 24.5,
      "learning_rate": 7.730586756978757e-05,
      "loss": 0.047,
      "positive_teacher_pairs": 26695.9,
      "query_h_norm_ratio": 0.4487603366374969,
      "relation_loss": 0.0060967457946389915,
      "selected_pairs": 26695.9,
      "step": 10510,
      "student_distance_mean": 26.904608726501465,
      "teacher_distance_mean": 0.8522963941097259,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0060967457946389915,
      "weighted_relation_loss": 0.0060967457946389915
    },
    {
      "action_query_batch_std": 0.4036260724067688,
      "action_query_norm": 43.579647827148435,
      "action_relation_loss": 0.005844427365809679,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.840972871842844,
      "flow_matching_loss": 0.040186213329434395,
      "grad_norm": 0.11794731020927429,
      "h_stock_valid_norm": 97.401318359375,
      "invalid_tail_samples": 22.2,
      "learning_rate": 7.72596801220604e-05,
      "loss": 0.046,
      "positive_teacher_pairs": 27227.3,
      "query_h_norm_ratio": 0.44742527306079866,
      "relation_loss": 0.005844427365809679,
      "selected_pairs": 27227.3,
      "step": 10520,
      "student_distance_mean": 26.88065357208252,
      "teacher_distance_mean": 0.850090378522873,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.005844427365809679,
      "weighted_relation_loss": 0.005844427365809679
    },
    {
      "action_query_batch_std": 0.40352822542190553,
      "action_query_norm": 43.58179626464844,
      "action_relation_loss": 0.006226208433508873,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.850327408793264,
      "flow_matching_loss": 0.041688636690378186,
      "grad_norm": 0.1442973017692566,
      "h_stock_valid_norm": 97.61147079467773,
      "invalid_tail_samples": 21.8,
      "learning_rate": 7.721345955125147e-05,
      "loss": 0.0479,
      "positive_teacher_pairs": 27309.7,
      "query_h_norm_ratio": 0.4464850664138794,
      "relation_loss": 0.006226208433508873,
      "selected_pairs": 27309.7,
      "step": 10530,
      "student_distance_mean": 26.894816589355468,
      "teacher_distance_mean": 0.8480269312858582,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.006226208433508873,
      "weighted_relation_loss": 0.006226208433508873
    },
    {
      "action_query_batch_std": 0.40515044927597044,
      "action_query_norm": 43.58236122131348,
      "action_relation_loss": 0.005708340881392359,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.859681945743686,
      "flow_matching_loss": 0.03857061602175236,
      "grad_norm": 0.09367487579584122,
      "h_stock_valid_norm": 97.13129348754883,
      "invalid_tail_samples": 24.8,
      "learning_rate": 7.71672059135231e-05,
      "loss": 0.0443,
      "positive_teacher_pairs": 26618.8,
      "query_h_norm_ratio": 0.44869895577430724,
      "relation_loss": 0.005708340881392359,
      "selected_pairs": 26618.8,
      "step": 10540,
      "student_distance_mean": 26.90813217163086,
      "teacher_distance_mean": 0.8570847153663635,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.005708340881392359,
      "weighted_relation_loss": 0.005708340881392359
    },
    {
      "action_query_batch_std": 0.4030830949544907,
      "action_query_norm": 43.574523544311525,
      "action_relation_loss": 0.00592774017713964,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.869036482694106,
      "flow_matching_loss": 0.042915310710668564,
      "grad_norm": 0.13401217758655548,
      "h_stock_valid_norm": 96.6454246520996,
      "invalid_tail_samples": 22.3,
      "learning_rate": 7.712091926507787e-05,
      "loss": 0.0488,
      "positive_teacher_pairs": 27201.6,
      "query_h_norm_ratio": 0.4508729875087738,
      "relation_loss": 0.00592774017713964,
      "selected_pairs": 27201.6,
      "step": 10550,
      "student_distance_mean": 26.827637672424316,
      "teacher_distance_mean": 0.8613663196563721,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.00592774017713964,
      "weighted_relation_loss": 0.00592774017713964
    },
    {
      "action_query_batch_std": 0.4121095687150955,
      "action_query_norm": 43.59930305480957,
      "action_relation_loss": 0.005753371911123395,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.878391019644528,
      "flow_matching_loss": 0.03949131295084953,
      "grad_norm": 0.10932410508394241,
      "h_stock_valid_norm": 96.29352188110352,
      "invalid_tail_samples": 24.3,
      "learning_rate": 7.70745996621584e-05,
      "loss": 0.0452,
      "positive_teacher_pairs": 26745.2,
      "query_h_norm_ratio": 0.4527783513069153,
      "relation_loss": 0.005753371911123395,
      "selected_pairs": 26745.2,
      "step": 10560,
      "student_distance_mean": 27.4434663772583,
      "teacher_distance_mean": 0.8613176703453064,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.005753371911123395,
      "weighted_relation_loss": 0.005753371911123395
    },
    {
      "action_query_batch_std": 0.4066144168376923,
      "action_query_norm": 43.58503684997559,
      "action_relation_loss": 0.005700116278603673,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.887745556594949,
      "flow_matching_loss": 0.040528320521116254,
      "grad_norm": 0.11890073865652084,
      "h_stock_valid_norm": 95.84504776000976,
      "invalid_tail_samples": 24.8,
      "learning_rate": 7.702824716104735e-05,
      "loss": 0.0462,
      "positive_teacher_pairs": 26614.4,
      "query_h_norm_ratio": 0.45474628210067747,
      "relation_loss": 0.005700116278603673,
      "selected_pairs": 26614.4,
      "step": 10570,
      "student_distance_mean": 27.13018379211426,
      "teacher_distance_mean": 0.8562812626361846,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.005700116278603673,
      "weighted_relation_loss": 0.005700116278603673
    },
    {
      "action_query_batch_std": 0.3995077431201935,
      "action_query_norm": 43.5622802734375,
      "action_relation_loss": 0.00590536049567163,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.897100093545369,
      "flow_matching_loss": 0.041659441962838176,
      "grad_norm": 0.11064911633729935,
      "h_stock_valid_norm": 95.9150619506836,
      "invalid_tail_samples": 26.2,
      "learning_rate": 7.698186181806743e-05,
      "loss": 0.0476,
      "positive_teacher_pairs": 26298.9,
      "query_h_norm_ratio": 0.4541794449090958,
      "relation_loss": 0.00590536049567163,
      "selected_pairs": 26298.9,
      "step": 10580,
      "student_distance_mean": 26.572826194763184,
      "teacher_distance_mean": 0.8501322150230408,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.00590536049567163,
      "weighted_relation_loss": 0.00590536049567163
    },
    {
      "action_query_batch_std": 0.40114885866641997,
      "action_query_norm": 43.56043586730957,
      "action_relation_loss": 0.005952455569058657,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.90645463049579,
      "flow_matching_loss": 0.041327406838536265,
      "grad_norm": 0.11555805802345276,
      "h_stock_valid_norm": 96.54740295410156,
      "invalid_tail_samples": 25.1,
      "learning_rate": 7.693544368958116e-05,
      "loss": 0.0473,
      "positive_teacher_pairs": 26553.1,
      "query_h_norm_ratio": 0.4511972159147263,
      "relation_loss": 0.005952455569058657,
      "selected_pairs": 26553.1,
      "step": 10590,
      "student_distance_mean": 26.69262065887451,
      "teacher_distance_mean": 0.8433753788471222,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.005952455569058657,
      "weighted_relation_loss": 0.005952455569058657
    },
    {
      "action_query_batch_std": 0.403649365901947,
      "action_query_norm": 43.565324783325195,
      "action_relation_loss": 0.005907151382416487,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.915809167446211,
      "flow_matching_loss": 0.041871244087815285,
      "grad_norm": 0.08484932035207748,
      "h_stock_valid_norm": 97.35021057128907,
      "invalid_tail_samples": 23.5,
      "learning_rate": 7.688899283199096e-05,
      "loss": 0.0478,
      "positive_teacher_pairs": 26923.4,
      "query_h_norm_ratio": 0.4475159883499146,
      "relation_loss": 0.005907151382416487,
      "selected_pairs": 26923.4,
      "step": 10600,
      "student_distance_mean": 26.926121139526366,
      "teacher_distance_mean": 0.8498095512390137,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.005907151382416487,
      "weighted_relation_loss": 0.005907151382416487
    },
    {
      "action_query_batch_std": 0.40472870469093325,
      "action_query_norm": 43.57400131225586,
      "action_relation_loss": 0.005966102052479982,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.925163704396633,
      "flow_matching_loss": 0.040659941360354425,
      "grad_norm": 0.1202586218714714,
      "h_stock_valid_norm": 97.14934616088867,
      "invalid_tail_samples": 26.0,
      "learning_rate": 7.684250930173901e-05,
      "loss": 0.0466,
      "positive_teacher_pairs": 26340.3,
      "query_h_norm_ratio": 0.4485302984714508,
      "relation_loss": 0.005966102052479982,
      "selected_pairs": 26340.3,
      "step": 10610,
      "student_distance_mean": 26.888482093811035,
      "teacher_distance_mean": 0.8576667428016662,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.005966102052479982,
      "weighted_relation_loss": 0.005966102052479982
    },
    {
      "action_query_batch_std": 0.4094879150390625,
      "action_query_norm": 43.58210220336914,
      "action_relation_loss": 0.006107937032356858,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.934518241347053,
      "flow_matching_loss": 0.042317334190011026,
      "grad_norm": 0.10348672419786453,
      "h_stock_valid_norm": 96.67322387695313,
      "invalid_tail_samples": 24.2,
      "learning_rate": 7.679599315530718e-05,
      "loss": 0.0484,
      "positive_teacher_pairs": 26771.3,
      "query_h_norm_ratio": 0.45082598030567167,
      "relation_loss": 0.006107937032356858,
      "selected_pairs": 26771.3,
      "step": 10620,
      "student_distance_mean": 27.32444038391113,
      "teacher_distance_mean": 0.8624542355537415,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.006107937032356858,
      "weighted_relation_loss": 0.006107937032356858
    },
    {
      "action_query_batch_std": 0.39748758673667905,
      "action_query_norm": 43.54385643005371,
      "action_relation_loss": 0.00590200605802238,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.943872778297475,
      "flow_matching_loss": 0.042569208145141604,
      "grad_norm": 0.10139279067516327,
      "h_stock_valid_norm": 96.43005294799805,
      "invalid_tail_samples": 26.0,
      "learning_rate": 7.674944444921695e-05,
      "loss": 0.0485,
      "positive_teacher_pairs": 26344.7,
      "query_h_norm_ratio": 0.4515654414892197,
      "relation_loss": 0.00590200605802238,
      "selected_pairs": 26344.7,
      "step": 10630,
      "student_distance_mean": 26.473980522155763,
      "teacher_distance_mean": 0.8465651869773865,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.00590200605802238,
      "weighted_relation_loss": 0.00590200605802238
    },
    {
      "action_query_batch_std": 0.40822865068912506,
      "action_query_norm": 43.56692657470703,
      "action_relation_loss": 0.00591787276789546,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.953227315247895,
      "flow_matching_loss": 0.039725810289382935,
      "grad_norm": 0.09466899186372757,
      "h_stock_valid_norm": 96.45812149047852,
      "invalid_tail_samples": 22.6,
      "learning_rate": 7.670286324002944e-05,
      "loss": 0.0456,
      "positive_teacher_pairs": 27124.6,
      "query_h_norm_ratio": 0.4516762673854828,
      "relation_loss": 0.00591787276789546,
      "selected_pairs": 27124.6,
      "step": 10640,
      "student_distance_mean": 26.98082447052002,
      "teacher_distance_mean": 0.8613380253314972,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.00591787276789546,
      "weighted_relation_loss": 0.00591787276789546
    },
    {
      "action_query_batch_std": 0.408795228600502,
      "action_query_norm": 43.568202590942384,
      "action_relation_loss": 0.00607836926355958,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.962581852198316,
      "flow_matching_loss": 0.04175392538309097,
      "grad_norm": 0.12339288741350174,
      "h_stock_valid_norm": 97.31854400634765,
      "invalid_tail_samples": 23.8,
      "learning_rate": 7.665624958434514e-05,
      "loss": 0.0478,
      "positive_teacher_pairs": 26850.8,
      "query_h_norm_ratio": 0.4476909816265106,
      "relation_loss": 0.00607836926355958,
      "selected_pairs": 26850.8,
      "step": 10650,
      "student_distance_mean": 27.16490535736084,
      "teacher_distance_mean": 0.8540648639202117,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.00607836926355958,
      "weighted_relation_loss": 0.00607836926355958
    },
    {
      "action_query_batch_std": 0.41131198704242705,
      "action_query_norm": 43.5728157043457,
      "action_relation_loss": 0.006143347918987274,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.971936389148738,
      "flow_matching_loss": 0.04164177142083645,
      "grad_norm": 0.1211351752281189,
      "h_stock_valid_norm": 97.89333648681641,
      "invalid_tail_samples": 22.9,
      "learning_rate": 7.66096035388041e-05,
      "loss": 0.0478,
      "positive_teacher_pairs": 27058.9,
      "query_h_norm_ratio": 0.44510733485221865,
      "relation_loss": 0.006143347918987274,
      "selected_pairs": 27058.9,
      "step": 10660,
      "student_distance_mean": 27.34041748046875,
      "teacher_distance_mean": 0.856555312871933,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.006143347918987274,
      "weighted_relation_loss": 0.006143347918987274
    },
    {
      "action_query_batch_std": 0.40488463044166567,
      "action_query_norm": 43.55143852233887,
      "action_relation_loss": 0.006105252262204886,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.981290926099158,
      "flow_matching_loss": 0.04045262597501278,
      "grad_norm": 0.11152675747871399,
      "h_stock_valid_norm": 98.39048614501954,
      "invalid_tail_samples": 27.7,
      "learning_rate": 7.656292516008563e-05,
      "loss": 0.0466,
      "positive_teacher_pairs": 25962.6,
      "query_h_norm_ratio": 0.4426431179046631,
      "relation_loss": 0.006105252262204886,
      "selected_pairs": 25962.6,
      "step": 10670,
      "student_distance_mean": 26.892790603637696,
      "teacher_distance_mean": 0.8555333554744721,
      "valid_action_samples": 228.3,
      "weighted_auxiliary_loss": 0.006105252262204886,
      "weighted_relation_loss": 0.006105252262204886
    },
    {
      "action_query_batch_std": 0.40153337717056276,
      "action_query_norm": 43.545101165771484,
      "action_relation_loss": 0.006024407688528299,
      "auxiliary_loss_weight": 1.0,
      "epoch": 9.99064546304958,
      "flow_matching_loss": 0.04111199527978897,
      "grad_norm": 0.1056644544005394,
      "h_stock_valid_norm": 97.50865631103515,
      "invalid_tail_samples": 22.7,
      "learning_rate": 7.651621450490837e-05,
      "loss": 0.0471,
      "positive_teacher_pairs": 27103.3,
      "query_h_norm_ratio": 0.44658643901348116,
      "relation_loss": 0.006024407688528299,
      "selected_pairs": 27103.3,
      "step": 10680,
      "student_distance_mean": 26.634421730041502,
      "teacher_distance_mean": 0.8528829038143158,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.006024407688528299,
      "weighted_relation_loss": 0.006024407688528299
    },
    {
      "action_query_batch_std": 0.40876721441745756,
      "action_query_norm": 43.55040817260742,
      "action_relation_loss": 0.005865624267607927,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.0,
      "flow_matching_loss": 0.040656565129756926,
      "grad_norm": 0.23379887640476227,
      "h_stock_valid_norm": 96.28896713256836,
      "invalid_tail_samples": 25.6,
      "learning_rate": 7.646947163003017e-05,
      "loss": 0.0465,
      "positive_teacher_pairs": 23529.8,
      "query_h_norm_ratio": 0.4522939145565033,
      "relation_loss": 0.005865624267607927,
      "selected_pairs": 23529.8,
      "step": 10690,
      "student_distance_mean": 27.20517578125,
      "teacher_distance_mean": 0.8557106494903565,
      "valid_action_samples": 210.5,
      "weighted_auxiliary_loss": 0.005865624267607927,
      "weighted_relation_loss": 0.005865624267607927
    },
    {
      "action_query_batch_std": 0.4061440348625183,
      "action_query_norm": 43.54532661437988,
      "action_relation_loss": 0.00537753775715828,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.00935453695042,
      "flow_matching_loss": 0.03704854045063257,
      "grad_norm": 0.13412857055664062,
      "h_stock_valid_norm": 96.40266189575195,
      "invalid_tail_samples": 22.6,
      "learning_rate": 7.642269659224803e-05,
      "loss": 0.0424,
      "positive_teacher_pairs": 27135.3,
      "query_h_norm_ratio": 0.45171003639698026,
      "relation_loss": 0.00537753775715828,
      "selected_pairs": 27135.3,
      "step": 10700,
      "student_distance_mean": 27.06438617706299,
      "teacher_distance_mean": 0.8576485335826873,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.00537753775715828,
      "weighted_relation_loss": 0.00537753775715828
    },
    {
      "action_query_batch_std": 0.4032337337732315,
      "action_query_norm": 43.53965110778809,
      "action_relation_loss": 0.005824546050280333,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.018709073900842,
      "flow_matching_loss": 0.03888855297118425,
      "grad_norm": 0.16560089588165283,
      "h_stock_valid_norm": 97.32679824829101,
      "invalid_tail_samples": 23.5,
      "learning_rate": 7.637588944839803e-05,
      "loss": 0.0447,
      "positive_teacher_pairs": 26921.8,
      "query_h_norm_ratio": 0.44735963344573976,
      "relation_loss": 0.005824546050280333,
      "selected_pairs": 26921.8,
      "step": 10710,
      "student_distance_mean": 26.792201805114747,
      "teacher_distance_mean": 0.850073766708374,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.005824546050280333,
      "weighted_relation_loss": 0.005824546050280333
    },
    {
      "action_query_batch_std": 0.40414975881576537,
      "action_query_norm": 43.53401527404785,
      "action_relation_loss": 0.0055385168176144365,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.028063610851262,
      "flow_matching_loss": 0.03817136362195015,
      "grad_norm": 0.14917708933353424,
      "h_stock_valid_norm": 97.65830307006836,
      "invalid_tail_samples": 27.3,
      "learning_rate": 7.632905025535528e-05,
      "loss": 0.0437,
      "positive_teacher_pairs": 26046.4,
      "query_h_norm_ratio": 0.44578298926353455,
      "relation_loss": 0.0055385168176144365,
      "selected_pairs": 26046.4,
      "step": 10720,
      "student_distance_mean": 26.85548515319824,
      "teacher_distance_mean": 0.8498138427734375,
      "valid_action_samples": 228.7,
      "weighted_auxiliary_loss": 0.0055385168176144365,
      "weighted_relation_loss": 0.0055385168176144365
    },
    {
      "action_query_batch_std": 0.4022328943014145,
      "action_query_norm": 43.526845169067386,
      "action_relation_loss": 0.005881230952218175,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.037418147801684,
      "flow_matching_loss": 0.03907719776034355,
      "grad_norm": 0.169084832072258,
      "h_stock_valid_norm": 97.34890518188476,
      "invalid_tail_samples": 23.8,
      "learning_rate": 7.628217907003378e-05,
      "loss": 0.045,
      "positive_teacher_pairs": 26847.7,
      "query_h_norm_ratio": 0.447125232219696,
      "relation_loss": 0.005881230952218175,
      "selected_pairs": 26847.7,
      "step": 10730,
      "student_distance_mean": 26.858416175842287,
      "teacher_distance_mean": 0.8556819379329681,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.005881230952218175,
      "weighted_relation_loss": 0.005881230952218175
    },
    {
      "action_query_batch_std": 0.4021587997674942,
      "action_query_norm": 43.53818969726562,
      "action_relation_loss": 0.005782137531787157,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.046772684752105,
      "flow_matching_loss": 0.04077157452702522,
      "grad_norm": 0.11742933839559555,
      "h_stock_valid_norm": 97.51297912597656,
      "invalid_tail_samples": 24.1,
      "learning_rate": 7.623527594938649e-05,
      "loss": 0.0466,
      "positive_teacher_pairs": 26786.3,
      "query_h_norm_ratio": 0.4464900255203247,
      "relation_loss": 0.005782137531787157,
      "selected_pairs": 26786.3,
      "step": 10740,
      "student_distance_mean": 26.71593532562256,
      "teacher_distance_mean": 0.8522308528423309,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.005782137531787157,
      "weighted_relation_loss": 0.005782137531787157
    },
    {
      "action_query_batch_std": 0.4018250614404678,
      "action_query_norm": 43.52995338439941,
      "action_relation_loss": 0.005604417156428098,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.056127221702527,
      "flow_matching_loss": 0.03846249096095562,
      "grad_norm": 0.1313321739435196,
      "h_stock_valid_norm": 97.30597305297852,
      "invalid_tail_samples": 23.6,
      "learning_rate": 7.618834095040509e-05,
      "loss": 0.0441,
      "positive_teacher_pairs": 26896.0,
      "query_h_norm_ratio": 0.4473555564880371,
      "relation_loss": 0.005604417156428098,
      "selected_pairs": 26896.0,
      "step": 10750,
      "student_distance_mean": 26.835369300842284,
      "teacher_distance_mean": 0.8515509784221649,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.005604417156428098,
      "weighted_relation_loss": 0.005604417156428098
    },
    {
      "action_query_batch_std": 0.3994269073009491,
      "action_query_norm": 43.524749755859375,
      "action_relation_loss": 0.0055759594775736335,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.065481758652947,
      "flow_matching_loss": 0.03808034285902977,
      "grad_norm": 0.1388663798570633,
      "h_stock_valid_norm": 97.3152229309082,
      "invalid_tail_samples": 21.2,
      "learning_rate": 7.614137413012e-05,
      "loss": 0.0437,
      "positive_teacher_pairs": 27458.3,
      "query_h_norm_ratio": 0.4472593992948532,
      "relation_loss": 0.0055759594775736335,
      "selected_pairs": 27458.3,
      "step": 10760,
      "student_distance_mean": 26.57678985595703,
      "teacher_distance_mean": 0.8458512544631958,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.0055759594775736335,
      "weighted_relation_loss": 0.0055759594775736335
    },
    {
      "action_query_batch_std": 0.4095829755067825,
      "action_query_norm": 43.54943733215332,
      "action_relation_loss": 0.0057081521023064855,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.074836295603367,
      "flow_matching_loss": 0.04220686890184879,
      "grad_norm": 0.08720701187849045,
      "h_stock_valid_norm": 96.7833137512207,
      "invalid_tail_samples": 24.7,
      "learning_rate": 7.609437554560042e-05,
      "loss": 0.0479,
      "positive_teacher_pairs": 26650.3,
      "query_h_norm_ratio": 0.449971878528595,
      "relation_loss": 0.0057081521023064855,
      "selected_pairs": 26650.3,
      "step": 10770,
      "student_distance_mean": 27.185256958007812,
      "teacher_distance_mean": 0.8649433672428131,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0057081521023064855,
      "weighted_relation_loss": 0.0057081521023064855
    },
    {
      "action_query_batch_std": 0.4029574304819107,
      "action_query_norm": 43.53190574645996,
      "action_relation_loss": 0.005695111351087689,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.084190832553789,
      "flow_matching_loss": 0.03988732434809208,
      "grad_norm": 0.1118173897266388,
      "h_stock_valid_norm": 97.02987136840821,
      "invalid_tail_samples": 24.9,
      "learning_rate": 7.604734525395398e-05,
      "loss": 0.0456,
      "positive_teacher_pairs": 26591.3,
      "query_h_norm_ratio": 0.4486504107713699,
      "relation_loss": 0.005695111351087689,
      "selected_pairs": 26591.3,
      "step": 10780,
      "student_distance_mean": 26.80887393951416,
      "teacher_distance_mean": 0.8512816071510315,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.005695111351087689,
      "weighted_relation_loss": 0.005695111351087689
    },
    {
      "action_query_batch_std": 0.4054546445608139,
      "action_query_norm": 43.54066009521485,
      "action_relation_loss": 0.00594143969938159,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.09354536950421,
      "flow_matching_loss": 0.03906295970082283,
      "grad_norm": 0.10500112175941467,
      "h_stock_valid_norm": 97.61150665283203,
      "invalid_tail_samples": 23.7,
      "learning_rate": 7.600028331232699e-05,
      "loss": 0.045,
      "positive_teacher_pairs": 26884.9,
      "query_h_norm_ratio": 0.44606299698352814,
      "relation_loss": 0.00594143969938159,
      "selected_pairs": 26884.9,
      "step": 10790,
      "student_distance_mean": 27.037656593322755,
      "teacher_distance_mean": 0.8588655471801758,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.00594143969938159,
      "weighted_relation_loss": 0.00594143969938159
    },
    {
      "action_query_batch_std": 0.4010607123374939,
      "action_query_norm": 43.53119049072266,
      "action_relation_loss": 0.005639960011467338,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.102899906454631,
      "flow_matching_loss": 0.04125418588519096,
      "grad_norm": 0.0989316925406456,
      "h_stock_valid_norm": 97.83460922241211,
      "invalid_tail_samples": 22.7,
      "learning_rate": 7.595318977790408e-05,
      "loss": 0.0469,
      "positive_teacher_pairs": 27106.9,
      "query_h_norm_ratio": 0.44495993554592134,
      "relation_loss": 0.005639960011467338,
      "selected_pairs": 27106.9,
      "step": 10800,
      "student_distance_mean": 26.699501991271973,
      "teacher_distance_mean": 0.8443946719169617,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.005639960011467338,
      "weighted_relation_loss": 0.005639960011467338
    },
    {
      "action_query_batch_std": 0.39792959988117216,
      "action_query_norm": 43.51302642822266,
      "action_relation_loss": 0.005697776004672051,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.112254443405051,
      "flow_matching_loss": 0.0386283740401268,
      "grad_norm": 0.12481214106082916,
      "h_stock_valid_norm": 96.99397277832031,
      "invalid_tail_samples": 23.3,
      "learning_rate": 7.590606470790836e-05,
      "loss": 0.0443,
      "positive_teacher_pairs": 26969.6,
      "query_h_norm_ratio": 0.4486210227012634,
      "relation_loss": 0.005697776004672051,
      "selected_pairs": 26969.6,
      "step": 10810,
      "student_distance_mean": 26.574737358093262,
      "teacher_distance_mean": 0.8534494161605835,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.005697776004672051,
      "weighted_relation_loss": 0.005697776004672051
    },
    {
      "action_query_batch_std": 0.4011968106031418,
      "action_query_norm": 43.51928367614746,
      "action_relation_loss": 0.005414991127327085,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.121608980355472,
      "flow_matching_loss": 0.04045886918902397,
      "grad_norm": 0.1848316788673401,
      "h_stock_valid_norm": 97.08987808227539,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.585890815960125e-05,
      "loss": 0.0459,
      "positive_teacher_pairs": 27032.6,
      "query_h_norm_ratio": 0.4482421845197678,
      "relation_loss": 0.005414991127327085,
      "selected_pairs": 27032.6,
      "step": 10820,
      "student_distance_mean": 26.71781063079834,
      "teacher_distance_mean": 0.8601637005805969,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.005414991127327085,
      "weighted_relation_loss": 0.005414991127327085
    },
    {
      "action_query_batch_std": 0.39736473858356475,
      "action_query_norm": 43.514509201049805,
      "action_relation_loss": 0.005752569297328591,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.130963517305894,
      "flow_matching_loss": 0.03814453221857548,
      "grad_norm": 0.13278736174106598,
      "h_stock_valid_norm": 97.0834342956543,
      "invalid_tail_samples": 22.6,
      "learning_rate": 7.581172019028238e-05,
      "loss": 0.0439,
      "positive_teacher_pairs": 27123.2,
      "query_h_norm_ratio": 0.44821885228157043,
      "relation_loss": 0.005752569297328591,
      "selected_pairs": 27123.2,
      "step": 10830,
      "student_distance_mean": 26.53275089263916,
      "teacher_distance_mean": 0.8467944622039795,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.005752569297328591,
      "weighted_relation_loss": 0.005752569297328591
    },
    {
      "action_query_batch_std": 0.3977074891328812,
      "action_query_norm": 43.50521583557129,
      "action_relation_loss": 0.0058311271946877245,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.140318054256314,
      "flow_matching_loss": 0.04126705080270767,
      "grad_norm": 0.10102858394384384,
      "h_stock_valid_norm": 96.5131217956543,
      "invalid_tail_samples": 22.5,
      "learning_rate": 7.576450085728959e-05,
      "loss": 0.0471,
      "positive_teacher_pairs": 27161.6,
      "query_h_norm_ratio": 0.45077290534973147,
      "relation_loss": 0.0058311271946877245,
      "selected_pairs": 27161.6,
      "step": 10840,
      "student_distance_mean": 26.447687339782714,
      "teacher_distance_mean": 0.8473447620868683,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0058311271946877245,
      "weighted_relation_loss": 0.0058311271946877245
    },
    {
      "action_query_batch_std": 0.40475685298442843,
      "action_query_norm": 43.518731689453126,
      "action_relation_loss": 0.005448401439934969,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.149672591206736,
      "flow_matching_loss": 0.03866420239210129,
      "grad_norm": 0.11909805983304977,
      "h_stock_valid_norm": 97.39655990600586,
      "invalid_tail_samples": 23.1,
      "learning_rate": 7.571725021799885e-05,
      "loss": 0.0441,
      "positive_teacher_pairs": 27010.4,
      "query_h_norm_ratio": 0.4468234747648239,
      "relation_loss": 0.005448401439934969,
      "selected_pairs": 27010.4,
      "step": 10850,
      "student_distance_mean": 26.962746238708498,
      "teacher_distance_mean": 0.8574159979820252,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.005448401439934969,
      "weighted_relation_loss": 0.005448401439934969
    },
    {
      "action_query_batch_std": 0.40190168023109435,
      "action_query_norm": 43.51083641052246,
      "action_relation_loss": 0.005727176740765572,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.159027128157156,
      "flow_matching_loss": 0.03917972445487976,
      "grad_norm": 0.14158621430397034,
      "h_stock_valid_norm": 97.70439224243164,
      "invalid_tail_samples": 24.6,
      "learning_rate": 7.566996832982409e-05,
      "loss": 0.0449,
      "positive_teacher_pairs": 26663.3,
      "query_h_norm_ratio": 0.44533281326293944,
      "relation_loss": 0.005727176740765572,
      "selected_pairs": 26663.3,
      "step": 10860,
      "student_distance_mean": 26.83717498779297,
      "teacher_distance_mean": 0.8552477717399597,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.005727176740765572,
      "weighted_relation_loss": 0.005727176740765572
    },
    {
      "action_query_batch_std": 0.4025778353214264,
      "action_query_norm": 43.503099822998045,
      "action_relation_loss": 0.005651610996574163,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.168381665107578,
      "flow_matching_loss": 0.036273615807294844,
      "grad_norm": 0.11954070627689362,
      "h_stock_valid_norm": 98.68658142089843,
      "invalid_tail_samples": 23.5,
      "learning_rate": 7.56226552502173e-05,
      "loss": 0.0419,
      "positive_teacher_pairs": 26916.8,
      "query_h_norm_ratio": 0.4408253312110901,
      "relation_loss": 0.005651610996574163,
      "selected_pairs": 26916.8,
      "step": 10870,
      "student_distance_mean": 26.762240982055665,
      "teacher_distance_mean": 0.8495739459991455,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.005651610996574163,
      "weighted_relation_loss": 0.005651610996574163
    },
    {
      "action_query_batch_std": 0.40143853724002837,
      "action_query_norm": 43.49663963317871,
      "action_relation_loss": 0.005371365696191788,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.177736202057998,
      "flow_matching_loss": 0.03847011737525463,
      "grad_norm": 0.12288455665111542,
      "h_stock_valid_norm": 99.19411163330078,
      "invalid_tail_samples": 24.2,
      "learning_rate": 7.557531103666833e-05,
      "loss": 0.0438,
      "positive_teacher_pairs": 26753.4,
      "query_h_norm_ratio": 0.43850513696670534,
      "relation_loss": 0.005371365696191788,
      "selected_pairs": 26753.4,
      "step": 10880,
      "student_distance_mean": 26.690612602233887,
      "teacher_distance_mean": 0.8533388555049897,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.005371365696191788,
      "weighted_relation_loss": 0.005371365696191788
    },
    {
      "action_query_batch_std": 0.4038134515285492,
      "action_query_norm": 43.51443176269531,
      "action_relation_loss": 0.005371458875015378,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.187090739008418,
      "flow_matching_loss": 0.039632995054125784,
      "grad_norm": 0.13957059383392334,
      "h_stock_valid_norm": 98.5924903869629,
      "invalid_tail_samples": 22.6,
      "learning_rate": 7.552793574670485e-05,
      "loss": 0.045,
      "positive_teacher_pairs": 27129.9,
      "query_h_norm_ratio": 0.4413612335920334,
      "relation_loss": 0.005371458875015378,
      "selected_pairs": 27129.9,
      "step": 10890,
      "student_distance_mean": 26.90852584838867,
      "teacher_distance_mean": 0.8630773186683655,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.005371458875015378,
      "weighted_relation_loss": 0.005371458875015378
    },
    {
      "action_query_batch_std": 0.4081374049186707,
      "action_query_norm": 43.523787307739255,
      "action_relation_loss": 0.005494784750044346,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.19644527595884,
      "flow_matching_loss": 0.04142856374382973,
      "grad_norm": 0.1439865082502365,
      "h_stock_valid_norm": 98.27024307250977,
      "invalid_tail_samples": 22.3,
      "learning_rate": 7.54805294378923e-05,
      "loss": 0.0469,
      "positive_teacher_pairs": 27207.3,
      "query_h_norm_ratio": 0.44290159046649935,
      "relation_loss": 0.005494784750044346,
      "selected_pairs": 27207.3,
      "step": 10900,
      "student_distance_mean": 27.055902862548827,
      "teacher_distance_mean": 0.8599816262722015,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.005494784750044346,
      "weighted_relation_loss": 0.005494784750044346
    },
    {
      "action_query_batch_std": 0.397214537858963,
      "action_query_norm": 43.49200859069824,
      "action_relation_loss": 0.005606845812872052,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.20579981290926,
      "flow_matching_loss": 0.04192771315574646,
      "grad_norm": 0.15060432255268097,
      "h_stock_valid_norm": 98.66356887817383,
      "invalid_tail_samples": 21.5,
      "learning_rate": 7.543309216783384e-05,
      "loss": 0.0475,
      "positive_teacher_pairs": 27387.4,
      "query_h_norm_ratio": 0.44081607460975647,
      "relation_loss": 0.005606845812872052,
      "selected_pairs": 27387.4,
      "step": 10910,
      "student_distance_mean": 26.35836582183838,
      "teacher_distance_mean": 0.8438433468341827,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.005606845812872052,
      "weighted_relation_loss": 0.005606845812872052
    },
    {
      "action_query_batch_std": 0.4028579145669937,
      "action_query_norm": 43.510603713989255,
      "action_relation_loss": 0.005317165097221732,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.215154349859683,
      "flow_matching_loss": 0.03870058357715607,
      "grad_norm": 0.14082810282707214,
      "h_stock_valid_norm": 99.06065063476562,
      "invalid_tail_samples": 25.0,
      "learning_rate": 7.53856239941702e-05,
      "loss": 0.044,
      "positive_teacher_pairs": 26578.0,
      "query_h_norm_ratio": 0.43923535346984866,
      "relation_loss": 0.005317165097221732,
      "selected_pairs": 26578.0,
      "step": 10920,
      "student_distance_mean": 26.935949516296386,
      "teacher_distance_mean": 0.865744686126709,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.005317165097221732,
      "weighted_relation_loss": 0.005317165097221732
    },
    {
      "action_query_batch_std": 0.4027653932571411,
      "action_query_norm": 43.50189056396484,
      "action_relation_loss": 0.005437653930857778,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.224508886810103,
      "flow_matching_loss": 0.03999660052359104,
      "grad_norm": 0.11410997062921524,
      "h_stock_valid_norm": 98.22679138183594,
      "invalid_tail_samples": 24.2,
      "learning_rate": 7.533812497457972e-05,
      "loss": 0.0454,
      "positive_teacher_pairs": 26759.7,
      "query_h_norm_ratio": 0.44288156628608705,
      "relation_loss": 0.005437653930857778,
      "selected_pairs": 26759.7,
      "step": 10930,
      "student_distance_mean": 26.77728385925293,
      "teacher_distance_mean": 0.8549629211425781,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.005437653930857778,
      "weighted_relation_loss": 0.005437653930857778
    },
    {
      "action_query_batch_std": 0.3997485816478729,
      "action_query_norm": 43.48128318786621,
      "action_relation_loss": 0.005474500311538577,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.233863423760523,
      "flow_matching_loss": 0.03940924741327763,
      "grad_norm": 0.09933727234601974,
      "h_stock_valid_norm": 96.45329360961914,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.529059516677814e-05,
      "loss": 0.0449,
      "positive_teacher_pairs": 27034.5,
      "query_h_norm_ratio": 0.4508209556341171,
      "relation_loss": 0.005474500311538577,
      "selected_pairs": 27034.5,
      "step": 10940,
      "student_distance_mean": 26.59802703857422,
      "teacher_distance_mean": 0.8459855616092682,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.005474500311538577,
      "weighted_relation_loss": 0.005474500311538577
    },
    {
      "action_query_batch_std": 0.40236601531505584,
      "action_query_norm": 43.48918991088867,
      "action_relation_loss": 0.005541802197694778,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.243217960710945,
      "flow_matching_loss": 0.040527719631791116,
      "grad_norm": 0.11603418737649918,
      "h_stock_valid_norm": 96.2376609802246,
      "invalid_tail_samples": 22.6,
      "learning_rate": 7.524303462851872e-05,
      "loss": 0.0461,
      "positive_teacher_pairs": 27127.5,
      "query_h_norm_ratio": 0.4519076615571976,
      "relation_loss": 0.005541802197694778,
      "selected_pairs": 27127.5,
      "step": 10950,
      "student_distance_mean": 26.836058044433592,
      "teacher_distance_mean": 0.8582009077072144,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.005541802197694778,
      "weighted_relation_loss": 0.005541802197694778
    },
    {
      "action_query_batch_std": 0.4047881096601486,
      "action_query_norm": 43.49819450378418,
      "action_relation_loss": 0.0055863774381577965,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.252572497661365,
      "flow_matching_loss": 0.03988020420074463,
      "grad_norm": 0.12462899088859558,
      "h_stock_valid_norm": 97.51371536254882,
      "invalid_tail_samples": 23.2,
      "learning_rate": 7.519544341759192e-05,
      "loss": 0.0455,
      "positive_teacher_pairs": 27010.5,
      "query_h_norm_ratio": 0.4460760891437531,
      "relation_loss": 0.0055863774381577965,
      "selected_pairs": 27010.5,
      "step": 10960,
      "student_distance_mean": 26.9051664352417,
      "teacher_distance_mean": 0.8612121105194092,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0055863774381577965,
      "weighted_relation_loss": 0.0055863774381577965
    },
    {
      "action_query_batch_std": 0.39626764357089994,
      "action_query_norm": 43.468404388427736,
      "action_relation_loss": 0.0055439996533095835,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.261927034611787,
      "flow_matching_loss": 0.040490259602665904,
      "grad_norm": 0.14016544818878174,
      "h_stock_valid_norm": 97.51148681640625,
      "invalid_tail_samples": 25.3,
      "learning_rate": 7.514782159182562e-05,
      "loss": 0.046,
      "positive_teacher_pairs": 26504.1,
      "query_h_norm_ratio": 0.44577980041503906,
      "relation_loss": 0.0055439996533095835,
      "selected_pairs": 26504.1,
      "step": 10970,
      "student_distance_mean": 26.36146354675293,
      "teacher_distance_mean": 0.8485355913639069,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0055439996533095835,
      "weighted_relation_loss": 0.0055439996533095835
    },
    {
      "action_query_batch_std": 0.3991474449634552,
      "action_query_norm": 43.47933807373047,
      "action_relation_loss": 0.005407763458788395,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.271281571562207,
      "flow_matching_loss": 0.03983690664172172,
      "grad_norm": 0.14417128264904022,
      "h_stock_valid_norm": 97.39322814941406,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.510016920908481e-05,
      "loss": 0.0452,
      "positive_teacher_pairs": 27034.4,
      "query_h_norm_ratio": 0.4464351862668991,
      "relation_loss": 0.005407763458788395,
      "selected_pairs": 27034.4,
      "step": 10980,
      "student_distance_mean": 26.626302337646486,
      "teacher_distance_mean": 0.848150759935379,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.005407763458788395,
      "weighted_relation_loss": 0.005407763458788395
    },
    {
      "action_query_batch_std": 0.4021621495485306,
      "action_query_norm": 43.487862777709964,
      "action_relation_loss": 0.005332903843373061,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.28063610851263,
      "flow_matching_loss": 0.038835467956960204,
      "grad_norm": 0.10520128160715103,
      "h_stock_valid_norm": 98.25595703125,
      "invalid_tail_samples": 26.4,
      "learning_rate": 7.505248632727158e-05,
      "loss": 0.0442,
      "positive_teacher_pairs": 26257.4,
      "query_h_norm_ratio": 0.44260563850402834,
      "relation_loss": 0.005332903843373061,
      "selected_pairs": 26257.4,
      "step": 10990,
      "student_distance_mean": 26.86888732910156,
      "teacher_distance_mean": 0.8582810640335083,
      "valid_action_samples": 229.6,
      "weighted_auxiliary_loss": 0.005332903843373061,
      "weighted_relation_loss": 0.005332903843373061
    },
    {
      "action_query_batch_std": 0.3983783423900604,
      "action_query_norm": 43.4716121673584,
      "action_relation_loss": 0.00564521593041718,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.28999064546305,
      "flow_matching_loss": 0.0408014066517353,
      "grad_norm": 0.13335701823234558,
      "h_stock_valid_norm": 99.02645263671874,
      "invalid_tail_samples": 27.1,
      "learning_rate": 7.50047730043252e-05,
      "loss": 0.0464,
      "positive_teacher_pairs": 26094.0,
      "query_h_norm_ratio": 0.4389949142932892,
      "relation_loss": 0.00564521593041718,
      "selected_pairs": 26094.0,
      "step": 11000,
      "student_distance_mean": 26.51783332824707,
      "teacher_distance_mean": 0.8524700999259949,
      "valid_action_samples": 228.9,
      "weighted_auxiliary_loss": 0.00564521593041718,
      "weighted_relation_loss": 0.00564521593041718
    },
    {
      "action_query_batch_std": 0.40079616010189056,
      "action_query_norm": 43.48108406066895,
      "action_relation_loss": 0.005546468868851661,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.29934518241347,
      "flow_matching_loss": 0.03909631296992302,
      "grad_norm": 0.13202618062496185,
      "h_stock_valid_norm": 97.93042221069337,
      "invalid_tail_samples": 25.0,
      "learning_rate": 7.495702929822183e-05,
      "loss": 0.0446,
      "positive_teacher_pairs": 26578.5,
      "query_h_norm_ratio": 0.44401590824127196,
      "relation_loss": 0.005546468868851661,
      "selected_pairs": 26578.5,
      "step": 11010,
      "student_distance_mean": 26.692982482910157,
      "teacher_distance_mean": 0.8577611565589904,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.005546468868851661,
      "weighted_relation_loss": 0.005546468868851661
    },
    {
      "action_query_batch_std": 0.4012141823768616,
      "action_query_norm": 43.4895076751709,
      "action_relation_loss": 0.005678082350641489,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.308699719363892,
      "flow_matching_loss": 0.036723121628165246,
      "grad_norm": 0.13811716437339783,
      "h_stock_valid_norm": 97.5587989807129,
      "invalid_tail_samples": 23.1,
      "learning_rate": 7.490925526697455e-05,
      "loss": 0.0424,
      "positive_teacher_pairs": 27011.2,
      "query_h_norm_ratio": 0.44578289687633516,
      "relation_loss": 0.005678082350641489,
      "selected_pairs": 27011.2,
      "step": 11020,
      "student_distance_mean": 26.64244728088379,
      "teacher_distance_mean": 0.853117561340332,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.005678082350641489,
      "weighted_relation_loss": 0.005678082350641489
    },
    {
      "action_query_batch_std": 0.3995800673961639,
      "action_query_norm": 43.4742992401123,
      "action_relation_loss": 0.005493230139836669,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.318054256314312,
      "flow_matching_loss": 0.04156782291829586,
      "grad_norm": 0.16858841478824615,
      "h_stock_valid_norm": 97.42541580200195,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.486145096863334e-05,
      "loss": 0.0471,
      "positive_teacher_pairs": 27032.0,
      "query_h_norm_ratio": 0.4462360322475433,
      "relation_loss": 0.005493230139836669,
      "selected_pairs": 27032.0,
      "step": 11030,
      "student_distance_mean": 26.57394504547119,
      "teacher_distance_mean": 0.8492246210575104,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.005493230139836669,
      "weighted_relation_loss": 0.005493230139836669
    },
    {
      "action_query_batch_std": 0.39873857200145724,
      "action_query_norm": 43.47280158996582,
      "action_relation_loss": 0.0056062770076096054,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.327408793264734,
      "flow_matching_loss": 0.041188498586416246,
      "grad_norm": 0.12613677978515625,
      "h_stock_valid_norm": 97.85935668945312,
      "invalid_tail_samples": 24.6,
      "learning_rate": 7.481361646128491e-05,
      "loss": 0.0468,
      "positive_teacher_pairs": 26665.8,
      "query_h_norm_ratio": 0.44424614906311033,
      "relation_loss": 0.0056062770076096054,
      "selected_pairs": 26665.8,
      "step": 11040,
      "student_distance_mean": 26.608185005187988,
      "teacher_distance_mean": 0.8562390983104706,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0056062770076096054,
      "weighted_relation_loss": 0.0056062770076096054
    },
    {
      "action_query_batch_std": 0.4035892397165298,
      "action_query_norm": 43.486810302734376,
      "action_relation_loss": 0.005600893497467041,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.336763330215154,
      "flow_matching_loss": 0.04208526015281677,
      "grad_norm": 0.10210476815700531,
      "h_stock_valid_norm": 97.35859069824218,
      "invalid_tail_samples": 23.1,
      "learning_rate": 7.476575180305271e-05,
      "loss": 0.0477,
      "positive_teacher_pairs": 27012.6,
      "query_h_norm_ratio": 0.4466689497232437,
      "relation_loss": 0.005600893497467041,
      "selected_pairs": 27012.6,
      "step": 11050,
      "student_distance_mean": 26.863460731506347,
      "teacher_distance_mean": 0.8572866916656494,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.005600893497467041,
      "weighted_relation_loss": 0.005600893497467041
    },
    {
      "action_query_batch_std": 0.39878080785274506,
      "action_query_norm": 43.46014633178711,
      "action_relation_loss": 0.005647797184064984,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.346117867165574,
      "flow_matching_loss": 0.03891260512173176,
      "grad_norm": 0.13034282624721527,
      "h_stock_valid_norm": 96.02498931884766,
      "invalid_tail_samples": 24.6,
      "learning_rate": 7.471785705209682e-05,
      "loss": 0.0446,
      "positive_teacher_pairs": 26667.4,
      "query_h_norm_ratio": 0.4526042193174362,
      "relation_loss": 0.005647797184064984,
      "selected_pairs": 26667.4,
      "step": 11060,
      "student_distance_mean": 26.511771774291994,
      "teacher_distance_mean": 0.8599551498889924,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.005647797184064984,
      "weighted_relation_loss": 0.005647797184064984
    },
    {
      "action_query_batch_std": 0.3889357656240463,
      "action_query_norm": 43.4447322845459,
      "action_relation_loss": 0.005662374431267381,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.355472404115996,
      "flow_matching_loss": 0.041246752440929416,
      "grad_norm": 0.13440407812595367,
      "h_stock_valid_norm": 96.96752166748047,
      "invalid_tail_samples": 23.6,
      "learning_rate": 7.466993226661387e-05,
      "loss": 0.0469,
      "positive_teacher_pairs": 26899.9,
      "query_h_norm_ratio": 0.4480528563261032,
      "relation_loss": 0.005662374431267381,
      "selected_pairs": 26899.9,
      "step": 11070,
      "student_distance_mean": 25.91212406158447,
      "teacher_distance_mean": 0.8388134479522705,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.005662374431267381,
      "weighted_relation_loss": 0.005662374431267381
    },
    {
      "action_query_batch_std": 0.39946673810482025,
      "action_query_norm": 43.47286605834961,
      "action_relation_loss": 0.00550599074922502,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.364826941066417,
      "flow_matching_loss": 0.04072929583489895,
      "grad_norm": 0.1343105584383011,
      "h_stock_valid_norm": 98.5252586364746,
      "invalid_tail_samples": 22.2,
      "learning_rate": 7.462197750483699e-05,
      "loss": 0.0462,
      "positive_teacher_pairs": 27222.4,
      "query_h_norm_ratio": 0.44123904705047606,
      "relation_loss": 0.00550599074922502,
      "selected_pairs": 27222.4,
      "step": 11080,
      "student_distance_mean": 26.650877952575684,
      "teacher_distance_mean": 0.8602357745170593,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.00550599074922502,
      "weighted_relation_loss": 0.00550599074922502
    },
    {
      "action_query_batch_std": 0.3989076018333435,
      "action_query_norm": 43.453582763671875,
      "action_relation_loss": 0.005545354913920164,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.374181478016839,
      "flow_matching_loss": 0.038376460783183575,
      "grad_norm": 0.1818622648715973,
      "h_stock_valid_norm": 97.46293640136719,
      "invalid_tail_samples": 23.2,
      "learning_rate": 7.457399282503574e-05,
      "loss": 0.0439,
      "positive_teacher_pairs": 27004.6,
      "query_h_norm_ratio": 0.4458646297454834,
      "relation_loss": 0.005545354913920164,
      "selected_pairs": 27004.6,
      "step": 11090,
      "student_distance_mean": 26.531440353393556,
      "teacher_distance_mean": 0.8552614986896515,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.005545354913920164,
      "weighted_relation_loss": 0.005545354913920164
    },
    {
      "action_query_batch_std": 0.40185407102108,
      "action_query_norm": 43.462269973754886,
      "action_relation_loss": 0.005498913861811161,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.383536014967259,
      "flow_matching_loss": 0.04107633493840694,
      "grad_norm": 0.14268824458122253,
      "h_stock_valid_norm": 95.97706756591796,
      "invalid_tail_samples": 21.9,
      "learning_rate": 7.452597828551604e-05,
      "loss": 0.0466,
      "positive_teacher_pairs": 27295.1,
      "query_h_norm_ratio": 0.45284813046455386,
      "relation_loss": 0.005498913861811161,
      "selected_pairs": 27295.1,
      "step": 11100,
      "student_distance_mean": 26.699041175842286,
      "teacher_distance_mean": 0.8580511569976806,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.005498913861811161,
      "weighted_relation_loss": 0.005498913861811161
    },
    {
      "action_query_batch_std": 0.4017152011394501,
      "action_query_norm": 43.47845306396484,
      "action_relation_loss": 0.005631649540737272,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.39289055191768,
      "flow_matching_loss": 0.038430586084723474,
      "grad_norm": 0.11795979738235474,
      "h_stock_valid_norm": 96.53008499145508,
      "invalid_tail_samples": 24.6,
      "learning_rate": 7.447793394462006e-05,
      "loss": 0.0441,
      "positive_teacher_pairs": 26667.7,
      "query_h_norm_ratio": 0.4504228442907333,
      "relation_loss": 0.005631649540737272,
      "selected_pairs": 26667.7,
      "step": 11110,
      "student_distance_mean": 26.746212005615234,
      "teacher_distance_mean": 0.8554830193519593,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.005631649540737272,
      "weighted_relation_loss": 0.005631649540737272
    },
    {
      "action_query_batch_std": 0.3912429749965668,
      "action_query_norm": 43.4380126953125,
      "action_relation_loss": 0.005966619029641152,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.402245088868101,
      "flow_matching_loss": 0.04073215276002884,
      "grad_norm": 0.10672017931938171,
      "h_stock_valid_norm": 98.04125137329102,
      "invalid_tail_samples": 22.1,
      "learning_rate": 7.442985986072624e-05,
      "loss": 0.0467,
      "positive_teacher_pairs": 27247.6,
      "query_h_norm_ratio": 0.4430652499198914,
      "relation_loss": 0.005966619029641152,
      "selected_pairs": 27247.6,
      "step": 11120,
      "student_distance_mean": 26.055989265441895,
      "teacher_distance_mean": 0.8485770404338837,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.005966619029641152,
      "weighted_relation_loss": 0.005966619029641152
    },
    {
      "action_query_batch_std": 0.39656278789043425,
      "action_query_norm": 43.45179443359375,
      "action_relation_loss": 0.005628225859254599,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.411599625818521,
      "flow_matching_loss": 0.03843233026564121,
      "grad_norm": 0.12618058919906616,
      "h_stock_valid_norm": 98.07267227172852,
      "invalid_tail_samples": 27.2,
      "learning_rate": 7.438175609224908e-05,
      "loss": 0.0441,
      "positive_teacher_pairs": 26073.2,
      "query_h_norm_ratio": 0.4430595189332962,
      "relation_loss": 0.005628225859254599,
      "selected_pairs": 26073.2,
      "step": 11130,
      "student_distance_mean": 26.33118667602539,
      "teacher_distance_mean": 0.8497159421443939,
      "valid_action_samples": 228.8,
      "weighted_auxiliary_loss": 0.005628225859254599,
      "weighted_relation_loss": 0.005628225859254599
    },
    {
      "action_query_batch_std": 0.39984543323516847,
      "action_query_norm": 43.45788078308105,
      "action_relation_loss": 0.005609385296702385,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.420954162768943,
      "flow_matching_loss": 0.041105829551815985,
      "grad_norm": 0.10199961066246033,
      "h_stock_valid_norm": 97.86776199340821,
      "invalid_tail_samples": 24.2,
      "learning_rate": 7.433362269763924e-05,
      "loss": 0.0467,
      "positive_teacher_pairs": 26759.0,
      "query_h_norm_ratio": 0.44405058920383456,
      "relation_loss": 0.005609385296702385,
      "selected_pairs": 26759.0,
      "step": 11140,
      "student_distance_mean": 26.709869384765625,
      "teacher_distance_mean": 0.8593740284442901,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.005609385296702385,
      "weighted_relation_loss": 0.005609385296702385
    },
    {
      "action_query_batch_std": 0.3976908653974533,
      "action_query_norm": 43.451886367797854,
      "action_relation_loss": 0.005626597022637725,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.430308699719363,
      "flow_matching_loss": 0.042587728798389436,
      "grad_norm": 0.12636448442935944,
      "h_stock_valid_norm": 97.81168212890626,
      "invalid_tail_samples": 23.4,
      "learning_rate": 7.428545973538329e-05,
      "loss": 0.0482,
      "positive_teacher_pairs": 26951.3,
      "query_h_norm_ratio": 0.4442457675933838,
      "relation_loss": 0.005626597022637725,
      "selected_pairs": 26951.3,
      "step": 11150,
      "student_distance_mean": 26.50030517578125,
      "teacher_distance_mean": 0.8539914846420288,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.005626597022637725,
      "weighted_relation_loss": 0.005626597022637725
    },
    {
      "action_query_batch_std": 0.3974191039800644,
      "action_query_norm": 43.449372100830075,
      "action_relation_loss": 0.005654814606532455,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.439663236669785,
      "flow_matching_loss": 0.03863570913672447,
      "grad_norm": 0.09954041987657547,
      "h_stock_valid_norm": 98.74540939331055,
      "invalid_tail_samples": 24.5,
      "learning_rate": 7.42372672640038e-05,
      "loss": 0.0443,
      "positive_teacher_pairs": 26686.2,
      "query_h_norm_ratio": 0.44001854956150055,
      "relation_loss": 0.005654814606532455,
      "selected_pairs": 26686.2,
      "step": 11160,
      "student_distance_mean": 26.37539920806885,
      "teacher_distance_mean": 0.8515841007232666,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.005654814606532455,
      "weighted_relation_loss": 0.005654814606532455
    },
    {
      "action_query_batch_std": 0.40088979005813596,
      "action_query_norm": 43.46713790893555,
      "action_relation_loss": 0.0054344477131962774,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.449017773620206,
      "flow_matching_loss": 0.03798250369727611,
      "grad_norm": 0.14837944507598877,
      "h_stock_valid_norm": 98.80022430419922,
      "invalid_tail_samples": 23.3,
      "learning_rate": 7.418904534205917e-05,
      "loss": 0.0434,
      "positive_teacher_pairs": 26969.3,
      "query_h_norm_ratio": 0.4399518042802811,
      "relation_loss": 0.0054344477131962774,
      "selected_pairs": 26969.3,
      "step": 11170,
      "student_distance_mean": 26.65992488861084,
      "teacher_distance_mean": 0.8621431946754455,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0054344477131962774,
      "weighted_relation_loss": 0.0054344477131962774
    },
    {
      "action_query_batch_std": 0.3949012517929077,
      "action_query_norm": 43.43818244934082,
      "action_relation_loss": 0.0057740201707929375,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.458372310570628,
      "flow_matching_loss": 0.04108920991420746,
      "grad_norm": 0.12356904149055481,
      "h_stock_valid_norm": 98.78751220703126,
      "invalid_tail_samples": 26.1,
      "learning_rate": 7.414079402814356e-05,
      "loss": 0.0469,
      "positive_teacher_pairs": 26328.1,
      "query_h_norm_ratio": 0.43971874117851256,
      "relation_loss": 0.0057740201707929375,
      "selected_pairs": 26328.1,
      "step": 11180,
      "student_distance_mean": 26.40140857696533,
      "teacher_distance_mean": 0.8522380232810974,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.0057740201707929375,
      "weighted_relation_loss": 0.0057740201707929375
    },
    {
      "action_query_batch_std": 0.4016359359025955,
      "action_query_norm": 43.451713180541994,
      "action_relation_loss": 0.00588544113561511,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.467726847521048,
      "flow_matching_loss": 0.04251212552189827,
      "grad_norm": 0.11775295436382294,
      "h_stock_valid_norm": 98.48682861328125,
      "invalid_tail_samples": 24.5,
      "learning_rate": 7.409251338088687e-05,
      "loss": 0.0484,
      "positive_teacher_pairs": 26691.4,
      "query_h_norm_ratio": 0.4411988347768784,
      "relation_loss": 0.00588544113561511,
      "selected_pairs": 26691.4,
      "step": 11190,
      "student_distance_mean": 26.643288803100585,
      "teacher_distance_mean": 0.8577737033367157,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.00588544113561511,
      "weighted_relation_loss": 0.00588544113561511
    },
    {
      "action_query_batch_std": 0.3993257761001587,
      "action_query_norm": 43.440617370605466,
      "action_relation_loss": 0.005578956566751003,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.477081384471468,
      "flow_matching_loss": 0.039115699008107185,
      "grad_norm": 0.14572732150554657,
      "h_stock_valid_norm": 98.07104187011718,
      "invalid_tail_samples": 23.4,
      "learning_rate": 7.404420345895467e-05,
      "loss": 0.0447,
      "positive_teacher_pairs": 26949.8,
      "query_h_norm_ratio": 0.44295373260974885,
      "relation_loss": 0.005578956566751003,
      "selected_pairs": 26949.8,
      "step": 11200,
      "student_distance_mean": 26.645465469360353,
      "teacher_distance_mean": 0.8533223688602447,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.005578956566751003,
      "weighted_relation_loss": 0.005578956566751003
    },
    {
      "action_query_batch_std": 0.3972890883684158,
      "action_query_norm": 43.43025703430176,
      "action_relation_loss": 0.006043131370097399,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.48643592142189,
      "flow_matching_loss": 0.04031968005001545,
      "grad_norm": 0.11914799362421036,
      "h_stock_valid_norm": 97.79324035644531,
      "invalid_tail_samples": 24.9,
      "learning_rate": 7.399586432104804e-05,
      "loss": 0.0464,
      "positive_teacher_pairs": 26594.2,
      "query_h_norm_ratio": 0.44410858154296873,
      "relation_loss": 0.006043131370097399,
      "selected_pairs": 26594.2,
      "step": 11210,
      "student_distance_mean": 26.435490989685057,
      "teacher_distance_mean": 0.8539423048496246,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.006043131370097399,
      "weighted_relation_loss": 0.006043131370097399
    },
    {
      "action_query_batch_std": 0.4024594098329544,
      "action_query_norm": 43.45586585998535,
      "action_relation_loss": 0.005444720387458801,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.49579045837231,
      "flow_matching_loss": 0.038792827352881434,
      "grad_norm": 0.11447182297706604,
      "h_stock_valid_norm": 98.34508666992187,
      "invalid_tail_samples": 22.0,
      "learning_rate": 7.394749602590358e-05,
      "loss": 0.0442,
      "positive_teacher_pairs": 27272.6,
      "query_h_norm_ratio": 0.4418741941452026,
      "relation_loss": 0.005444720387458801,
      "selected_pairs": 27272.6,
      "step": 11220,
      "student_distance_mean": 26.757485008239748,
      "teacher_distance_mean": 0.8540492653846741,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.005444720387458801,
      "weighted_relation_loss": 0.005444720387458801
    },
    {
      "action_query_batch_std": 0.40306131541728973,
      "action_query_norm": 43.44824638366699,
      "action_relation_loss": 0.005609378824010492,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.505144995322732,
      "flow_matching_loss": 0.03842054232954979,
      "grad_norm": 0.13039253652095795,
      "h_stock_valid_norm": 98.4769790649414,
      "invalid_tail_samples": 22.0,
      "learning_rate": 7.389909863229336e-05,
      "loss": 0.044,
      "positive_teacher_pairs": 27263.8,
      "query_h_norm_ratio": 0.44120272994041443,
      "relation_loss": 0.005609378824010492,
      "selected_pairs": 27263.8,
      "step": 11230,
      "student_distance_mean": 26.884928703308105,
      "teacher_distance_mean": 0.8640173971652985,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.005609378824010492,
      "weighted_relation_loss": 0.005609378824010492
    },
    {
      "action_query_batch_std": 0.3950970947742462,
      "action_query_norm": 43.42394104003906,
      "action_relation_loss": 0.005481587583199143,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.514499532273152,
      "flow_matching_loss": 0.03930031582713127,
      "grad_norm": 0.10858806222677231,
      "h_stock_valid_norm": 98.67640228271485,
      "invalid_tail_samples": 22.0,
      "learning_rate": 7.385067219902477e-05,
      "loss": 0.0448,
      "positive_teacher_pairs": 27270.3,
      "query_h_norm_ratio": 0.4400713235139847,
      "relation_loss": 0.005481587583199143,
      "selected_pairs": 27270.3,
      "step": 11240,
      "student_distance_mean": 26.32529010772705,
      "teacher_distance_mean": 0.847406131029129,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.005481587583199143,
      "weighted_relation_loss": 0.005481587583199143
    },
    {
      "action_query_batch_std": 0.3976445943117142,
      "action_query_norm": 43.43174781799316,
      "action_relation_loss": 0.005745384003967046,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.523854069223573,
      "flow_matching_loss": 0.037809405103325845,
      "grad_norm": 0.16466999053955078,
      "h_stock_valid_norm": 99.14763641357422,
      "invalid_tail_samples": 22.9,
      "learning_rate": 7.380221678494049e-05,
      "loss": 0.0436,
      "positive_teacher_pairs": 27061.9,
      "query_h_norm_ratio": 0.4380535423755646,
      "relation_loss": 0.005745384003967046,
      "selected_pairs": 27061.9,
      "step": 11250,
      "student_distance_mean": 26.54374580383301,
      "teacher_distance_mean": 0.855303817987442,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.005745384003967046,
      "weighted_relation_loss": 0.005745384003967046
    },
    {
      "action_query_batch_std": 0.3957475334405899,
      "action_query_norm": 43.417779541015626,
      "action_relation_loss": 0.0054720448795706035,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.533208606173995,
      "flow_matching_loss": 0.03942074477672577,
      "grad_norm": 0.1201554611325264,
      "h_stock_valid_norm": 98.5764289855957,
      "invalid_tail_samples": 25.0,
      "learning_rate": 7.37537324489184e-05,
      "loss": 0.0449,
      "positive_teacher_pairs": 26575.1,
      "query_h_norm_ratio": 0.44045327603816986,
      "relation_loss": 0.0054720448795706035,
      "selected_pairs": 26575.1,
      "step": 11260,
      "student_distance_mean": 26.30845489501953,
      "teacher_distance_mean": 0.8489171624183655,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0054720448795706035,
      "weighted_relation_loss": 0.0054720448795706035
    },
    {
      "action_query_batch_std": 0.3949801534414291,
      "action_query_norm": 43.41209831237793,
      "action_relation_loss": 0.0056192063260823485,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.542563143124415,
      "flow_matching_loss": 0.03941589929163456,
      "grad_norm": 0.13334135711193085,
      "h_stock_valid_norm": 97.66920700073243,
      "invalid_tail_samples": 26.5,
      "learning_rate": 7.370521924987155e-05,
      "loss": 0.045,
      "positive_teacher_pairs": 26226.8,
      "query_h_norm_ratio": 0.44449091851711275,
      "relation_loss": 0.0056192063260823485,
      "selected_pairs": 26226.8,
      "step": 11270,
      "student_distance_mean": 26.36070327758789,
      "teacher_distance_mean": 0.8482461750507355,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.0056192063260823485,
      "weighted_relation_loss": 0.0056192063260823485
    },
    {
      "action_query_batch_std": 0.4064858973026276,
      "action_query_norm": 43.450317764282225,
      "action_relation_loss": 0.005525414692237973,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.551917680074837,
      "flow_matching_loss": 0.04084364138543606,
      "grad_norm": 0.13975855708122253,
      "h_stock_valid_norm": 97.39785385131836,
      "invalid_tail_samples": 24.6,
      "learning_rate": 7.365667724674806e-05,
      "loss": 0.0464,
      "positive_teacher_pairs": 26668.4,
      "query_h_norm_ratio": 0.4461142897605896,
      "relation_loss": 0.005525414692237973,
      "selected_pairs": 26668.4,
      "step": 11280,
      "student_distance_mean": 27.061362075805665,
      "teacher_distance_mean": 0.8651190161705017,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.005525414692237973,
      "weighted_relation_loss": 0.005525414692237973
    },
    {
      "action_query_batch_std": 0.40100550949573516,
      "action_query_norm": 43.429574966430664,
      "action_relation_loss": 0.005470419442281127,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.561272217025257,
      "flow_matching_loss": 0.039723947644233704,
      "grad_norm": 0.17285460233688354,
      "h_stock_valid_norm": 97.09806213378906,
      "invalid_tail_samples": 24.6,
      "learning_rate": 7.360810649853105e-05,
      "loss": 0.0452,
      "positive_teacher_pairs": 26665.5,
      "query_h_norm_ratio": 0.44727965295314787,
      "relation_loss": 0.005470419442281127,
      "selected_pairs": 26665.5,
      "step": 11290,
      "student_distance_mean": 26.719965553283693,
      "teacher_distance_mean": 0.8588155686855317,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.005470419442281127,
      "weighted_relation_loss": 0.005470419442281127
    },
    {
      "action_query_batch_std": 0.39577949941158297,
      "action_query_norm": 43.40710182189942,
      "action_relation_loss": 0.005406502820551396,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.570626753975677,
      "flow_matching_loss": 0.03797499425709248,
      "grad_norm": 0.11854871362447739,
      "h_stock_valid_norm": 97.48455200195312,
      "invalid_tail_samples": 24.3,
      "learning_rate": 7.355950706423853e-05,
      "loss": 0.0434,
      "positive_teacher_pairs": 26735.5,
      "query_h_norm_ratio": 0.44527637362480166,
      "relation_loss": 0.005406502820551396,
      "selected_pairs": 26735.5,
      "step": 11300,
      "student_distance_mean": 26.35177688598633,
      "teacher_distance_mean": 0.853323882818222,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.005406502820551396,
      "weighted_relation_loss": 0.005406502820551396
    },
    {
      "action_query_batch_std": 0.3951857715845108,
      "action_query_norm": 43.41529655456543,
      "action_relation_loss": 0.005742869852110743,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.5799812909261,
      "flow_matching_loss": 0.03840591926127672,
      "grad_norm": 0.09535308182239532,
      "h_stock_valid_norm": 98.0067741394043,
      "invalid_tail_samples": 23.6,
      "learning_rate": 7.351087900292342e-05,
      "loss": 0.0441,
      "positive_teacher_pairs": 26895.8,
      "query_h_norm_ratio": 0.44298538863658904,
      "relation_loss": 0.005742869852110743,
      "selected_pairs": 26895.8,
      "step": 11310,
      "student_distance_mean": 26.317312240600586,
      "teacher_distance_mean": 0.850238835811615,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.005742869852110743,
      "weighted_relation_loss": 0.005742869852110743
    },
    {
      "action_query_batch_std": 0.3982147812843323,
      "action_query_norm": 43.417538452148435,
      "action_relation_loss": 0.005403858330100775,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.58933582787652,
      "flow_matching_loss": 0.036948786303400996,
      "grad_norm": 0.11359504610300064,
      "h_stock_valid_norm": 97.91726684570312,
      "invalid_tail_samples": 24.2,
      "learning_rate": 7.346222237367339e-05,
      "loss": 0.0424,
      "positive_teacher_pairs": 26762.2,
      "query_h_norm_ratio": 0.4434149295091629,
      "relation_loss": 0.005403858330100775,
      "selected_pairs": 26762.2,
      "step": 11320,
      "student_distance_mean": 26.457823181152342,
      "teacher_distance_mean": 0.8522630512714386,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.005403858330100775,
      "weighted_relation_loss": 0.005403858330100775
    },
    {
      "action_query_batch_std": 0.397231787443161,
      "action_query_norm": 43.41801986694336,
      "action_relation_loss": 0.005435870261862874,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.598690364826941,
      "flow_matching_loss": 0.03907175734639168,
      "grad_norm": 0.10510862618684769,
      "h_stock_valid_norm": 97.84904251098632,
      "invalid_tail_samples": 22.0,
      "learning_rate": 7.341353723561081e-05,
      "loss": 0.0445,
      "positive_teacher_pairs": 27264.6,
      "query_h_norm_ratio": 0.4437298893928528,
      "relation_loss": 0.005435870261862874,
      "selected_pairs": 27264.6,
      "step": 11330,
      "student_distance_mean": 26.435701179504395,
      "teacher_distance_mean": 0.8512579798698425,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.005435870261862874,
      "weighted_relation_loss": 0.005435870261862874
    },
    {
      "action_query_batch_std": 0.3957409828901291,
      "action_query_norm": 43.40039596557617,
      "action_relation_loss": 0.00559621793217957,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.608044901777362,
      "flow_matching_loss": 0.038909298554062846,
      "grad_norm": 0.12127543985843658,
      "h_stock_valid_norm": 97.62198867797852,
      "invalid_tail_samples": 23.3,
      "learning_rate": 7.336482364789277e-05,
      "loss": 0.0445,
      "positive_teacher_pairs": 26961.9,
      "query_h_norm_ratio": 0.44458308815956116,
      "relation_loss": 0.00559621793217957,
      "selected_pairs": 26961.9,
      "step": 11340,
      "student_distance_mean": 26.484485054016112,
      "teacher_distance_mean": 0.856107485294342,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.00559621793217957,
      "weighted_relation_loss": 0.00559621793217957
    },
    {
      "action_query_batch_std": 0.39983078837394714,
      "action_query_norm": 43.41231918334961,
      "action_relation_loss": 0.0053129266947507855,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.617399438727784,
      "flow_matching_loss": 0.03888633940368891,
      "grad_norm": 0.11149301379919052,
      "h_stock_valid_norm": 97.40747146606445,
      "invalid_tail_samples": 25.6,
      "learning_rate": 7.331608166971082e-05,
      "loss": 0.0442,
      "positive_teacher_pairs": 26433.9,
      "query_h_norm_ratio": 0.4456817924976349,
      "relation_loss": 0.0053129266947507855,
      "selected_pairs": 26433.9,
      "step": 11350,
      "student_distance_mean": 26.68557205200195,
      "teacher_distance_mean": 0.8584909200668335,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0053129266947507855,
      "weighted_relation_loss": 0.0053129266947507855
    },
    {
      "action_query_batch_std": 0.3938766986131668,
      "action_query_norm": 43.4004711151123,
      "action_relation_loss": 0.005344880046322942,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.626753975678204,
      "flow_matching_loss": 0.03855610340833664,
      "grad_norm": 0.1014646589756012,
      "h_stock_valid_norm": 96.58154373168945,
      "invalid_tail_samples": 24.3,
      "learning_rate": 7.326731136029108e-05,
      "loss": 0.0439,
      "positive_teacher_pairs": 26737.2,
      "query_h_norm_ratio": 0.44936929941177367,
      "relation_loss": 0.005344880046322942,
      "selected_pairs": 26737.2,
      "step": 11360,
      "student_distance_mean": 26.276711845397948,
      "teacher_distance_mean": 0.8467535376548767,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.005344880046322942,
      "weighted_relation_loss": 0.005344880046322942
    },
    {
      "action_query_batch_std": 0.39451958537101744,
      "action_query_norm": 43.39851684570313,
      "action_relation_loss": 0.0055167937185615305,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.636108512628624,
      "flow_matching_loss": 0.039585322886705396,
      "grad_norm": 0.15469036996364594,
      "h_stock_valid_norm": 97.39295654296875,
      "invalid_tail_samples": 24.1,
      "learning_rate": 7.321851277889408e-05,
      "loss": 0.0451,
      "positive_teacher_pairs": 26780.4,
      "query_h_norm_ratio": 0.4456084817647934,
      "relation_loss": 0.0055167937185615305,
      "selected_pairs": 26780.4,
      "step": 11370,
      "student_distance_mean": 26.263662910461427,
      "teacher_distance_mean": 0.85236074924469,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0055167937185615305,
      "weighted_relation_loss": 0.0055167937185615305
    },
    {
      "action_query_batch_std": 0.3929506540298462,
      "action_query_norm": 43.388740158081056,
      "action_relation_loss": 0.005624583922326565,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.645463049579046,
      "flow_matching_loss": 0.03955766707658768,
      "grad_norm": 0.12930931150913239,
      "h_stock_valid_norm": 98.08514556884765,
      "invalid_tail_samples": 23.5,
      "learning_rate": 7.316968598481469e-05,
      "loss": 0.0452,
      "positive_teacher_pairs": 26923.5,
      "query_h_norm_ratio": 0.44236798882484435,
      "relation_loss": 0.005624583922326565,
      "selected_pairs": 26923.5,
      "step": 11380,
      "student_distance_mean": 26.09217414855957,
      "teacher_distance_mean": 0.8464615225791932,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.005624583922326565,
      "weighted_relation_loss": 0.005624583922326565
    },
    {
      "action_query_batch_std": 0.39520834684371947,
      "action_query_norm": 43.39359703063965,
      "action_relation_loss": 0.00535825495608151,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.654817586529466,
      "flow_matching_loss": 0.039563381485641,
      "grad_norm": 0.12110721319913864,
      "h_stock_valid_norm": 97.20785751342774,
      "invalid_tail_samples": 23.3,
      "learning_rate": 7.312083103738206e-05,
      "loss": 0.0449,
      "positive_teacher_pairs": 26967.1,
      "query_h_norm_ratio": 0.4464077323675156,
      "relation_loss": 0.00535825495608151,
      "selected_pairs": 26967.1,
      "step": 11390,
      "student_distance_mean": 26.285673141479492,
      "teacher_distance_mean": 0.8588544011116028,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.00535825495608151,
      "weighted_relation_loss": 0.00535825495608151
    },
    {
      "action_query_batch_std": 0.3948898732662201,
      "action_query_norm": 43.38780174255371,
      "action_relation_loss": 0.005666306475177407,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.664172123479888,
      "flow_matching_loss": 0.04040917716920376,
      "grad_norm": 0.1195027083158493,
      "h_stock_valid_norm": 97.20913391113281,
      "invalid_tail_samples": 25.4,
      "learning_rate": 7.307194799595958e-05,
      "loss": 0.0461,
      "positive_teacher_pairs": 26484.3,
      "query_h_norm_ratio": 0.44634685218334197,
      "relation_loss": 0.005666306475177407,
      "selected_pairs": 26484.3,
      "step": 11400,
      "student_distance_mean": 26.257622909545898,
      "teacher_distance_mean": 0.8608473777770996,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.005666306475177407,
      "weighted_relation_loss": 0.005666306475177407
    },
    {
      "action_query_batch_std": 0.40112208425998686,
      "action_query_norm": 43.40962677001953,
      "action_relation_loss": 0.0053189839236438274,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.673526660430309,
      "flow_matching_loss": 0.037363117560744286,
      "grad_norm": 0.13891556859016418,
      "h_stock_valid_norm": 97.77227630615235,
      "invalid_tail_samples": 25.3,
      "learning_rate": 7.302303691994473e-05,
      "loss": 0.0427,
      "positive_teacher_pairs": 26506.0,
      "query_h_norm_ratio": 0.44399007260799406,
      "relation_loss": 0.0053189839236438274,
      "selected_pairs": 26506.0,
      "step": 11410,
      "student_distance_mean": 26.700860214233398,
      "teacher_distance_mean": 0.8692938268184662,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0053189839236438274,
      "weighted_relation_loss": 0.0053189839236438274
    },
    {
      "action_query_batch_std": 0.3914316713809967,
      "action_query_norm": 43.378458404541014,
      "action_relation_loss": 0.005183294368907809,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.68288119738073,
      "flow_matching_loss": 0.03837498985230923,
      "grad_norm": 0.13877524435520172,
      "h_stock_valid_norm": 97.45549774169922,
      "invalid_tail_samples": 25.6,
      "learning_rate": 7.29740978687691e-05,
      "loss": 0.0436,
      "positive_teacher_pairs": 26430.6,
      "query_h_norm_ratio": 0.4451185017824173,
      "relation_loss": 0.005183294368907809,
      "selected_pairs": 26430.6,
      "step": 11420,
      "student_distance_mean": 26.185778427124024,
      "teacher_distance_mean": 0.8578297555446625,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.005183294368907809,
      "weighted_relation_loss": 0.005183294368907809
    },
    {
      "action_query_batch_std": 0.3934828996658325,
      "action_query_norm": 43.38559150695801,
      "action_relation_loss": 0.005620965315029025,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.69223573433115,
      "flow_matching_loss": 0.03619625810533762,
      "grad_norm": 0.11119335144758224,
      "h_stock_valid_norm": 97.24398574829101,
      "invalid_tail_samples": 24.6,
      "learning_rate": 7.292513090189825e-05,
      "loss": 0.0418,
      "positive_teacher_pairs": 26664.2,
      "query_h_norm_ratio": 0.44615519642829893,
      "relation_loss": 0.005620965315029025,
      "selected_pairs": 26664.2,
      "step": 11430,
      "student_distance_mean": 26.205932235717775,
      "teacher_distance_mean": 0.8540283203125,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.005620965315029025,
      "weighted_relation_loss": 0.005620965315029025
    },
    {
      "action_query_batch_std": 0.3944386124610901,
      "action_query_norm": 43.37530784606933,
      "action_relation_loss": 0.005533705232664943,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.701590271281571,
      "flow_matching_loss": 0.03784457817673683,
      "grad_norm": 0.1367662400007248,
      "h_stock_valid_norm": 97.29236068725587,
      "invalid_tail_samples": 22.9,
      "learning_rate": 7.287613607883163e-05,
      "loss": 0.0434,
      "positive_teacher_pairs": 27060.1,
      "query_h_norm_ratio": 0.44582839906215666,
      "relation_loss": 0.005533705232664943,
      "selected_pairs": 27060.1,
      "step": 11440,
      "student_distance_mean": 26.282774353027342,
      "teacher_distance_mean": 0.8544480800628662,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.005533705232664943,
      "weighted_relation_loss": 0.005533705232664943
    },
    {
      "action_query_batch_std": 0.40113445222377775,
      "action_query_norm": 43.393463134765625,
      "action_relation_loss": 0.005712818028405309,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.710944808231993,
      "flow_matching_loss": 0.037330925837159155,
      "grad_norm": 0.10021620988845825,
      "h_stock_valid_norm": 97.44545516967773,
      "invalid_tail_samples": 21.4,
      "learning_rate": 7.282711345910263e-05,
      "loss": 0.043,
      "positive_teacher_pairs": 27407.9,
      "query_h_norm_ratio": 0.44531471729278566,
      "relation_loss": 0.005712818028405309,
      "selected_pairs": 27407.9,
      "step": 11450,
      "student_distance_mean": 26.54774627685547,
      "teacher_distance_mean": 0.8588090658187866,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.005712818028405309,
      "weighted_relation_loss": 0.005712818028405309
    },
    {
      "action_query_batch_std": 0.39250077605247496,
      "action_query_norm": 43.378308868408205,
      "action_relation_loss": 0.005407335748896003,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.720299345182413,
      "flow_matching_loss": 0.03856344483792782,
      "grad_norm": 0.1566353589296341,
      "h_stock_valid_norm": 97.55082855224609,
      "invalid_tail_samples": 23.1,
      "learning_rate": 7.27780631022783e-05,
      "loss": 0.044,
      "positive_teacher_pairs": 27029.1,
      "query_h_norm_ratio": 0.4446804732084274,
      "relation_loss": 0.005407335748896003,
      "selected_pairs": 27029.1,
      "step": 11460,
      "student_distance_mean": 26.174481010437013,
      "teacher_distance_mean": 0.852228581905365,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.005407335748896003,
      "weighted_relation_loss": 0.005407335748896003
    },
    {
      "action_query_batch_std": 0.3937743782997131,
      "action_query_norm": 43.3690990447998,
      "action_relation_loss": 0.005343320919200778,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.729653882132835,
      "flow_matching_loss": 0.035520962253212926,
      "grad_norm": 0.13851694762706757,
      "h_stock_valid_norm": 97.59963912963867,
      "invalid_tail_samples": 24.1,
      "learning_rate": 7.272898506795948e-05,
      "loss": 0.0409,
      "positive_teacher_pairs": 26784.8,
      "query_h_norm_ratio": 0.44436080753803253,
      "relation_loss": 0.005343320919200778,
      "selected_pairs": 26784.8,
      "step": 11470,
      "student_distance_mean": 26.096435737609863,
      "teacher_distance_mean": 0.8464912354946137,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.005343320919200778,
      "weighted_relation_loss": 0.005343320919200778
    },
    {
      "action_query_batch_std": 0.4005725860595703,
      "action_query_norm": 43.389608001708986,
      "action_relation_loss": 0.005372336506843567,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.739008419083255,
      "flow_matching_loss": 0.03693309649825096,
      "grad_norm": 0.10480175167322159,
      "h_stock_valid_norm": 98.0125,
      "invalid_tail_samples": 24.6,
      "learning_rate": 7.267987941578058e-05,
      "loss": 0.0423,
      "positive_teacher_pairs": 26668.7,
      "query_h_norm_ratio": 0.4426958441734314,
      "relation_loss": 0.005372336506843567,
      "selected_pairs": 26668.7,
      "step": 11480,
      "student_distance_mean": 26.6458438873291,
      "teacher_distance_mean": 0.8543910682201385,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.005372336506843567,
      "weighted_relation_loss": 0.005372336506843567
    },
    {
      "action_query_batch_std": 0.3972699522972107,
      "action_query_norm": 43.3855484008789,
      "action_relation_loss": 0.005330456327646971,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.748362956033676,
      "flow_matching_loss": 0.039320912770926954,
      "grad_norm": 0.13176625967025757,
      "h_stock_valid_norm": 97.49301452636719,
      "invalid_tail_samples": 22.2,
      "learning_rate": 7.263074620540962e-05,
      "loss": 0.0447,
      "positive_teacher_pairs": 27226.3,
      "query_h_norm_ratio": 0.4450137346982956,
      "relation_loss": 0.005330456327646971,
      "selected_pairs": 27226.3,
      "step": 11490,
      "student_distance_mean": 26.514300918579103,
      "teacher_distance_mean": 0.8648564755916596,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.005330456327646971,
      "weighted_relation_loss": 0.005330456327646971
    },
    {
      "action_query_batch_std": 0.3982286751270294,
      "action_query_norm": 43.390214920043945,
      "action_relation_loss": 0.005445822514593602,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.757717492984098,
      "flow_matching_loss": 0.037696986272931096,
      "grad_norm": 0.15663301944732666,
      "h_stock_valid_norm": 97.85044021606446,
      "invalid_tail_samples": 24.3,
      "learning_rate": 7.25815854965481e-05,
      "loss": 0.0431,
      "positive_teacher_pairs": 26736.3,
      "query_h_norm_ratio": 0.4434385269880295,
      "relation_loss": 0.005445822514593602,
      "selected_pairs": 26736.3,
      "step": 11500,
      "student_distance_mean": 26.532129096984864,
      "teacher_distance_mean": 0.8544734239578247,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.005445822514593602,
      "weighted_relation_loss": 0.005445822514593602
    },
    {
      "action_query_batch_std": 0.39486391842365265,
      "action_query_norm": 43.36747093200684,
      "action_relation_loss": 0.0054876433685421945,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.767072029934518,
      "flow_matching_loss": 0.03853655271232128,
      "grad_norm": 0.10974443703889847,
      "h_stock_valid_norm": 97.64218521118164,
      "invalid_tail_samples": 23.9,
      "learning_rate": 7.253239734893089e-05,
      "loss": 0.044,
      "positive_teacher_pairs": 26833.8,
      "query_h_norm_ratio": 0.44415052235126495,
      "relation_loss": 0.0054876433685421945,
      "selected_pairs": 26833.8,
      "step": 11510,
      "student_distance_mean": 26.32802963256836,
      "teacher_distance_mean": 0.847986352443695,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0054876433685421945,
      "weighted_relation_loss": 0.0054876433685421945
    },
    {
      "action_query_batch_std": 0.397415891289711,
      "action_query_norm": 43.38536262512207,
      "action_relation_loss": 0.005304489145055413,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.77642656688494,
      "flow_matching_loss": 0.03882186748087406,
      "grad_norm": 0.1116744875907898,
      "h_stock_valid_norm": 97.92800827026367,
      "invalid_tail_samples": 25.6,
      "learning_rate": 7.248318182232623e-05,
      "loss": 0.0441,
      "positive_teacher_pairs": 26438.2,
      "query_h_norm_ratio": 0.44303747117519376,
      "relation_loss": 0.005304489145055413,
      "selected_pairs": 26438.2,
      "step": 11520,
      "student_distance_mean": 26.495722579956055,
      "teacher_distance_mean": 0.859623658657074,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.005304489145055413,
      "weighted_relation_loss": 0.005304489145055413
    },
    {
      "action_query_batch_std": 0.4044923484325409,
      "action_query_norm": 43.39791412353516,
      "action_relation_loss": 0.005154945375397802,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.78578110383536,
      "flow_matching_loss": 0.03829097300767899,
      "grad_norm": 0.10175664722919464,
      "h_stock_valid_norm": 97.7290657043457,
      "invalid_tail_samples": 23.7,
      "learning_rate": 7.243393897653565e-05,
      "loss": 0.0434,
      "positive_teacher_pairs": 26876.3,
      "query_h_norm_ratio": 0.44407233595848083,
      "relation_loss": 0.005154945375397802,
      "selected_pairs": 26876.3,
      "step": 11530,
      "student_distance_mean": 26.927593421936034,
      "teacher_distance_mean": 0.8685523629188537,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.005154945375397802,
      "weighted_relation_loss": 0.005154945375397802
    },
    {
      "action_query_batch_std": 0.3951113820075989,
      "action_query_norm": 43.37297668457031,
      "action_relation_loss": 0.005295793060213327,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.795135640785782,
      "flow_matching_loss": 0.03753584250807762,
      "grad_norm": 0.11581915616989136,
      "h_stock_valid_norm": 96.76932525634766,
      "invalid_tail_samples": 21.1,
      "learning_rate": 7.23846688713938e-05,
      "loss": 0.0428,
      "positive_teacher_pairs": 27482.0,
      "query_h_norm_ratio": 0.44821334779262545,
      "relation_loss": 0.005295793060213327,
      "selected_pairs": 27482.0,
      "step": 11540,
      "student_distance_mean": 26.317764663696288,
      "teacher_distance_mean": 0.8504108428955078,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.005295793060213327,
      "weighted_relation_loss": 0.005295793060213327
    },
    {
      "action_query_batch_std": 0.39630702435970305,
      "action_query_norm": 43.37300567626953,
      "action_relation_loss": 0.005253634648397565,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.804490177736202,
      "flow_matching_loss": 0.03856731802225113,
      "grad_norm": 0.10914041101932526,
      "h_stock_valid_norm": 96.70830383300782,
      "invalid_tail_samples": 22.4,
      "learning_rate": 7.233537156676854e-05,
      "loss": 0.0438,
      "positive_teacher_pairs": 27172.1,
      "query_h_norm_ratio": 0.44849691689014437,
      "relation_loss": 0.005253634648397565,
      "selected_pairs": 27172.1,
      "step": 11550,
      "student_distance_mean": 26.19786319732666,
      "teacher_distance_mean": 0.850616317987442,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.005253634648397565,
      "weighted_relation_loss": 0.005253634648397565
    },
    {
      "action_query_batch_std": 0.39401452243328094,
      "action_query_norm": 43.36951751708985,
      "action_relation_loss": 0.005452852696180344,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.813844714686622,
      "flow_matching_loss": 0.0378058772534132,
      "grad_norm": 0.11577805131673813,
      "h_stock_valid_norm": 96.73747482299805,
      "invalid_tail_samples": 25.7,
      "learning_rate": 7.228604712256076e-05,
      "loss": 0.0433,
      "positive_teacher_pairs": 26418.3,
      "query_h_norm_ratio": 0.4483247876167297,
      "relation_loss": 0.005452852696180344,
      "selected_pairs": 26418.3,
      "step": 11560,
      "student_distance_mean": 26.357403182983397,
      "teacher_distance_mean": 0.8511880815029145,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.005452852696180344,
      "weighted_relation_loss": 0.005452852696180344
    },
    {
      "action_query_batch_std": 0.39936298727989195,
      "action_query_norm": 43.37478942871094,
      "action_relation_loss": 0.005335966683924198,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.823199251637044,
      "flow_matching_loss": 0.03746623247861862,
      "grad_norm": 0.08953700214624405,
      "h_stock_valid_norm": 96.4037254333496,
      "invalid_tail_samples": 23.6,
      "learning_rate": 7.223669559870426e-05,
      "loss": 0.0428,
      "positive_teacher_pairs": 26900.5,
      "query_h_norm_ratio": 0.4499324053525925,
      "relation_loss": 0.005335966683924198,
      "selected_pairs": 26900.5,
      "step": 11570,
      "student_distance_mean": 26.410520172119142,
      "teacher_distance_mean": 0.8606429159641266,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.005335966683924198,
      "weighted_relation_loss": 0.005335966683924198
    },
    {
      "action_query_batch_std": 0.3934097796678543,
      "action_query_norm": 43.357137298583986,
      "action_relation_loss": 0.00550535349175334,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.832553788587465,
      "flow_matching_loss": 0.03793460391461849,
      "grad_norm": 0.0989905372262001,
      "h_stock_valid_norm": 96.3583984375,
      "invalid_tail_samples": 24.3,
      "learning_rate": 7.218731705516585e-05,
      "loss": 0.0434,
      "positive_teacher_pairs": 26733.7,
      "query_h_norm_ratio": 0.4499613970518112,
      "relation_loss": 0.00550535349175334,
      "selected_pairs": 26733.7,
      "step": 11580,
      "student_distance_mean": 26.118821716308595,
      "teacher_distance_mean": 0.8507877707481384,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.00550535349175334,
      "weighted_relation_loss": 0.00550535349175334
    },
    {
      "action_query_batch_std": 0.3994679987430573,
      "action_query_norm": 43.37953605651855,
      "action_relation_loss": 0.005134207662194967,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.841908325537887,
      "flow_matching_loss": 0.037846092507243155,
      "grad_norm": 0.11969675123691559,
      "h_stock_valid_norm": 96.41016006469727,
      "invalid_tail_samples": 23.2,
      "learning_rate": 7.21379115519451e-05,
      "loss": 0.043,
      "positive_teacher_pairs": 26987.5,
      "query_h_norm_ratio": 0.4499505192041397,
      "relation_loss": 0.005134207662194967,
      "selected_pairs": 26987.5,
      "step": 11590,
      "student_distance_mean": 26.611955451965333,
      "teacher_distance_mean": 0.861273729801178,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.005134207662194967,
      "weighted_relation_loss": 0.005134207662194967
    },
    {
      "action_query_batch_std": 0.3970337510108948,
      "action_query_norm": 43.37481575012207,
      "action_relation_loss": 0.005446944106370211,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.851262862488307,
      "flow_matching_loss": 0.038362330943346026,
      "grad_norm": 0.12598532438278198,
      "h_stock_valid_norm": 96.92995910644531,
      "invalid_tail_samples": 21.9,
      "learning_rate": 7.20884791490743e-05,
      "loss": 0.0438,
      "positive_teacher_pairs": 27298.1,
      "query_h_norm_ratio": 0.44749996066093445,
      "relation_loss": 0.005446944106370211,
      "selected_pairs": 27298.1,
      "step": 11600,
      "student_distance_mean": 26.54110469818115,
      "teacher_distance_mean": 0.8628146529197693,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.005446944106370211,
      "weighted_relation_loss": 0.005446944106370211
    },
    {
      "action_query_batch_std": 0.39089569747447966,
      "action_query_norm": 43.35073585510254,
      "action_relation_loss": 0.005451801093295216,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.860617399438727,
      "flow_matching_loss": 0.03870668224990368,
      "grad_norm": 0.1336631178855896,
      "h_stock_valid_norm": 97.93713684082032,
      "invalid_tail_samples": 23.6,
      "learning_rate": 7.203901990661857e-05,
      "loss": 0.0442,
      "positive_teacher_pairs": 26891.0,
      "query_h_norm_ratio": 0.44264284074306487,
      "relation_loss": 0.005451801093295216,
      "selected_pairs": 26891.0,
      "step": 11610,
      "student_distance_mean": 26.079194831848145,
      "teacher_distance_mean": 0.8529490411281586,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.005451801093295216,
      "weighted_relation_loss": 0.005451801093295216
    },
    {
      "action_query_batch_std": 0.3946472704410553,
      "action_query_norm": 43.35773887634277,
      "action_relation_loss": 0.00529144643805921,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.869971936389149,
      "flow_matching_loss": 0.037245315685868266,
      "grad_norm": 0.14054147899150848,
      "h_stock_valid_norm": 98.37749404907227,
      "invalid_tail_samples": 24.3,
      "learning_rate": 7.198953388467549e-05,
      "loss": 0.0425,
      "positive_teacher_pairs": 26734.3,
      "query_h_norm_ratio": 0.4407316744327545,
      "relation_loss": 0.00529144643805921,
      "selected_pairs": 26734.3,
      "step": 11620,
      "student_distance_mean": 26.32527084350586,
      "teacher_distance_mean": 0.8574946641921997,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.00529144643805921,
      "weighted_relation_loss": 0.00529144643805921
    },
    {
      "action_query_batch_std": 0.39740959405899046,
      "action_query_norm": 43.35948867797852,
      "action_relation_loss": 0.005304268095642328,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.87932647333957,
      "flow_matching_loss": 0.04019889123737812,
      "grad_norm": 0.10633699595928192,
      "h_stock_valid_norm": 97.91376419067383,
      "invalid_tail_samples": 23.1,
      "learning_rate": 7.194002114337526e-05,
      "loss": 0.0455,
      "positive_teacher_pairs": 27008.5,
      "query_h_norm_ratio": 0.4428375571966171,
      "relation_loss": 0.005304268095642328,
      "selected_pairs": 27008.5,
      "step": 11630,
      "student_distance_mean": 26.387920951843263,
      "teacher_distance_mean": 0.8608181118965149,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.005304268095642328,
      "weighted_relation_loss": 0.005304268095642328
    },
    {
      "action_query_batch_std": 0.3961388021707535,
      "action_query_norm": 43.35915870666504,
      "action_relation_loss": 0.005553642194718123,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.888681010289991,
      "flow_matching_loss": 0.03814212828874588,
      "grad_norm": 0.12802089750766754,
      "h_stock_valid_norm": 97.24991455078126,
      "invalid_tail_samples": 23.8,
      "learning_rate": 7.189048174288054e-05,
      "loss": 0.0437,
      "positive_teacher_pairs": 26854.2,
      "query_h_norm_ratio": 0.4458577811717987,
      "relation_loss": 0.005553642194718123,
      "selected_pairs": 26854.2,
      "step": 11640,
      "student_distance_mean": 26.32182788848877,
      "teacher_distance_mean": 0.8548666894435882,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.005553642194718123,
      "weighted_relation_loss": 0.005553642194718123
    },
    {
      "action_query_batch_std": 0.39392470121383666,
      "action_query_norm": 43.355831909179685,
      "action_relation_loss": 0.005673520406708122,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.898035547240411,
      "flow_matching_loss": 0.03942546434700489,
      "grad_norm": 0.08394467830657959,
      "h_stock_valid_norm": 97.32002182006836,
      "invalid_tail_samples": 26.2,
      "learning_rate": 7.184091574338636e-05,
      "loss": 0.0451,
      "positive_teacher_pairs": 26297.4,
      "query_h_norm_ratio": 0.44549948573112486,
      "relation_loss": 0.005673520406708122,
      "selected_pairs": 26297.4,
      "step": 11650,
      "student_distance_mean": 26.22248363494873,
      "teacher_distance_mean": 0.8543736100196838,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.005673520406708122,
      "weighted_relation_loss": 0.005673520406708122
    },
    {
      "action_query_batch_std": 0.39625310301780703,
      "action_query_norm": 43.3597957611084,
      "action_relation_loss": 0.005439593736082315,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.907390084190833,
      "flow_matching_loss": 0.03741797581315041,
      "grad_norm": 0.1354467123746872,
      "h_stock_valid_norm": 97.51900177001953,
      "invalid_tail_samples": 24.9,
      "learning_rate": 7.179132320512009e-05,
      "loss": 0.0429,
      "positive_teacher_pairs": 26598.3,
      "query_h_norm_ratio": 0.44463412165641786,
      "relation_loss": 0.005439593736082315,
      "selected_pairs": 26598.3,
      "step": 11660,
      "student_distance_mean": 26.474971961975097,
      "teacher_distance_mean": 0.8550334513187409,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.005439593736082315,
      "weighted_relation_loss": 0.005439593736082315
    },
    {
      "action_query_batch_std": 0.3959159940481186,
      "action_query_norm": 43.3543399810791,
      "action_relation_loss": 0.005339613230898976,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.916744621141254,
      "flow_matching_loss": 0.03705797865986824,
      "grad_norm": 0.12573325634002686,
      "h_stock_valid_norm": 97.71460418701172,
      "invalid_tail_samples": 21.8,
      "learning_rate": 7.174170418834134e-05,
      "loss": 0.0424,
      "positive_teacher_pairs": 27315.0,
      "query_h_norm_ratio": 0.4436886668205261,
      "relation_loss": 0.005339613230898976,
      "selected_pairs": 27315.0,
      "step": 11670,
      "student_distance_mean": 26.356549644470213,
      "teacher_distance_mean": 0.8480920016765594,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.005339613230898976,
      "weighted_relation_loss": 0.005339613230898976
    },
    {
      "action_query_batch_std": 0.39148551821708677,
      "action_query_norm": 43.33482475280762,
      "action_relation_loss": 0.0050859999842941764,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.926099158091674,
      "flow_matching_loss": 0.03891835734248161,
      "grad_norm": 0.12806111574172974,
      "h_stock_valid_norm": 97.10193481445313,
      "invalid_tail_samples": 21.5,
      "learning_rate": 7.169205875334189e-05,
      "loss": 0.044,
      "positive_teacher_pairs": 27383.0,
      "query_h_norm_ratio": 0.44629454612731934,
      "relation_loss": 0.0050859999842941764,
      "selected_pairs": 27383.0,
      "step": 11680,
      "student_distance_mean": 26.02122554779053,
      "teacher_distance_mean": 0.8554121732711792,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.0050859999842941764,
      "weighted_relation_loss": 0.0050859999842941764
    },
    {
      "action_query_batch_std": 0.39392242431640623,
      "action_query_norm": 43.34520950317383,
      "action_relation_loss": 0.005213543632999062,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.935453695042096,
      "flow_matching_loss": 0.03683587536215782,
      "grad_norm": 0.13297410309314728,
      "h_stock_valid_norm": 96.67772750854492,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.164238696044562e-05,
      "loss": 0.042,
      "positive_teacher_pairs": 27036.3,
      "query_h_norm_ratio": 0.4483515053987503,
      "relation_loss": 0.005213543632999062,
      "selected_pairs": 27036.3,
      "step": 11690,
      "student_distance_mean": 26.157705116271973,
      "teacher_distance_mean": 0.8590748608112335,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.005213543632999062,
      "weighted_relation_loss": 0.005213543632999062
    },
    {
      "action_query_batch_std": 0.39196884334087373,
      "action_query_norm": 43.33218498229981,
      "action_relation_loss": 0.005131010664626956,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.944808231992516,
      "flow_matching_loss": 0.03682214096188545,
      "grad_norm": 0.10351303964853287,
      "h_stock_valid_norm": 97.03767929077148,
      "invalid_tail_samples": 26.0,
      "learning_rate": 7.159268887000846e-05,
      "loss": 0.042,
      "positive_teacher_pairs": 26338.3,
      "query_h_norm_ratio": 0.4465607166290283,
      "relation_loss": 0.005131010664626956,
      "selected_pairs": 26338.3,
      "step": 11700,
      "student_distance_mean": 26.234356307983397,
      "teacher_distance_mean": 0.8590348958969116,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.005131010664626956,
      "weighted_relation_loss": 0.005131010664626956
    },
    {
      "action_query_batch_std": 0.39484746754169464,
      "action_query_norm": 43.34145965576172,
      "action_relation_loss": 0.005250862101092935,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.954162768942938,
      "flow_matching_loss": 0.040012812986969945,
      "grad_norm": 0.14872398972511292,
      "h_stock_valid_norm": 98.24918670654297,
      "invalid_tail_samples": 21.9,
      "learning_rate": 7.154296454241827e-05,
      "loss": 0.0453,
      "positive_teacher_pairs": 27289.7,
      "query_h_norm_ratio": 0.441150838136673,
      "relation_loss": 0.005250862101092935,
      "selected_pairs": 27289.7,
      "step": 11710,
      "student_distance_mean": 26.37118453979492,
      "teacher_distance_mean": 0.8570830762386322,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.005250862101092935,
      "weighted_relation_loss": 0.005250862101092935
    },
    {
      "action_query_batch_std": 0.39435208439826963,
      "action_query_norm": 43.32897033691406,
      "action_relation_loss": 0.005122712906450033,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.963517305893358,
      "flow_matching_loss": 0.036429771780967714,
      "grad_norm": 0.1533401906490326,
      "h_stock_valid_norm": 99.20276412963867,
      "invalid_tail_samples": 22.3,
      "learning_rate": 7.149321403809479e-05,
      "loss": 0.0416,
      "positive_teacher_pairs": 27196.9,
      "query_h_norm_ratio": 0.4367742300033569,
      "relation_loss": 0.005122712906450033,
      "selected_pairs": 27196.9,
      "step": 11720,
      "student_distance_mean": 26.093544006347656,
      "teacher_distance_mean": 0.8533395886421203,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.005122712906450033,
      "weighted_relation_loss": 0.005122712906450033
    },
    {
      "action_query_batch_std": 0.3929480344057083,
      "action_query_norm": 43.32135238647461,
      "action_relation_loss": 0.0052759750746190544,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.97287184284378,
      "flow_matching_loss": 0.03762905262410641,
      "grad_norm": 0.12885552644729614,
      "h_stock_valid_norm": 99.21853408813476,
      "invalid_tail_samples": 22.7,
      "learning_rate": 7.144343741748956e-05,
      "loss": 0.0429,
      "positive_teacher_pairs": 27101.6,
      "query_h_norm_ratio": 0.4366271734237671,
      "relation_loss": 0.0052759750746190544,
      "selected_pairs": 27101.6,
      "step": 11730,
      "student_distance_mean": 26.061008262634278,
      "teacher_distance_mean": 0.8548927664756775,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0052759750746190544,
      "weighted_relation_loss": 0.0052759750746190544
    },
    {
      "action_query_batch_std": 0.39575866758823397,
      "action_query_norm": 43.32905578613281,
      "action_relation_loss": 0.005597927374765277,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.9822263797942,
      "flow_matching_loss": 0.039270567893981936,
      "grad_norm": 0.09251406788825989,
      "h_stock_valid_norm": 98.51119842529297,
      "invalid_tail_samples": 26.9,
      "learning_rate": 7.139363474108589e-05,
      "loss": 0.0449,
      "positive_teacher_pairs": 26139.4,
      "query_h_norm_ratio": 0.4398433566093445,
      "relation_loss": 0.005597927374765277,
      "selected_pairs": 26139.4,
      "step": 11740,
      "student_distance_mean": 26.356848335266115,
      "teacher_distance_mean": 0.8564966440200805,
      "valid_action_samples": 229.1,
      "weighted_auxiliary_loss": 0.005597927374765277,
      "weighted_relation_loss": 0.005597927374765277
    },
    {
      "action_query_batch_std": 0.3937501072883606,
      "action_query_norm": 43.31826820373535,
      "action_relation_loss": 0.005252562463283539,
      "auxiliary_loss_weight": 1.0,
      "epoch": 10.99158091674462,
      "flow_matching_loss": 0.03832867406308651,
      "grad_norm": 0.1540326625108719,
      "h_stock_valid_norm": 97.9691390991211,
      "invalid_tail_samples": 24.5,
      "learning_rate": 7.134380606939874e-05,
      "loss": 0.0436,
      "positive_teacher_pairs": 26686.4,
      "query_h_norm_ratio": 0.4421675235033035,
      "relation_loss": 0.005252562463283539,
      "selected_pairs": 26686.4,
      "step": 11750,
      "student_distance_mean": 26.193461990356447,
      "teacher_distance_mean": 0.8528771162033081,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.005252562463283539,
      "weighted_relation_loss": 0.005252562463283539
    },
    {
      "action_query_batch_std": 0.39541733264923096,
      "action_query_norm": 43.32230224609375,
      "action_relation_loss": 0.0054823501966893675,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.000935453695043,
      "flow_matching_loss": 0.03933380246162414,
      "grad_norm": 0.12655167281627655,
      "h_stock_valid_norm": 97.69266052246094,
      "invalid_tail_samples": 20.7,
      "learning_rate": 7.12939514629746e-05,
      "loss": 0.0448,
      "positive_teacher_pairs": 24618.5,
      "query_h_norm_ratio": 0.4434614539146423,
      "relation_loss": 0.0054823501966893675,
      "selected_pairs": 24618.5,
      "step": 11760,
      "student_distance_mean": 26.305195999145507,
      "teacher_distance_mean": 0.85646271109581,
      "valid_action_samples": 215.4,
      "weighted_auxiliary_loss": 0.0054823501966893675,
      "weighted_relation_loss": 0.0054823501966893675
    },
    {
      "action_query_batch_std": 0.39475283920764925,
      "action_query_norm": 43.32142372131348,
      "action_relation_loss": 0.005232586292549968,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.010289990645463,
      "flow_matching_loss": 0.03573279976844788,
      "grad_norm": 0.12739650905132294,
      "h_stock_valid_norm": 98.23633422851563,
      "invalid_tail_samples": 25.0,
      "learning_rate": 7.124407098239155e-05,
      "loss": 0.041,
      "positive_teacher_pairs": 26585.4,
      "query_h_norm_ratio": 0.44099524021148684,
      "relation_loss": 0.005232586292549968,
      "selected_pairs": 26585.4,
      "step": 11770,
      "student_distance_mean": 26.298497581481932,
      "teacher_distance_mean": 0.8616881668567657,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.005232586292549968,
      "weighted_relation_loss": 0.005232586292549968
    },
    {
      "action_query_batch_std": 0.39695592820644376,
      "action_query_norm": 43.32275619506836,
      "action_relation_loss": 0.005063410010188818,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.019644527595885,
      "flow_matching_loss": 0.03722209297120571,
      "grad_norm": 0.14023789763450623,
      "h_stock_valid_norm": 98.45308685302734,
      "invalid_tail_samples": 25.4,
      "learning_rate": 7.119416468825908e-05,
      "loss": 0.0423,
      "positive_teacher_pairs": 26483.4,
      "query_h_norm_ratio": 0.4400365173816681,
      "relation_loss": 0.005063410010188818,
      "selected_pairs": 26483.4,
      "step": 11780,
      "student_distance_mean": 26.3856990814209,
      "teacher_distance_mean": 0.8644467175006867,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.005063410010188818,
      "weighted_relation_loss": 0.005063410010188818
    },
    {
      "action_query_batch_std": 0.39754767119884493,
      "action_query_norm": 43.32714767456055,
      "action_relation_loss": 0.0051406282931566235,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.028999064546305,
      "flow_matching_loss": 0.037131094932556154,
      "grad_norm": 0.09976471960544586,
      "h_stock_valid_norm": 98.07760009765624,
      "invalid_tail_samples": 25.6,
      "learning_rate": 7.114423264121804e-05,
      "loss": 0.0423,
      "positive_teacher_pairs": 26434.3,
      "query_h_norm_ratio": 0.4417726993560791,
      "relation_loss": 0.0051406282931566235,
      "selected_pairs": 26434.3,
      "step": 11790,
      "student_distance_mean": 26.462367057800293,
      "teacher_distance_mean": 0.861061441898346,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0051406282931566235,
      "weighted_relation_loss": 0.0051406282931566235
    },
    {
      "action_query_batch_std": 0.3993500292301178,
      "action_query_norm": 43.342505645751956,
      "action_relation_loss": 0.004953748267143965,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.038353601496725,
      "flow_matching_loss": 0.03752595074474811,
      "grad_norm": 0.11979164928197861,
      "h_stock_valid_norm": 97.34227981567383,
      "invalid_tail_samples": 23.3,
      "learning_rate": 7.109427490194056e-05,
      "loss": 0.0425,
      "positive_teacher_pairs": 26965.4,
      "query_h_norm_ratio": 0.44526101648807526,
      "relation_loss": 0.004953748267143965,
      "selected_pairs": 26965.4,
      "step": 11800,
      "student_distance_mean": 26.558272743225096,
      "teacher_distance_mean": 0.8571049630641937,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.004953748267143965,
      "weighted_relation_loss": 0.004953748267143965
    },
    {
      "action_query_batch_std": 0.3912256181240082,
      "action_query_norm": 43.30975875854492,
      "action_relation_loss": 0.005113888857886195,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.047708138447147,
      "flow_matching_loss": 0.038395054265856744,
      "grad_norm": 0.15267595648765564,
      "h_stock_valid_norm": 97.4833480834961,
      "invalid_tail_samples": 23.1,
      "learning_rate": 7.104429153113001e-05,
      "loss": 0.0435,
      "positive_teacher_pairs": 27017.2,
      "query_h_norm_ratio": 0.4442893356084824,
      "relation_loss": 0.005113888857886195,
      "selected_pairs": 27017.2,
      "step": 11810,
      "student_distance_mean": 26.14146556854248,
      "teacher_distance_mean": 0.8491233348846435,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.005113888857886195,
      "weighted_relation_loss": 0.005113888857886195
    },
    {
      "action_query_batch_std": 0.39039104878902436,
      "action_query_norm": 43.28659896850586,
      "action_relation_loss": 0.00500251823104918,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.057062675397567,
      "flow_matching_loss": 0.03801536709070206,
      "grad_norm": 0.14819154143333435,
      "h_stock_valid_norm": 98.67163848876953,
      "invalid_tail_samples": 25.6,
      "learning_rate": 7.099428258952092e-05,
      "loss": 0.043,
      "positive_teacher_pairs": 26448.5,
      "query_h_norm_ratio": 0.43869668543338775,
      "relation_loss": 0.00500251823104918,
      "selected_pairs": 26448.5,
      "step": 11820,
      "student_distance_mean": 25.881286430358887,
      "teacher_distance_mean": 0.8499236583709717,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.00500251823104918,
      "weighted_relation_loss": 0.00500251823104918
    },
    {
      "action_query_batch_std": 0.395557102560997,
      "action_query_norm": 43.31456298828125,
      "action_relation_loss": 0.005174690065905452,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.06641721234799,
      "flow_matching_loss": 0.03643084652721882,
      "grad_norm": 0.13721388578414917,
      "h_stock_valid_norm": 99.0665885925293,
      "invalid_tail_samples": 21.8,
      "learning_rate": 7.094424813787883e-05,
      "loss": 0.0416,
      "positive_teacher_pairs": 27315.3,
      "query_h_norm_ratio": 0.43723191022872926,
      "relation_loss": 0.005174690065905452,
      "selected_pairs": 27315.3,
      "step": 11830,
      "student_distance_mean": 26.478437423706055,
      "teacher_distance_mean": 0.8595586359500885,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.005174690065905452,
      "weighted_relation_loss": 0.005174690065905452
    },
    {
      "action_query_batch_std": 0.39001242220401766,
      "action_query_norm": 43.28893280029297,
      "action_relation_loss": 0.005115184234455228,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.07577174929841,
      "flow_matching_loss": 0.03665050212293863,
      "grad_norm": 0.14469829201698303,
      "h_stock_valid_norm": 98.13437576293946,
      "invalid_tail_samples": 22.7,
      "learning_rate": 7.089418823700034e-05,
      "loss": 0.0418,
      "positive_teacher_pairs": 27105.6,
      "query_h_norm_ratio": 0.44113147258758545,
      "relation_loss": 0.005115184234455228,
      "selected_pairs": 27105.6,
      "step": 11840,
      "student_distance_mean": 25.911308097839356,
      "teacher_distance_mean": 0.8474994778633118,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.005115184234455228,
      "weighted_relation_loss": 0.005115184234455228
    },
    {
      "action_query_batch_std": 0.3904522806406021,
      "action_query_norm": 43.2980899810791,
      "action_relation_loss": 0.005063606891781092,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.085126286248832,
      "flow_matching_loss": 0.036715658567845824,
      "grad_norm": 0.10989801585674286,
      "h_stock_valid_norm": 97.43192291259766,
      "invalid_tail_samples": 23.9,
      "learning_rate": 7.084410294771298e-05,
      "loss": 0.0418,
      "positive_teacher_pairs": 26826.0,
      "query_h_norm_ratio": 0.4443957805633545,
      "relation_loss": 0.005063606891781092,
      "selected_pairs": 26826.0,
      "step": 11850,
      "student_distance_mean": 25.8668493270874,
      "teacher_distance_mean": 0.8460217475891113,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.005063606891781092,
      "weighted_relation_loss": 0.005063606891781092
    },
    {
      "action_query_batch_std": 0.39820168912410736,
      "action_query_norm": 43.32034606933594,
      "action_relation_loss": 0.0048901497852057215,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.094480823199252,
      "flow_matching_loss": 0.03590697795152664,
      "grad_norm": 0.1292329579591751,
      "h_stock_valid_norm": 97.50313873291016,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.079399233087504e-05,
      "loss": 0.0408,
      "positive_teacher_pairs": 27038.3,
      "query_h_norm_ratio": 0.4443041354417801,
      "relation_loss": 0.0048901497852057215,
      "selected_pairs": 27038.3,
      "step": 11860,
      "student_distance_mean": 26.58576374053955,
      "teacher_distance_mean": 0.8613465607166291,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0048901497852057215,
      "weighted_relation_loss": 0.0048901497852057215
    },
    {
      "action_query_batch_std": 0.3971074253320694,
      "action_query_norm": 43.30570983886719,
      "action_relation_loss": 0.005276500619947911,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.103835360149672,
      "flow_matching_loss": 0.03784380294382572,
      "grad_norm": 0.11187582463026047,
      "h_stock_valid_norm": 97.3828956604004,
      "invalid_tail_samples": 23.3,
      "learning_rate": 7.074385644737568e-05,
      "loss": 0.0431,
      "positive_teacher_pairs": 26963.6,
      "query_h_norm_ratio": 0.4446964621543884,
      "relation_loss": 0.005276500619947911,
      "selected_pairs": 26963.6,
      "step": 11870,
      "student_distance_mean": 26.31156063079834,
      "teacher_distance_mean": 0.8582494497299195,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.005276500619947911,
      "weighted_relation_loss": 0.005276500619947911
    },
    {
      "action_query_batch_std": 0.3916697114706039,
      "action_query_norm": 43.287713623046876,
      "action_relation_loss": 0.004953307844698429,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.113189897100094,
      "flow_matching_loss": 0.03605248816311359,
      "grad_norm": 0.11928297579288483,
      "h_stock_valid_norm": 97.84237442016601,
      "invalid_tail_samples": 20.0,
      "learning_rate": 7.069369535813473e-05,
      "loss": 0.041,
      "positive_teacher_pairs": 27737.2,
      "query_h_norm_ratio": 0.44243095219135287,
      "relation_loss": 0.004953307844698429,
      "selected_pairs": 27737.2,
      "step": 11880,
      "student_distance_mean": 26.02848014831543,
      "teacher_distance_mean": 0.8514590561389923,
      "valid_action_samples": 236.0,
      "weighted_auxiliary_loss": 0.004953307844698429,
      "weighted_relation_loss": 0.004953307844698429
    },
    {
      "action_query_batch_std": 0.3947626709938049,
      "action_query_norm": 43.30787582397461,
      "action_relation_loss": 0.004895878070965409,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.122544434050514,
      "flow_matching_loss": 0.0367103947326541,
      "grad_norm": 0.09790302813053131,
      "h_stock_valid_norm": 97.95752487182617,
      "invalid_tail_samples": 24.8,
      "learning_rate": 7.06435091241026e-05,
      "loss": 0.0416,
      "positive_teacher_pairs": 26622.0,
      "query_h_norm_ratio": 0.4421109020709991,
      "relation_loss": 0.004895878070965409,
      "selected_pairs": 26622.0,
      "step": 11890,
      "student_distance_mean": 26.36038475036621,
      "teacher_distance_mean": 0.8585293114185333,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.004895878070965409,
      "weighted_relation_loss": 0.004895878070965409
    },
    {
      "action_query_batch_std": 0.3955798089504242,
      "action_query_norm": 43.28981094360351,
      "action_relation_loss": 0.00516556496731937,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.131898971000936,
      "flow_matching_loss": 0.03837300725281238,
      "grad_norm": 0.11459191143512726,
      "h_stock_valid_norm": 97.49153594970703,
      "invalid_tail_samples": 25.6,
      "learning_rate": 7.059329780626034e-05,
      "loss": 0.0435,
      "positive_teacher_pairs": 26446.0,
      "query_h_norm_ratio": 0.44404386878013613,
      "relation_loss": 0.00516556496731937,
      "selected_pairs": 26446.0,
      "step": 11900,
      "student_distance_mean": 26.319147300720214,
      "teacher_distance_mean": 0.8582212150096893,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.00516556496731937,
      "weighted_relation_loss": 0.00516556496731937
    },
    {
      "action_query_batch_std": 0.3970104902982712,
      "action_query_norm": 43.30522117614746,
      "action_relation_loss": 0.004992022551596165,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.141253507951356,
      "flow_matching_loss": 0.037521634623408316,
      "grad_norm": 0.11887358874082565,
      "h_stock_valid_norm": 97.91681747436523,
      "invalid_tail_samples": 26.6,
      "learning_rate": 7.054306146561944e-05,
      "loss": 0.0425,
      "positive_teacher_pairs": 26214.8,
      "query_h_norm_ratio": 0.4422706991434097,
      "relation_loss": 0.004992022551596165,
      "selected_pairs": 26214.8,
      "step": 11910,
      "student_distance_mean": 26.556788063049318,
      "teacher_distance_mean": 0.8596799552440644,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.004992022551596165,
      "weighted_relation_loss": 0.004992022551596165
    },
    {
      "action_query_batch_std": 0.40027720630168917,
      "action_query_norm": 43.31602439880371,
      "action_relation_loss": 0.004947463888674975,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.150608044901777,
      "flow_matching_loss": 0.03791380636394024,
      "grad_norm": 0.1204686164855957,
      "h_stock_valid_norm": 98.65142669677735,
      "invalid_tail_samples": 24.2,
      "learning_rate": 7.049280016322178e-05,
      "loss": 0.0429,
      "positive_teacher_pairs": 26755.9,
      "query_h_norm_ratio": 0.43908547461032865,
      "relation_loss": 0.004947463888674975,
      "selected_pairs": 26755.9,
      "step": 11920,
      "student_distance_mean": 26.54510440826416,
      "teacher_distance_mean": 0.864751011133194,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.004947463888674975,
      "weighted_relation_loss": 0.004947463888674975
    },
    {
      "action_query_batch_std": 0.3938003957271576,
      "action_query_norm": 43.29311943054199,
      "action_relation_loss": 0.0048976581078022715,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.159962581852199,
      "flow_matching_loss": 0.03498994559049606,
      "grad_norm": 0.10953320562839508,
      "h_stock_valid_norm": 98.46726837158204,
      "invalid_tail_samples": 23.1,
      "learning_rate": 7.044251396013957e-05,
      "loss": 0.0399,
      "positive_teacher_pairs": 27012.8,
      "query_h_norm_ratio": 0.4396780848503113,
      "relation_loss": 0.0048976581078022715,
      "selected_pairs": 27012.8,
      "step": 11930,
      "student_distance_mean": 26.308676719665527,
      "teacher_distance_mean": 0.8578653573989868,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0048976581078022715,
      "weighted_relation_loss": 0.0048976581078022715
    },
    {
      "action_query_batch_std": 0.39596163630485537,
      "action_query_norm": 43.29406814575195,
      "action_relation_loss": 0.005090203974395991,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.169317118802619,
      "flow_matching_loss": 0.03588472083210945,
      "grad_norm": 0.12451260536909103,
      "h_stock_valid_norm": 98.37560729980468,
      "invalid_tail_samples": 24.0,
      "learning_rate": 7.039220291747528e-05,
      "loss": 0.041,
      "positive_teacher_pairs": 26807.0,
      "query_h_norm_ratio": 0.4400927066802979,
      "relation_loss": 0.005090203974395991,
      "selected_pairs": 26807.0,
      "step": 11940,
      "student_distance_mean": 26.300724220275878,
      "teacher_distance_mean": 0.8619182348251343,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.005090203974395991,
      "weighted_relation_loss": 0.005090203974395991
    },
    {
      "action_query_batch_std": 0.3912541300058365,
      "action_query_norm": 43.27351379394531,
      "action_relation_loss": 0.004998782090842724,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.17867165575304,
      "flow_matching_loss": 0.03624804839491844,
      "grad_norm": 0.09845209866762161,
      "h_stock_valid_norm": 98.37691497802734,
      "invalid_tail_samples": 26.9,
      "learning_rate": 7.034186709636159e-05,
      "loss": 0.0412,
      "positive_teacher_pairs": 26144.8,
      "query_h_norm_ratio": 0.4398778140544891,
      "relation_loss": 0.004998782090842724,
      "selected_pairs": 26144.8,
      "step": 11950,
      "student_distance_mean": 26.20755252838135,
      "teacher_distance_mean": 0.8542828261852264,
      "valid_action_samples": 229.1,
      "weighted_auxiliary_loss": 0.004998782090842724,
      "weighted_relation_loss": 0.004998782090842724
    },
    {
      "action_query_batch_std": 0.3903429746627808,
      "action_query_norm": 43.260843658447264,
      "action_relation_loss": 0.0051628212910145525,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.188026192703461,
      "flow_matching_loss": 0.03731740601360798,
      "grad_norm": 0.1054418683052063,
      "h_stock_valid_norm": 98.25718536376954,
      "invalid_tail_samples": 24.1,
      "learning_rate": 7.029150655796129e-05,
      "loss": 0.0425,
      "positive_teacher_pairs": 26782.3,
      "query_h_norm_ratio": 0.4402838796377182,
      "relation_loss": 0.0051628212910145525,
      "selected_pairs": 26782.3,
      "step": 11960,
      "student_distance_mean": 26.11296501159668,
      "teacher_distance_mean": 0.8489801406860351,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0051628212910145525,
      "weighted_relation_loss": 0.0051628212910145525
    },
    {
      "action_query_batch_std": 0.3958775222301483,
      "action_query_norm": 43.287784576416016,
      "action_relation_loss": 0.0049330254551023245,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.197380729653883,
      "flow_matching_loss": 0.035624073818325996,
      "grad_norm": 0.13854897022247314,
      "h_stock_valid_norm": 98.2381576538086,
      "invalid_tail_samples": 24.6,
      "learning_rate": 7.024112136346712e-05,
      "loss": 0.0406,
      "positive_teacher_pairs": 26667.8,
      "query_h_norm_ratio": 0.4406464844942093,
      "relation_loss": 0.0049330254551023245,
      "selected_pairs": 26667.8,
      "step": 11970,
      "student_distance_mean": 26.385720825195314,
      "teacher_distance_mean": 0.8655943691730499,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0049330254551023245,
      "weighted_relation_loss": 0.0049330254551023245
    },
    {
      "action_query_batch_std": 0.39268221855163576,
      "action_query_norm": 43.275396728515624,
      "action_relation_loss": 0.005266670370474458,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.206735266604303,
      "flow_matching_loss": 0.038199570029973984,
      "grad_norm": 0.13102993369102478,
      "h_stock_valid_norm": 97.6479248046875,
      "invalid_tail_samples": 23.3,
      "learning_rate": 7.01907115741019e-05,
      "loss": 0.0435,
      "positive_teacher_pairs": 26966.5,
      "query_h_norm_ratio": 0.4431808054447174,
      "relation_loss": 0.005266670370474458,
      "selected_pairs": 26966.5,
      "step": 11980,
      "student_distance_mean": 26.121292495727538,
      "teacher_distance_mean": 0.856056410074234,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.005266670370474458,
      "weighted_relation_loss": 0.005266670370474458
    },
    {
      "action_query_batch_std": 0.39074126780033114,
      "action_query_norm": 43.27534408569336,
      "action_relation_loss": 0.005225788475945592,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.216089803554723,
      "flow_matching_loss": 0.03830321803689003,
      "grad_norm": 0.122207872569561,
      "h_stock_valid_norm": 97.92716979980469,
      "invalid_tail_samples": 24.5,
      "learning_rate": 7.014027725111825e-05,
      "loss": 0.0435,
      "positive_teacher_pairs": 26689.1,
      "query_h_norm_ratio": 0.4419177144765854,
      "relation_loss": 0.005225788475945592,
      "selected_pairs": 26689.1,
      "step": 11990,
      "student_distance_mean": 25.95803871154785,
      "teacher_distance_mean": 0.8444814503192901,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.005225788475945592,
      "weighted_relation_loss": 0.005225788475945592
    },
    {
      "action_query_batch_std": 0.39495979249477386,
      "action_query_norm": 43.266770935058595,
      "action_relation_loss": 0.0050074461381882426,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.225444340505145,
      "flow_matching_loss": 0.037850698456168175,
      "grad_norm": 0.13373208045959473,
      "h_stock_valid_norm": 98.31688842773437,
      "invalid_tail_samples": 24.8,
      "learning_rate": 7.008981845579864e-05,
      "loss": 0.0429,
      "positive_teacher_pairs": 26622.4,
      "query_h_norm_ratio": 0.4400783538818359,
      "relation_loss": 0.0050074461381882426,
      "selected_pairs": 26622.4,
      "step": 12000,
      "student_distance_mean": 26.33807506561279,
      "teacher_distance_mean": 0.8568101048469543,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0050074461381882426,
      "weighted_relation_loss": 0.0050074461381882426
    },
    {
      "action_query_batch_std": 0.39201649129390714,
      "action_query_norm": 43.26553497314453,
      "action_relation_loss": 0.00501189767383039,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.234798877455566,
      "flow_matching_loss": 0.03495824187994003,
      "grad_norm": 0.13734543323516846,
      "h_stock_valid_norm": 98.56323623657227,
      "invalid_tail_samples": 24.2,
      "learning_rate": 7.003933524945528e-05,
      "loss": 0.04,
      "positive_teacher_pairs": 26753.4,
      "query_h_norm_ratio": 0.43896839320659636,
      "relation_loss": 0.00501189767383039,
      "selected_pairs": 26753.4,
      "step": 12010,
      "student_distance_mean": 26.121831130981445,
      "teacher_distance_mean": 0.849613219499588,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.00501189767383039,
      "weighted_relation_loss": 0.00501189767383039
    },
    {
      "action_query_batch_std": 0.4019458472728729,
      "action_query_norm": 43.29221458435059,
      "action_relation_loss": 0.005141291441395879,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.244153414405988,
      "flow_matching_loss": 0.03556951731443405,
      "grad_norm": 0.12249351292848587,
      "h_stock_valid_norm": 98.53129119873047,
      "invalid_tail_samples": 25.9,
      "learning_rate": 6.998882769342998e-05,
      "loss": 0.0407,
      "positive_teacher_pairs": 26365.0,
      "query_h_norm_ratio": 0.4393802464008331,
      "relation_loss": 0.005141291441395879,
      "selected_pairs": 26365.0,
      "step": 12020,
      "student_distance_mean": 26.832532119750976,
      "teacher_distance_mean": 0.8713494837284088,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.005141291441395879,
      "weighted_relation_loss": 0.005141291441395879
    },
    {
      "action_query_batch_std": 0.3926795691251755,
      "action_query_norm": 43.27454833984375,
      "action_relation_loss": 0.005128890089690685,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.253507951356408,
      "flow_matching_loss": 0.03515053391456604,
      "grad_norm": 0.10405939817428589,
      "h_stock_valid_norm": 97.78085098266601,
      "invalid_tail_samples": 21.8,
      "learning_rate": 6.993829584909423e-05,
      "loss": 0.0403,
      "positive_teacher_pairs": 27320.1,
      "query_h_norm_ratio": 0.44257002472877505,
      "relation_loss": 0.005128890089690685,
      "selected_pairs": 27320.1,
      "step": 12030,
      "student_distance_mean": 26.2465087890625,
      "teacher_distance_mean": 0.851344245672226,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.005128890089690685,
      "weighted_relation_loss": 0.005128890089690685
    },
    {
      "action_query_batch_std": 0.39108688235282896,
      "action_query_norm": 43.25433692932129,
      "action_relation_loss": 0.0052017497830092905,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.262862488306828,
      "flow_matching_loss": 0.036447730660438535,
      "grad_norm": 0.08882968872785568,
      "h_stock_valid_norm": 97.78771133422852,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.988773977784895e-05,
      "loss": 0.0416,
      "positive_teacher_pairs": 26660.9,
      "query_h_norm_ratio": 0.4423313826322556,
      "relation_loss": 0.0052017497830092905,
      "selected_pairs": 26660.9,
      "step": 12040,
      "student_distance_mean": 26.119876289367674,
      "teacher_distance_mean": 0.8502229750156403,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0052017497830092905,
      "weighted_relation_loss": 0.0052017497830092905
    },
    {
      "action_query_batch_std": 0.39609489738941195,
      "action_query_norm": 43.26284904479981,
      "action_relation_loss": 0.005053097847849131,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.27221702525725,
      "flow_matching_loss": 0.03752260766923428,
      "grad_norm": 0.10238988697528839,
      "h_stock_valid_norm": 98.3206687927246,
      "invalid_tail_samples": 23.7,
      "learning_rate": 6.983715954112454e-05,
      "loss": 0.0426,
      "positive_teacher_pairs": 26876.1,
      "query_h_norm_ratio": 0.44002423286437986,
      "relation_loss": 0.005053097847849131,
      "selected_pairs": 26876.1,
      "step": 12050,
      "student_distance_mean": 26.3425350189209,
      "teacher_distance_mean": 0.856883031129837,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.005053097847849131,
      "weighted_relation_loss": 0.005053097847849131
    },
    {
      "action_query_batch_std": 0.39391595125198364,
      "action_query_norm": 43.25994529724121,
      "action_relation_loss": 0.004868713533505797,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.28157156220767,
      "flow_matching_loss": 0.034682847931981085,
      "grad_norm": 0.13884924352169037,
      "h_stock_valid_norm": 98.01582489013671,
      "invalid_tail_samples": 22.9,
      "learning_rate": 6.978655520038079e-05,
      "loss": 0.0396,
      "positive_teacher_pairs": 27069.4,
      "query_h_norm_ratio": 0.4413621872663498,
      "relation_loss": 0.004868713533505797,
      "selected_pairs": 27069.4,
      "step": 12060,
      "student_distance_mean": 26.175421905517577,
      "teacher_distance_mean": 0.8537349283695221,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.004868713533505797,
      "weighted_relation_loss": 0.004868713533505797
    },
    {
      "action_query_batch_std": 0.39154103994369505,
      "action_query_norm": 43.25664291381836,
      "action_relation_loss": 0.005052685225382447,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.290926099158092,
      "flow_matching_loss": 0.03453033175319433,
      "grad_norm": 0.11279004067182541,
      "h_stock_valid_norm": 98.64677810668945,
      "invalid_tail_samples": 22.8,
      "learning_rate": 6.97359268171067e-05,
      "loss": 0.0396,
      "positive_teacher_pairs": 27086.9,
      "query_h_norm_ratio": 0.43850276470184324,
      "relation_loss": 0.005052685225382447,
      "selected_pairs": 27086.9,
      "step": 12070,
      "student_distance_mean": 26.08594379425049,
      "teacher_distance_mean": 0.8534065186977386,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.005052685225382447,
      "weighted_relation_loss": 0.005052685225382447
    },
    {
      "action_query_batch_std": 0.3928090810775757,
      "action_query_norm": 43.24965057373047,
      "action_relation_loss": 0.004967618174850941,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.300280636108512,
      "flow_matching_loss": 0.03723468780517578,
      "grad_norm": 0.12096048146486282,
      "h_stock_valid_norm": 99.41700820922851,
      "invalid_tail_samples": 23.7,
      "learning_rate": 6.968527445282056e-05,
      "loss": 0.0422,
      "positive_teacher_pairs": 26880.7,
      "query_h_norm_ratio": 0.43503697514533995,
      "relation_loss": 0.004967618174850941,
      "selected_pairs": 26880.7,
      "step": 12080,
      "student_distance_mean": 26.171921157836913,
      "teacher_distance_mean": 0.8522169291973114,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.004967618174850941,
      "weighted_relation_loss": 0.004967618174850941
    },
    {
      "action_query_batch_std": 0.3995010942220688,
      "action_query_norm": 43.286637496948245,
      "action_relation_loss": 0.004908141680061817,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.309635173058934,
      "flow_matching_loss": 0.037307660467922686,
      "grad_norm": 0.1419667750597,
      "h_stock_valid_norm": 99.6223045349121,
      "invalid_tail_samples": 21.8,
      "learning_rate": 6.963459816906977e-05,
      "loss": 0.0422,
      "positive_teacher_pairs": 27314.1,
      "query_h_norm_ratio": 0.4345096230506897,
      "relation_loss": 0.004908141680061817,
      "selected_pairs": 27314.1,
      "step": 12090,
      "student_distance_mean": 26.621150779724122,
      "teacher_distance_mean": 0.8632580935955048,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.004908141680061817,
      "weighted_relation_loss": 0.004908141680061817
    },
    {
      "action_query_batch_std": 0.39796222746372223,
      "action_query_norm": 43.28226737976074,
      "action_relation_loss": 0.005165234860032797,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.318989710009355,
      "flow_matching_loss": 0.03598622437566519,
      "grad_norm": 0.12335299700498581,
      "h_stock_valid_norm": 99.21156005859375,
      "invalid_tail_samples": 22.1,
      "learning_rate": 6.958389802743077e-05,
      "loss": 0.0412,
      "positive_teacher_pairs": 27246.4,
      "query_h_norm_ratio": 0.4362644344568253,
      "relation_loss": 0.005165234860032797,
      "selected_pairs": 27246.4,
      "step": 12100,
      "student_distance_mean": 26.484374046325684,
      "teacher_distance_mean": 0.8585791885852814,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.005165234860032797,
      "weighted_relation_loss": 0.005165234860032797
    },
    {
      "action_query_batch_std": 0.3977943956851959,
      "action_query_norm": 43.26085929870605,
      "action_relation_loss": 0.005048148613423109,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.328344246959775,
      "flow_matching_loss": 0.03554386477917433,
      "grad_norm": 0.1387982815504074,
      "h_stock_valid_norm": 98.68876647949219,
      "invalid_tail_samples": 22.4,
      "learning_rate": 6.953317408950903e-05,
      "loss": 0.0406,
      "positive_teacher_pairs": 27183.5,
      "query_h_norm_ratio": 0.4383591920137405,
      "relation_loss": 0.005048148613423109,
      "selected_pairs": 27183.5,
      "step": 12110,
      "student_distance_mean": 26.460780525207518,
      "teacher_distance_mean": 0.8591339409351348,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.005048148613423109,
      "weighted_relation_loss": 0.005048148613423109
    },
    {
      "action_query_batch_std": 0.3964278489351273,
      "action_query_norm": 43.26216125488281,
      "action_relation_loss": 0.004826748883351684,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.337698783910197,
      "flow_matching_loss": 0.035959886573255065,
      "grad_norm": 0.1028980016708374,
      "h_stock_valid_norm": 98.49436416625977,
      "invalid_tail_samples": 24.7,
      "learning_rate": 6.94824264169389e-05,
      "loss": 0.0408,
      "positive_teacher_pairs": 26641.9,
      "query_h_norm_ratio": 0.4392379641532898,
      "relation_loss": 0.004826748883351684,
      "selected_pairs": 26641.9,
      "step": 12120,
      "student_distance_mean": 26.47251377105713,
      "teacher_distance_mean": 0.8639210283756256,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.004826748883351684,
      "weighted_relation_loss": 0.004826748883351684
    },
    {
      "action_query_batch_std": 0.3902143746614456,
      "action_query_norm": 43.23656158447265,
      "action_relation_loss": 0.00513441557995975,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.347053320860617,
      "flow_matching_loss": 0.036689494550228116,
      "grad_norm": 0.11962411552667618,
      "h_stock_valid_norm": 98.52844848632813,
      "invalid_tail_samples": 23.1,
      "learning_rate": 6.94316550713836e-05,
      "loss": 0.0418,
      "positive_teacher_pairs": 27012.4,
      "query_h_norm_ratio": 0.43882502019405367,
      "relation_loss": 0.00513441557995975,
      "selected_pairs": 27012.4,
      "step": 12130,
      "student_distance_mean": 25.957309913635253,
      "teacher_distance_mean": 0.850759357213974,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.00513441557995975,
      "weighted_relation_loss": 0.00513441557995975
    },
    {
      "action_query_batch_std": 0.3885468155145645,
      "action_query_norm": 43.2287353515625,
      "action_relation_loss": 0.00478592119179666,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.356407857811039,
      "flow_matching_loss": 0.03766095563769341,
      "grad_norm": 0.11844255775213242,
      "h_stock_valid_norm": 98.38014297485351,
      "invalid_tail_samples": 29.0,
      "learning_rate": 6.938086011453513e-05,
      "loss": 0.0424,
      "positive_teacher_pairs": 25660.6,
      "query_h_norm_ratio": 0.4394106298685074,
      "relation_loss": 0.00478592119179666,
      "selected_pairs": 25660.6,
      "step": 12140,
      "student_distance_mean": 25.91597957611084,
      "teacher_distance_mean": 0.8510256111621857,
      "valid_action_samples": 227.0,
      "weighted_auxiliary_loss": 0.00478592119179666,
      "weighted_relation_loss": 0.00478592119179666
    },
    {
      "action_query_batch_std": 0.39566839337348936,
      "action_query_norm": 43.25392646789551,
      "action_relation_loss": 0.004823703272268176,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.36576239476146,
      "flow_matching_loss": 0.03699925988912582,
      "grad_norm": 0.13293109834194183,
      "h_stock_valid_norm": 97.45060195922852,
      "invalid_tail_samples": 23.4,
      "learning_rate": 6.93300416081141e-05,
      "loss": 0.0418,
      "positive_teacher_pairs": 26953.9,
      "query_h_norm_ratio": 0.44385949373245237,
      "relation_loss": 0.004823703272268176,
      "selected_pairs": 26953.9,
      "step": 12150,
      "student_distance_mean": 26.41044502258301,
      "teacher_distance_mean": 0.8636250019073486,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.004823703272268176,
      "weighted_relation_loss": 0.004823703272268176
    },
    {
      "action_query_batch_std": 0.39356083869934083,
      "action_query_norm": 43.24648590087891,
      "action_relation_loss": 0.005089247692376375,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.37511693171188,
      "flow_matching_loss": 0.037284331396222115,
      "grad_norm": 0.117668017745018,
      "h_stock_valid_norm": 97.40893096923828,
      "invalid_tail_samples": 26.3,
      "learning_rate": 6.927919961386984e-05,
      "loss": 0.0424,
      "positive_teacher_pairs": 26287.6,
      "query_h_norm_ratio": 0.44397581815719606,
      "relation_loss": 0.005089247692376375,
      "selected_pairs": 26287.6,
      "step": 12160,
      "student_distance_mean": 26.152441215515136,
      "teacher_distance_mean": 0.8609374344348908,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.005089247692376375,
      "weighted_relation_loss": 0.005089247692376375
    },
    {
      "action_query_batch_std": 0.3863066405057907,
      "action_query_norm": 43.21664047241211,
      "action_relation_loss": 0.004865253809839487,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.384471468662301,
      "flow_matching_loss": 0.03766367733478546,
      "grad_norm": 0.11265858262777328,
      "h_stock_valid_norm": 98.409912109375,
      "invalid_tail_samples": 23.5,
      "learning_rate": 6.922833419358013e-05,
      "loss": 0.0425,
      "positive_teacher_pairs": 26914.2,
      "query_h_norm_ratio": 0.43915637731552126,
      "relation_loss": 0.004865253809839487,
      "selected_pairs": 26914.2,
      "step": 12170,
      "student_distance_mean": 25.571881675720213,
      "teacher_distance_mean": 0.8442794978618622,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.004865253809839487,
      "weighted_relation_loss": 0.004865253809839487
    },
    {
      "action_query_batch_std": 0.388028210401535,
      "action_query_norm": 43.22294158935547,
      "action_relation_loss": 0.004820586834102869,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.393826005612722,
      "flow_matching_loss": 0.03404917679727078,
      "grad_norm": 0.11180293560028076,
      "h_stock_valid_norm": 98.66128692626953,
      "invalid_tail_samples": 24.1,
      "learning_rate": 6.917744540905124e-05,
      "loss": 0.0389,
      "positive_teacher_pairs": 26782.5,
      "query_h_norm_ratio": 0.43809674978256224,
      "relation_loss": 0.004820586834102869,
      "selected_pairs": 26782.5,
      "step": 12180,
      "student_distance_mean": 25.865150451660156,
      "teacher_distance_mean": 0.8494931817054748,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.004820586834102869,
      "weighted_relation_loss": 0.004820586834102869
    },
    {
      "action_query_batch_std": 0.3967994451522827,
      "action_query_norm": 43.23941154479981,
      "action_relation_loss": 0.004709007358178497,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.403180542563144,
      "flow_matching_loss": 0.03727144598960876,
      "grad_norm": 0.10183194279670715,
      "h_stock_valid_norm": 98.66813125610352,
      "invalid_tail_samples": 23.9,
      "learning_rate": 6.912653332211787e-05,
      "loss": 0.042,
      "positive_teacher_pairs": 26827.0,
      "query_h_norm_ratio": 0.4382350414991379,
      "relation_loss": 0.004709007358178497,
      "selected_pairs": 26827.0,
      "step": 12190,
      "student_distance_mean": 26.39738883972168,
      "teacher_distance_mean": 0.8645116329193115,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.004709007358178497,
      "weighted_relation_loss": 0.004709007358178497
    },
    {
      "action_query_batch_std": 0.3946790397167206,
      "action_query_norm": 43.23581428527832,
      "action_relation_loss": 0.004747529048472643,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.412535079513564,
      "flow_matching_loss": 0.038000718131661414,
      "grad_norm": 0.15511275827884674,
      "h_stock_valid_norm": 98.0624382019043,
      "invalid_tail_samples": 20.2,
      "learning_rate": 6.9075597994643e-05,
      "loss": 0.0427,
      "positive_teacher_pairs": 27693.7,
      "query_h_norm_ratio": 0.44090413451194765,
      "relation_loss": 0.004747529048472643,
      "selected_pairs": 27693.7,
      "step": 12200,
      "student_distance_mean": 26.323147583007813,
      "teacher_distance_mean": 0.8612628638744354,
      "valid_action_samples": 235.8,
      "weighted_auxiliary_loss": 0.004747529048472643,
      "weighted_relation_loss": 0.004747529048472643
    },
    {
      "action_query_batch_std": 0.3913245052099228,
      "action_query_norm": 43.217190170288085,
      "action_relation_loss": 0.005004731379449368,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.421889616463986,
      "flow_matching_loss": 0.03847192190587521,
      "grad_norm": 0.11314640939235687,
      "h_stock_valid_norm": 97.95198211669921,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.902463948851787e-05,
      "loss": 0.0435,
      "positive_teacher_pairs": 26669.6,
      "query_h_norm_ratio": 0.44121178388595583,
      "relation_loss": 0.005004731379449368,
      "selected_pairs": 26669.6,
      "step": 12210,
      "student_distance_mean": 26.02598171234131,
      "teacher_distance_mean": 0.8561084032058716,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.005004731379449368,
      "weighted_relation_loss": 0.005004731379449368
    },
    {
      "action_query_batch_std": 0.3886971205472946,
      "action_query_norm": 43.221825790405276,
      "action_relation_loss": 0.004766782512888312,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.431244153414406,
      "flow_matching_loss": 0.03489334937185049,
      "grad_norm": 0.09353353083133698,
      "h_stock_valid_norm": 98.186181640625,
      "invalid_tail_samples": 24.1,
      "learning_rate": 6.897365786566184e-05,
      "loss": 0.0397,
      "positive_teacher_pairs": 26777.7,
      "query_h_norm_ratio": 0.44020640552043916,
      "relation_loss": 0.004766782512888312,
      "selected_pairs": 26777.7,
      "step": 12220,
      "student_distance_mean": 25.88567409515381,
      "teacher_distance_mean": 0.8530638813972473,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.004766782512888312,
      "weighted_relation_loss": 0.004766782512888312
    },
    {
      "action_query_batch_std": 0.390503454208374,
      "action_query_norm": 43.219702911376956,
      "action_relation_loss": 0.004803014406934381,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.440598690364826,
      "flow_matching_loss": 0.03459474295377731,
      "grad_norm": 0.11325564235448837,
      "h_stock_valid_norm": 97.2181999206543,
      "invalid_tail_samples": 22.6,
      "learning_rate": 6.892265318802242e-05,
      "loss": 0.0394,
      "positive_teacher_pairs": 27125.6,
      "query_h_norm_ratio": 0.444565212726593,
      "relation_loss": 0.004803014406934381,
      "selected_pairs": 27125.6,
      "step": 12230,
      "student_distance_mean": 26.02288684844971,
      "teacher_distance_mean": 0.858221036195755,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.004803014406934381,
      "weighted_relation_loss": 0.004803014406934381
    },
    {
      "action_query_batch_std": 0.3832142263650894,
      "action_query_norm": 43.19272575378418,
      "action_relation_loss": 0.005078418273478746,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.449953227315248,
      "flow_matching_loss": 0.034655149281024936,
      "grad_norm": 0.09376522898674011,
      "h_stock_valid_norm": 98.15504302978516,
      "invalid_tail_samples": 26.6,
      "learning_rate": 6.887162551757507e-05,
      "loss": 0.0397,
      "positive_teacher_pairs": 26224.0,
      "query_h_norm_ratio": 0.4400550574064255,
      "relation_loss": 0.005078418273478746,
      "selected_pairs": 26224.0,
      "step": 12240,
      "student_distance_mean": 25.485936546325682,
      "teacher_distance_mean": 0.8462480425834655,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.005078418273478746,
      "weighted_relation_loss": 0.005078418273478746
    },
    {
      "action_query_batch_std": 0.3868696212768555,
      "action_query_norm": 43.210737228393555,
      "action_relation_loss": 0.004969890275970101,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.459307764265668,
      "flow_matching_loss": 0.03407703097909689,
      "grad_norm": 0.08744029700756073,
      "h_stock_valid_norm": 98.91918106079102,
      "invalid_tail_samples": 24.4,
      "learning_rate": 6.882057491632326e-05,
      "loss": 0.039,
      "positive_teacher_pairs": 26712.1,
      "query_h_norm_ratio": 0.4368308842182159,
      "relation_loss": 0.004969890275970101,
      "selected_pairs": 26712.1,
      "step": 12250,
      "student_distance_mean": 25.83360481262207,
      "teacher_distance_mean": 0.8552780032157898,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.004969890275970101,
      "weighted_relation_loss": 0.004969890275970101
    },
    {
      "action_query_batch_std": 0.3889527887105942,
      "action_query_norm": 43.2156421661377,
      "action_relation_loss": 0.0049963119905442,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.46866230121609,
      "flow_matching_loss": 0.037296558544039723,
      "grad_norm": 0.10713019967079163,
      "h_stock_valid_norm": 98.86332626342774,
      "invalid_tail_samples": 22.6,
      "learning_rate": 6.876950144629824e-05,
      "loss": 0.0423,
      "positive_teacher_pairs": 27131.5,
      "query_h_norm_ratio": 0.4371315032243729,
      "relation_loss": 0.0049963119905442,
      "selected_pairs": 27131.5,
      "step": 12260,
      "student_distance_mean": 25.929247093200683,
      "teacher_distance_mean": 0.8607246398925781,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.0049963119905442,
      "weighted_relation_loss": 0.0049963119905442
    },
    {
      "action_query_batch_std": 0.38730633854866026,
      "action_query_norm": 43.1997127532959,
      "action_relation_loss": 0.004748828988522291,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.47801683816651,
      "flow_matching_loss": 0.03636624477803707,
      "grad_norm": 0.12845376133918762,
      "h_stock_valid_norm": 97.93523864746093,
      "invalid_tail_samples": 21.7,
      "learning_rate": 6.87184051695591e-05,
      "loss": 0.0411,
      "positive_teacher_pairs": 27336.5,
      "query_h_norm_ratio": 0.44111362397670745,
      "relation_loss": 0.004748828988522291,
      "selected_pairs": 27336.5,
      "step": 12270,
      "student_distance_mean": 25.76647491455078,
      "teacher_distance_mean": 0.8535094738006592,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.004748828988522291,
      "weighted_relation_loss": 0.004748828988522291
    },
    {
      "action_query_batch_std": 0.3874271124601364,
      "action_query_norm": 43.19747428894043,
      "action_relation_loss": 0.0050609071739017965,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.487371375116933,
      "flow_matching_loss": 0.03662220984697342,
      "grad_norm": 0.10149404406547546,
      "h_stock_valid_norm": 97.75097885131837,
      "invalid_tail_samples": 23.6,
      "learning_rate": 6.866728614819268e-05,
      "loss": 0.0417,
      "positive_teacher_pairs": 26895.2,
      "query_h_norm_ratio": 0.44191745221614837,
      "relation_loss": 0.0050609071739017965,
      "selected_pairs": 26895.2,
      "step": 12280,
      "student_distance_mean": 25.888685035705567,
      "teacher_distance_mean": 0.8515496909618377,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0050609071739017965,
      "weighted_relation_loss": 0.0050609071739017965
    },
    {
      "action_query_batch_std": 0.3874475181102753,
      "action_query_norm": 43.1928768157959,
      "action_relation_loss": 0.004902410274371505,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.496725912067353,
      "flow_matching_loss": 0.0363305764272809,
      "grad_norm": 0.1424245983362198,
      "h_stock_valid_norm": 98.24378204345703,
      "invalid_tail_samples": 22.7,
      "learning_rate": 6.861614444431337e-05,
      "loss": 0.0412,
      "positive_teacher_pairs": 27101.7,
      "query_h_norm_ratio": 0.43965342938899993,
      "relation_loss": 0.004902410274371505,
      "selected_pairs": 27101.7,
      "step": 12290,
      "student_distance_mean": 25.948368644714357,
      "teacher_distance_mean": 0.8566925942897796,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.004902410274371505,
      "weighted_relation_loss": 0.004902410274371505
    },
    {
      "action_query_batch_std": 0.3908392310142517,
      "action_query_norm": 43.2146800994873,
      "action_relation_loss": 0.004845360480248928,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.506080449017773,
      "flow_matching_loss": 0.03545353189110756,
      "grad_norm": 0.13942256569862366,
      "h_stock_valid_norm": 98.5016357421875,
      "invalid_tail_samples": 22.5,
      "learning_rate": 6.856498012006318e-05,
      "loss": 0.0403,
      "positive_teacher_pairs": 27151.1,
      "query_h_norm_ratio": 0.43872477412223815,
      "relation_loss": 0.004845360480248928,
      "selected_pairs": 27151.1,
      "step": 12300,
      "student_distance_mean": 26.047673988342286,
      "teacher_distance_mean": 0.8550793647766113,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.004845360480248928,
      "weighted_relation_loss": 0.004845360480248928
    },
    {
      "action_query_batch_std": 0.3880732834339142,
      "action_query_norm": 43.19618721008301,
      "action_relation_loss": 0.004874918097630143,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.515434985968195,
      "flow_matching_loss": 0.03662467747926712,
      "grad_norm": 0.13667376339435577,
      "h_stock_valid_norm": 97.78208618164062,
      "invalid_tail_samples": 21.9,
      "learning_rate": 6.851379323761157e-05,
      "loss": 0.0415,
      "positive_teacher_pairs": 27295.4,
      "query_h_norm_ratio": 0.44176715314388276,
      "relation_loss": 0.004874918097630143,
      "selected_pairs": 27295.4,
      "step": 12310,
      "student_distance_mean": 25.747113037109376,
      "teacher_distance_mean": 0.8521512746810913,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.004874918097630143,
      "weighted_relation_loss": 0.004874918097630143
    },
    {
      "action_query_batch_std": 0.38737650215625763,
      "action_query_norm": 43.192101669311526,
      "action_relation_loss": 0.004982928326353431,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.524789522918615,
      "flow_matching_loss": 0.03529445193707943,
      "grad_norm": 0.1353573501110077,
      "h_stock_valid_norm": 97.94796905517578,
      "invalid_tail_samples": 22.1,
      "learning_rate": 6.846258385915545e-05,
      "loss": 0.0403,
      "positive_teacher_pairs": 27245.4,
      "query_h_norm_ratio": 0.4409737318754196,
      "relation_loss": 0.004982928326353431,
      "selected_pairs": 27245.4,
      "step": 12320,
      "student_distance_mean": 25.775188636779784,
      "teacher_distance_mean": 0.8608643233776092,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.004982928326353431,
      "weighted_relation_loss": 0.004982928326353431
    },
    {
      "action_query_batch_std": 0.39835159182548524,
      "action_query_norm": 43.228376388549805,
      "action_relation_loss": 0.0049856768921017645,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.534144059869037,
      "flow_matching_loss": 0.037060846015810965,
      "grad_norm": 0.12042859941720963,
      "h_stock_valid_norm": 98.65183944702149,
      "invalid_tail_samples": 24.7,
      "learning_rate": 6.841135204691902e-05,
      "loss": 0.042,
      "positive_teacher_pairs": 26643.9,
      "query_h_norm_ratio": 0.43819558024406435,
      "relation_loss": 0.0049856768921017645,
      "selected_pairs": 26643.9,
      "step": 12330,
      "student_distance_mean": 26.537847328186036,
      "teacher_distance_mean": 0.866787600517273,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0049856768921017645,
      "weighted_relation_loss": 0.0049856768921017645
    },
    {
      "action_query_batch_std": 0.3886413395404816,
      "action_query_norm": 43.18959083557129,
      "action_relation_loss": 0.004813375370576978,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.543498596819457,
      "flow_matching_loss": 0.03432031534612179,
      "grad_norm": 0.1253133863210678,
      "h_stock_valid_norm": 98.78429107666015,
      "invalid_tail_samples": 25.5,
      "learning_rate": 6.836009786315377e-05,
      "loss": 0.0391,
      "positive_teacher_pairs": 26457.6,
      "query_h_norm_ratio": 0.43721458315849304,
      "relation_loss": 0.004813375370576978,
      "selected_pairs": 26457.6,
      "step": 12340,
      "student_distance_mean": 25.924137687683107,
      "teacher_distance_mean": 0.8514745652675628,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.004813375370576978,
      "weighted_relation_loss": 0.004813375370576978
    },
    {
      "action_query_batch_std": 0.39097069203853607,
      "action_query_norm": 43.19344139099121,
      "action_relation_loss": 0.004773106565698982,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.552853133769878,
      "flow_matching_loss": 0.03687068037688732,
      "grad_norm": 0.11154551059007645,
      "h_stock_valid_norm": 97.67199172973633,
      "invalid_tail_samples": 23.5,
      "learning_rate": 6.830882137013839e-05,
      "loss": 0.0416,
      "positive_teacher_pairs": 26918.0,
      "query_h_norm_ratio": 0.4422420382499695,
      "relation_loss": 0.004773106565698982,
      "selected_pairs": 26918.0,
      "step": 12350,
      "student_distance_mean": 26.06060791015625,
      "teacher_distance_mean": 0.8668803215026856,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.004773106565698982,
      "weighted_relation_loss": 0.004773106565698982
    },
    {
      "action_query_batch_std": 0.3877862632274628,
      "action_query_norm": 43.19368476867676,
      "action_relation_loss": 0.004853325011208653,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.5622076707203,
      "flow_matching_loss": 0.03513121996074915,
      "grad_norm": 0.11012646555900574,
      "h_stock_valid_norm": 97.26293411254883,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.825752263017863e-05,
      "loss": 0.04,
      "positive_teacher_pairs": 26665.1,
      "query_h_norm_ratio": 0.4440945237874985,
      "relation_loss": 0.004853325011208653,
      "selected_pairs": 26665.1,
      "step": 12360,
      "student_distance_mean": 25.80409507751465,
      "teacher_distance_mean": 0.8502260863780975,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.004853325011208653,
      "weighted_relation_loss": 0.004853325011208653
    },
    {
      "action_query_batch_std": 0.38914522230625154,
      "action_query_norm": 43.187782287597656,
      "action_relation_loss": 0.004662699531763792,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.57156220767072,
      "flow_matching_loss": 0.03577819205820561,
      "grad_norm": 0.11385368555784225,
      "h_stock_valid_norm": 97.35119400024413,
      "invalid_tail_samples": 24.1,
      "learning_rate": 6.820620170560731e-05,
      "loss": 0.0404,
      "positive_teacher_pairs": 26782.0,
      "query_h_norm_ratio": 0.4436326950788498,
      "relation_loss": 0.004662699531763792,
      "selected_pairs": 26782.0,
      "step": 12370,
      "student_distance_mean": 25.931839179992675,
      "teacher_distance_mean": 0.8571113288402558,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.004662699531763792,
      "weighted_relation_loss": 0.004662699531763792
    },
    {
      "action_query_batch_std": 0.3856846570968628,
      "action_query_norm": 43.16583366394043,
      "action_relation_loss": 0.00488476431928575,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.580916744621142,
      "flow_matching_loss": 0.03494501765817404,
      "grad_norm": 0.12739089131355286,
      "h_stock_valid_norm": 97.44319915771484,
      "invalid_tail_samples": 21.9,
      "learning_rate": 6.815485865878418e-05,
      "loss": 0.0398,
      "positive_teacher_pairs": 27293.4,
      "query_h_norm_ratio": 0.44298777282238005,
      "relation_loss": 0.00488476431928575,
      "selected_pairs": 27293.4,
      "step": 12380,
      "student_distance_mean": 25.67938804626465,
      "teacher_distance_mean": 0.8525603771209717,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.00488476431928575,
      "weighted_relation_loss": 0.00488476431928575
    },
    {
      "action_query_batch_std": 0.38703511357307435,
      "action_query_norm": 43.17501029968262,
      "action_relation_loss": 0.004883430339396,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.590271281571562,
      "flow_matching_loss": 0.03336926251649856,
      "grad_norm": 0.11579976230859756,
      "h_stock_valid_norm": 97.59675674438476,
      "invalid_tail_samples": 23.5,
      "learning_rate": 6.810349355209588e-05,
      "loss": 0.0383,
      "positive_teacher_pairs": 26920.8,
      "query_h_norm_ratio": 0.442384272813797,
      "relation_loss": 0.004883430339396,
      "selected_pairs": 26920.8,
      "step": 12390,
      "student_distance_mean": 25.738611793518068,
      "teacher_distance_mean": 0.8506368637084961,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.004883430339396,
      "weighted_relation_loss": 0.004883430339396
    },
    {
      "action_query_batch_std": 0.3898759543895721,
      "action_query_norm": 43.18751602172851,
      "action_relation_loss": 0.004810015857219696,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.599625818521982,
      "flow_matching_loss": 0.03676989674568176,
      "grad_norm": 0.10494931042194366,
      "h_stock_valid_norm": 97.43874435424804,
      "invalid_tail_samples": 21.6,
      "learning_rate": 6.805210644795588e-05,
      "loss": 0.0416,
      "positive_teacher_pairs": 27360.9,
      "query_h_norm_ratio": 0.44323583841323855,
      "relation_loss": 0.004810015857219696,
      "selected_pairs": 27360.9,
      "step": 12400,
      "student_distance_mean": 26.01726722717285,
      "teacher_distance_mean": 0.8543310225009918,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.004810015857219696,
      "weighted_relation_loss": 0.004810015857219696
    },
    {
      "action_query_batch_std": 0.38900744915008545,
      "action_query_norm": 43.18329849243164,
      "action_relation_loss": 0.004705598531290889,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.608980355472404,
      "flow_matching_loss": 0.03567432556301355,
      "grad_norm": 0.09587412327528,
      "h_stock_valid_norm": 97.3144920349121,
      "invalid_tail_samples": 20.5,
      "learning_rate": 6.800069740880433e-05,
      "loss": 0.0404,
      "positive_teacher_pairs": 27626.9,
      "query_h_norm_ratio": 0.4437572717666626,
      "relation_loss": 0.004705598531290889,
      "selected_pairs": 27626.9,
      "step": 12410,
      "student_distance_mean": 25.919590568542482,
      "teacher_distance_mean": 0.8532103717327117,
      "valid_action_samples": 235.5,
      "weighted_auxiliary_loss": 0.004705598531290889,
      "weighted_relation_loss": 0.004705598531290889
    },
    {
      "action_query_batch_std": 0.3836304247379303,
      "action_query_norm": 43.159786987304685,
      "action_relation_loss": 0.005075827101245522,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.618334892422824,
      "flow_matching_loss": 0.03590586371719837,
      "grad_norm": 0.17414139211177826,
      "h_stock_valid_norm": 97.8381362915039,
      "invalid_tail_samples": 25.0,
      "learning_rate": 6.794926649710807e-05,
      "loss": 0.041,
      "positive_teacher_pairs": 26575.5,
      "query_h_norm_ratio": 0.4411431759595871,
      "relation_loss": 0.005075827101245522,
      "selected_pairs": 26575.5,
      "step": 12420,
      "student_distance_mean": 25.47805233001709,
      "teacher_distance_mean": 0.8489383459091187,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.005075827101245522,
      "weighted_relation_loss": 0.005075827101245522
    },
    {
      "action_query_batch_std": 0.3890911877155304,
      "action_query_norm": 43.172562789916995,
      "action_relation_loss": 0.005100289499387145,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.627689429373246,
      "flow_matching_loss": 0.03516851682215929,
      "grad_norm": 0.1259092539548874,
      "h_stock_valid_norm": 98.4355972290039,
      "invalid_tail_samples": 23.6,
      "learning_rate": 6.789781377536052e-05,
      "loss": 0.0403,
      "positive_teacher_pairs": 26898.8,
      "query_h_norm_ratio": 0.43858826756477354,
      "relation_loss": 0.005100289499387145,
      "selected_pairs": 26898.8,
      "step": 12430,
      "student_distance_mean": 25.91706790924072,
      "teacher_distance_mean": 0.8564731478691101,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.005100289499387145,
      "weighted_relation_loss": 0.005100289499387145
    },
    {
      "action_query_batch_std": 0.39068078994750977,
      "action_query_norm": 43.17847023010254,
      "action_relation_loss": 0.004855873947963119,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.637043966323667,
      "flow_matching_loss": 0.03668820634484291,
      "grad_norm": 0.11099249124526978,
      "h_stock_valid_norm": 98.31669387817382,
      "invalid_tail_samples": 23.2,
      "learning_rate": 6.784633930608158e-05,
      "loss": 0.0415,
      "positive_teacher_pairs": 26990.0,
      "query_h_norm_ratio": 0.43917875587940214,
      "relation_loss": 0.004855873947963119,
      "selected_pairs": 26990.0,
      "step": 12440,
      "student_distance_mean": 26.014722442626955,
      "teacher_distance_mean": 0.8504588544368744,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.004855873947963119,
      "weighted_relation_loss": 0.004855873947963119
    },
    {
      "action_query_batch_std": 0.3848626524209976,
      "action_query_norm": 43.15357093811035,
      "action_relation_loss": 0.005555765656754375,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.646398503274089,
      "flow_matching_loss": 0.03682552352547645,
      "grad_norm": 0.11625462025403976,
      "h_stock_valid_norm": 97.77876281738281,
      "invalid_tail_samples": 27.0,
      "learning_rate": 6.779484315181759e-05,
      "loss": 0.0424,
      "positive_teacher_pairs": 26112.4,
      "query_h_norm_ratio": 0.4413447052240372,
      "relation_loss": 0.005555765656754375,
      "selected_pairs": 26112.4,
      "step": 12450,
      "student_distance_mean": 25.553354835510255,
      "teacher_distance_mean": 0.8444285988807678,
      "valid_action_samples": 229.0,
      "weighted_auxiliary_loss": 0.005555765656754375,
      "weighted_relation_loss": 0.005555765656754375
    },
    {
      "action_query_batch_std": 0.389121475815773,
      "action_query_norm": 43.163828659057614,
      "action_relation_loss": 0.005011129844933748,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.655753040224509,
      "flow_matching_loss": 0.03605103753507137,
      "grad_norm": 0.10771086066961288,
      "h_stock_valid_norm": 97.47791442871093,
      "invalid_tail_samples": 23.0,
      "learning_rate": 6.774332537514122e-05,
      "loss": 0.0411,
      "positive_teacher_pairs": 27047.1,
      "query_h_norm_ratio": 0.44281065464019775,
      "relation_loss": 0.005011129844933748,
      "selected_pairs": 27047.1,
      "step": 12460,
      "student_distance_mean": 25.75515193939209,
      "teacher_distance_mean": 0.8564913392066955,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.005011129844933748,
      "weighted_relation_loss": 0.005011129844933748
    },
    {
      "action_query_batch_std": 0.3844337075948715,
      "action_query_norm": 43.143608474731444,
      "action_relation_loss": 0.004898809920996427,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.665107577174929,
      "flow_matching_loss": 0.03604909144341946,
      "grad_norm": 0.10678597539663315,
      "h_stock_valid_norm": 97.13795318603516,
      "invalid_tail_samples": 25.8,
      "learning_rate": 6.769178603865143e-05,
      "loss": 0.0409,
      "positive_teacher_pairs": 26392.8,
      "query_h_norm_ratio": 0.44415405988693235,
      "relation_loss": 0.004898809920996427,
      "selected_pairs": 26392.8,
      "step": 12470,
      "student_distance_mean": 25.530629348754882,
      "teacher_distance_mean": 0.8426762104034424,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.004898809920996427,
      "weighted_relation_loss": 0.004898809920996427
    },
    {
      "action_query_batch_std": 0.38709868788719176,
      "action_query_norm": 43.154323959350585,
      "action_relation_loss": 0.004936201684176922,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.674462114125351,
      "flow_matching_loss": 0.036996080353856084,
      "grad_norm": 0.10442667454481125,
      "h_stock_valid_norm": 96.68854064941407,
      "invalid_tail_samples": 22.1,
      "learning_rate": 6.764022520497337e-05,
      "loss": 0.0419,
      "positive_teacher_pairs": 27253.1,
      "query_h_norm_ratio": 0.4463287204504013,
      "relation_loss": 0.004936201684176922,
      "selected_pairs": 27253.1,
      "step": 12480,
      "student_distance_mean": 25.796585464477538,
      "teacher_distance_mean": 0.8486266553401947,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.004936201684176922,
      "weighted_relation_loss": 0.004936201684176922
    },
    {
      "action_query_batch_std": 0.38855124413967135,
      "action_query_norm": 43.15898246765137,
      "action_relation_loss": 0.005076670972630382,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.683816651075771,
      "flow_matching_loss": 0.036908704414963724,
      "grad_norm": 0.11656897515058517,
      "h_stock_valid_norm": 96.11629257202148,
      "invalid_tail_samples": 25.8,
      "learning_rate": 6.758864293675833e-05,
      "loss": 0.042,
      "positive_teacher_pairs": 26385.2,
      "query_h_norm_ratio": 0.4490300536155701,
      "relation_loss": 0.005076670972630382,
      "selected_pairs": 26385.2,
      "step": 12490,
      "student_distance_mean": 25.903744316101076,
      "teacher_distance_mean": 0.8500442504882812,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.005076670972630382,
      "weighted_relation_loss": 0.005076670972630382
    },
    {
      "action_query_batch_std": 0.38986354172229765,
      "action_query_norm": 43.15625915527344,
      "action_relation_loss": 0.004813744965940714,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.693171188026193,
      "flow_matching_loss": 0.03643207997083664,
      "grad_norm": 0.14705613255500793,
      "h_stock_valid_norm": 96.86329650878906,
      "invalid_tail_samples": 23.7,
      "learning_rate": 6.753703929668362e-05,
      "loss": 0.0412,
      "positive_teacher_pairs": 26874.4,
      "query_h_norm_ratio": 0.445542111992836,
      "relation_loss": 0.004813744965940714,
      "selected_pairs": 26874.4,
      "step": 12500,
      "student_distance_mean": 25.885469818115233,
      "teacher_distance_mean": 0.8551168203353882,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.004813744965940714,
      "weighted_relation_loss": 0.004813744965940714
    },
    {
      "action_query_batch_std": 0.3897085964679718,
      "action_query_norm": 43.156167602539064,
      "action_relation_loss": 0.004813571181148291,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.702525724976613,
      "flow_matching_loss": 0.03452309183776379,
      "grad_norm": 0.10810815542936325,
      "h_stock_valid_norm": 97.35102462768555,
      "invalid_tail_samples": 24.1,
      "learning_rate": 6.748541434745253e-05,
      "loss": 0.0393,
      "positive_teacher_pairs": 26785.2,
      "query_h_norm_ratio": 0.44330733418464663,
      "relation_loss": 0.004813571181148291,
      "selected_pairs": 26785.2,
      "step": 12510,
      "student_distance_mean": 25.769840431213378,
      "teacher_distance_mean": 0.8509969055652619,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.004813571181148291,
      "weighted_relation_loss": 0.004813571181148291
    },
    {
      "action_query_batch_std": 0.3918161004781723,
      "action_query_norm": 43.16356353759765,
      "action_relation_loss": 0.0047151186969131235,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.711880261927035,
      "flow_matching_loss": 0.035388358123600486,
      "grad_norm": 0.10244794934988022,
      "h_stock_valid_norm": 97.40381164550782,
      "invalid_tail_samples": 21.9,
      "learning_rate": 6.743376815179424e-05,
      "loss": 0.0401,
      "positive_teacher_pairs": 27300.0,
      "query_h_norm_ratio": 0.44314214289188386,
      "relation_loss": 0.0047151186969131235,
      "selected_pairs": 27300.0,
      "step": 12520,
      "student_distance_mean": 26.089097213745116,
      "teacher_distance_mean": 0.8494042336940766,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.0047151186969131235,
      "weighted_relation_loss": 0.0047151186969131235
    },
    {
      "action_query_batch_std": 0.38929826617240904,
      "action_query_norm": 43.15934371948242,
      "action_relation_loss": 0.004932105774059892,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.721234798877456,
      "flow_matching_loss": 0.037514643743634224,
      "grad_norm": 0.13085462152957916,
      "h_stock_valid_norm": 97.281103515625,
      "invalid_tail_samples": 20.3,
      "learning_rate": 6.738210077246376e-05,
      "loss": 0.0424,
      "positive_teacher_pairs": 27669.9,
      "query_h_norm_ratio": 0.44365967214107516,
      "relation_loss": 0.004932105774059892,
      "selected_pairs": 27669.9,
      "step": 12530,
      "student_distance_mean": 25.85884952545166,
      "teacher_distance_mean": 0.8494055509567261,
      "valid_action_samples": 235.7,
      "weighted_auxiliary_loss": 0.004932105774059892,
      "weighted_relation_loss": 0.004932105774059892
    },
    {
      "action_query_batch_std": 0.3921744763851166,
      "action_query_norm": 43.16047744750976,
      "action_relation_loss": 0.004793598875403404,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.730589335827876,
      "flow_matching_loss": 0.03627433329820633,
      "grad_norm": 0.12028191238641739,
      "h_stock_valid_norm": 97.80946197509766,
      "invalid_tail_samples": 26.0,
      "learning_rate": 6.733041227224181e-05,
      "loss": 0.0411,
      "positive_teacher_pairs": 26340.6,
      "query_h_norm_ratio": 0.4412823557853699,
      "relation_loss": 0.004793598875403404,
      "selected_pairs": 26340.6,
      "step": 12540,
      "student_distance_mean": 26.15982303619385,
      "teacher_distance_mean": 0.8600144147872925,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.004793598875403404,
      "weighted_relation_loss": 0.004793598875403404
    },
    {
      "action_query_batch_std": 0.39422036707401276,
      "action_query_norm": 43.168196487426755,
      "action_relation_loss": 0.004673920013010502,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.739943872778298,
      "flow_matching_loss": 0.03597562499344349,
      "grad_norm": 0.10928499698638916,
      "h_stock_valid_norm": 98.983447265625,
      "invalid_tail_samples": 20.9,
      "learning_rate": 6.72787027139348e-05,
      "loss": 0.0406,
      "positive_teacher_pairs": 27527.7,
      "query_h_norm_ratio": 0.43611789047718047,
      "relation_loss": 0.004673920013010502,
      "selected_pairs": 27527.7,
      "step": 12550,
      "student_distance_mean": 26.177703857421875,
      "teacher_distance_mean": 0.8598371744155884,
      "valid_action_samples": 235.1,
      "weighted_auxiliary_loss": 0.004673920013010502,
      "weighted_relation_loss": 0.004673920013010502
    },
    {
      "action_query_batch_std": 0.38718817234039304,
      "action_query_norm": 43.12747573852539,
      "action_relation_loss": 0.004841738566756249,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.749298409728718,
      "flow_matching_loss": 0.035319932363927366,
      "grad_norm": 0.12923529744148254,
      "h_stock_valid_norm": 98.44679794311523,
      "invalid_tail_samples": 26.0,
      "learning_rate": 6.72269721603747e-05,
      "loss": 0.0402,
      "positive_teacher_pairs": 26340.6,
      "query_h_norm_ratio": 0.43809067606925967,
      "relation_loss": 0.004841738566756249,
      "selected_pairs": 26340.6,
      "step": 12560,
      "student_distance_mean": 25.784235191345214,
      "teacher_distance_mean": 0.8561491966247559,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.004841738566756249,
      "weighted_relation_loss": 0.004841738566756249
    },
    {
      "action_query_batch_std": 0.3885307818651199,
      "action_query_norm": 43.140198135375975,
      "action_relation_loss": 0.0048566179815679785,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.75865294667914,
      "flow_matching_loss": 0.03271820954978466,
      "grad_norm": 0.12022505700588226,
      "h_stock_valid_norm": 97.57799072265625,
      "invalid_tail_samples": 22.9,
      "learning_rate": 6.717522067441904e-05,
      "loss": 0.0376,
      "positive_teacher_pairs": 27059.0,
      "query_h_norm_ratio": 0.4421120911836624,
      "relation_loss": 0.0048566179815679785,
      "selected_pairs": 27059.0,
      "step": 12570,
      "student_distance_mean": 25.854004287719725,
      "teacher_distance_mean": 0.8545519232749939,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0048566179815679785,
      "weighted_relation_loss": 0.0048566179815679785
    },
    {
      "action_query_batch_std": 0.39409770667552946,
      "action_query_norm": 43.15835304260254,
      "action_relation_loss": 0.005000242963433266,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.76800748362956,
      "flow_matching_loss": 0.035603438317775724,
      "grad_norm": 0.11752024292945862,
      "h_stock_valid_norm": 97.52516479492188,
      "invalid_tail_samples": 22.3,
      "learning_rate": 6.712344831895074e-05,
      "loss": 0.0406,
      "positive_teacher_pairs": 27197.7,
      "query_h_norm_ratio": 0.4425387501716614,
      "relation_loss": 0.005000242963433266,
      "selected_pairs": 27197.7,
      "step": 12580,
      "student_distance_mean": 26.22073745727539,
      "teacher_distance_mean": 0.8543322265148163,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.005000242963433266,
      "weighted_relation_loss": 0.005000242963433266
    },
    {
      "action_query_batch_std": 0.3959300607442856,
      "action_query_norm": 43.15124588012695,
      "action_relation_loss": 0.0047385191544890405,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.77736202057998,
      "flow_matching_loss": 0.03560601193457842,
      "grad_norm": 0.08736510574817657,
      "h_stock_valid_norm": 97.3834098815918,
      "invalid_tail_samples": 24.3,
      "learning_rate": 6.707165515687811e-05,
      "loss": 0.0403,
      "positive_teacher_pairs": 26733.8,
      "query_h_norm_ratio": 0.4431088387966156,
      "relation_loss": 0.0047385191544890405,
      "selected_pairs": 26733.8,
      "step": 12590,
      "student_distance_mean": 26.318320655822752,
      "teacher_distance_mean": 0.869545578956604,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0047385191544890405,
      "weighted_relation_loss": 0.0047385191544890405
    },
    {
      "action_query_batch_std": 0.38813886046409607,
      "action_query_norm": 43.12968978881836,
      "action_relation_loss": 0.0047956667840480804,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.786716557530402,
      "flow_matching_loss": 0.03292504902929068,
      "grad_norm": 0.13628286123275757,
      "h_stock_valid_norm": 97.55255661010742,
      "invalid_tail_samples": 23.2,
      "learning_rate": 6.70198412511347e-05,
      "loss": 0.0377,
      "positive_teacher_pairs": 26995.4,
      "query_h_norm_ratio": 0.4421203911304474,
      "relation_loss": 0.0047956667840480804,
      "selected_pairs": 26995.4,
      "step": 12600,
      "student_distance_mean": 25.826380920410156,
      "teacher_distance_mean": 0.8521739065647125,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0047956667840480804,
      "weighted_relation_loss": 0.0047956667840480804
    },
    {
      "action_query_batch_std": 0.3860958516597748,
      "action_query_norm": 43.1182243347168,
      "action_relation_loss": 0.0047200356610119345,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.796071094480823,
      "flow_matching_loss": 0.036023267358541486,
      "grad_norm": 0.141715869307518,
      "h_stock_valid_norm": 97.10979080200195,
      "invalid_tail_samples": 25.1,
      "learning_rate": 6.69680066646793e-05,
      "loss": 0.0407,
      "positive_teacher_pairs": 26547.1,
      "query_h_norm_ratio": 0.4440243989229202,
      "relation_loss": 0.0047200356610119345,
      "selected_pairs": 26547.1,
      "step": 12610,
      "student_distance_mean": 25.665165519714357,
      "teacher_distance_mean": 0.8456077456474305,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0047200356610119345,
      "weighted_relation_loss": 0.0047200356610119345
    },
    {
      "action_query_batch_std": 0.3852545231580734,
      "action_query_norm": 43.10961837768555,
      "action_relation_loss": 0.0051979562267661095,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.805425631431245,
      "flow_matching_loss": 0.03532188143581152,
      "grad_norm": 0.1052018404006958,
      "h_stock_valid_norm": 97.0485450744629,
      "invalid_tail_samples": 22.1,
      "learning_rate": 6.691615146049583e-05,
      "loss": 0.0405,
      "positive_teacher_pairs": 27256.1,
      "query_h_norm_ratio": 0.4442120373249054,
      "relation_loss": 0.0051979562267661095,
      "selected_pairs": 27256.1,
      "step": 12620,
      "student_distance_mean": 25.651333618164063,
      "teacher_distance_mean": 0.8447196364402771,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0051979562267661095,
      "weighted_relation_loss": 0.0051979562267661095
    },
    {
      "action_query_batch_std": 0.3920631974935532,
      "action_query_norm": 43.131574630737305,
      "action_relation_loss": 0.004874042375013233,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.814780168381665,
      "flow_matching_loss": 0.03686222527176142,
      "grad_norm": 0.10982178151607513,
      "h_stock_valid_norm": 97.5042495727539,
      "invalid_tail_samples": 25.1,
      "learning_rate": 6.686427570159324e-05,
      "loss": 0.0417,
      "positive_teacher_pairs": 26552.8,
      "query_h_norm_ratio": 0.4423582911491394,
      "relation_loss": 0.004874042375013233,
      "selected_pairs": 26552.8,
      "step": 12630,
      "student_distance_mean": 26.083205795288087,
      "teacher_distance_mean": 0.8558188796043396,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.004874042375013233,
      "weighted_relation_loss": 0.004874042375013233
    },
    {
      "action_query_batch_std": 0.39099459946155546,
      "action_query_norm": 43.137102127075195,
      "action_relation_loss": 0.0048155120573937895,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.824134705332087,
      "flow_matching_loss": 0.034847603738307954,
      "grad_norm": 0.158603698015213,
      "h_stock_valid_norm": 97.99680709838867,
      "invalid_tail_samples": 23.2,
      "learning_rate": 6.681237945100548e-05,
      "loss": 0.0397,
      "positive_teacher_pairs": 26994.9,
      "query_h_norm_ratio": 0.44019280970096586,
      "relation_loss": 0.0048155120573937895,
      "selected_pairs": 26994.9,
      "step": 12640,
      "student_distance_mean": 26.124226379394532,
      "teacher_distance_mean": 0.8528707265853882,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0048155120573937895,
      "weighted_relation_loss": 0.0048155120573937895
    },
    {
      "action_query_batch_std": 0.38440240919589996,
      "action_query_norm": 43.10652046203613,
      "action_relation_loss": 0.004955045273527503,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.833489242282507,
      "flow_matching_loss": 0.036119967326521875,
      "grad_norm": 0.14312361180782318,
      "h_stock_valid_norm": 98.48640365600586,
      "invalid_tail_samples": 24.0,
      "learning_rate": 6.676046277179139e-05,
      "loss": 0.0411,
      "positive_teacher_pairs": 26810.6,
      "query_h_norm_ratio": 0.43769248127937316,
      "relation_loss": 0.004955045273527503,
      "selected_pairs": 26810.6,
      "step": 12650,
      "student_distance_mean": 25.561709785461424,
      "teacher_distance_mean": 0.8394081294536591,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.004955045273527503,
      "weighted_relation_loss": 0.004955045273527503
    },
    {
      "action_query_batch_std": 0.3915784448385239,
      "action_query_norm": 43.13160095214844,
      "action_relation_loss": 0.004976786067709326,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.842843779232927,
      "flow_matching_loss": 0.03765041008591652,
      "grad_norm": 0.15406984090805054,
      "h_stock_valid_norm": 98.99320068359376,
      "invalid_tail_samples": 22.5,
      "learning_rate": 6.670852572703462e-05,
      "loss": 0.0426,
      "positive_teacher_pairs": 27151.7,
      "query_h_norm_ratio": 0.4357051491737366,
      "relation_loss": 0.004976786067709326,
      "selected_pairs": 27151.7,
      "step": 12660,
      "student_distance_mean": 25.90116481781006,
      "teacher_distance_mean": 0.8494193375110626,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.004976786067709326,
      "weighted_relation_loss": 0.004976786067709326
    },
    {
      "action_query_batch_std": 0.38981625735759734,
      "action_query_norm": 43.11791839599609,
      "action_relation_loss": 0.0049773703794926405,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.85219831618335,
      "flow_matching_loss": 0.037895596399903295,
      "grad_norm": 0.13214142620563507,
      "h_stock_valid_norm": 98.39393157958985,
      "invalid_tail_samples": 26.1,
      "learning_rate": 6.665656837984359e-05,
      "loss": 0.0429,
      "positive_teacher_pairs": 26317.5,
      "query_h_norm_ratio": 0.43821996748447417,
      "relation_loss": 0.0049773703794926405,
      "selected_pairs": 26317.5,
      "step": 12670,
      "student_distance_mean": 26.026344299316406,
      "teacher_distance_mean": 0.8551294147968292,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.0049773703794926405,
      "weighted_relation_loss": 0.0049773703794926405
    },
    {
      "action_query_batch_std": 0.3841795831918716,
      "action_query_norm": 43.1010326385498,
      "action_relation_loss": 0.004940007021650672,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.86155285313377,
      "flow_matching_loss": 0.03530110828578472,
      "grad_norm": 0.11819635331630707,
      "h_stock_valid_norm": 97.91302032470703,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.660459079335135e-05,
      "loss": 0.0402,
      "positive_teacher_pairs": 26664.8,
      "query_h_norm_ratio": 0.4402011692523956,
      "relation_loss": 0.004940007021650672,
      "selected_pairs": 26664.8,
      "step": 12680,
      "student_distance_mean": 25.701562690734864,
      "teacher_distance_mean": 0.8449822902679444,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.004940007021650672,
      "weighted_relation_loss": 0.004940007021650672
    },
    {
      "action_query_batch_std": 0.38578910529613497,
      "action_query_norm": 43.11127471923828,
      "action_relation_loss": 0.004907558811828494,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.870907390084191,
      "flow_matching_loss": 0.03506213035434484,
      "grad_norm": 0.11208745837211609,
      "h_stock_valid_norm": 98.01110153198242,
      "invalid_tail_samples": 23.2,
      "learning_rate": 6.655259303071558e-05,
      "loss": 0.04,
      "positive_teacher_pairs": 26989.2,
      "query_h_norm_ratio": 0.4398625761270523,
      "relation_loss": 0.004907558811828494,
      "selected_pairs": 26989.2,
      "step": 12690,
      "student_distance_mean": 25.738972091674803,
      "teacher_distance_mean": 0.8401512026786804,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.004907558811828494,
      "weighted_relation_loss": 0.004907558811828494
    },
    {
      "action_query_batch_std": 0.39500809609889986,
      "action_query_norm": 43.134179306030276,
      "action_relation_loss": 0.0047579821664839985,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.880261927034612,
      "flow_matching_loss": 0.035664252191782,
      "grad_norm": 0.13184206187725067,
      "h_stock_valid_norm": 98.39558944702148,
      "invalid_tail_samples": 22.1,
      "learning_rate": 6.650057515511848e-05,
      "loss": 0.0404,
      "positive_teacher_pairs": 27255.2,
      "query_h_norm_ratio": 0.4383854061365128,
      "relation_loss": 0.0047579821664839985,
      "selected_pairs": 27255.2,
      "step": 12700,
      "student_distance_mean": 26.369264221191408,
      "teacher_distance_mean": 0.8681148290634155,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0047579821664839985,
      "weighted_relation_loss": 0.0047579821664839985
    },
    {
      "action_query_batch_std": 0.38342928886413574,
      "action_query_norm": 43.08749008178711,
      "action_relation_loss": 0.004983611777424813,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.889616463985032,
      "flow_matching_loss": 0.03763079084455967,
      "grad_norm": 0.11840737611055374,
      "h_stock_valid_norm": 99.17396545410156,
      "invalid_tail_samples": 23.7,
      "learning_rate": 6.644853722976667e-05,
      "loss": 0.0426,
      "positive_teacher_pairs": 26874.4,
      "query_h_norm_ratio": 0.4344662308692932,
      "relation_loss": 0.004983611777424813,
      "selected_pairs": 26874.4,
      "step": 12710,
      "student_distance_mean": 25.37904644012451,
      "teacher_distance_mean": 0.8405336320400238,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.004983611777424813,
      "weighted_relation_loss": 0.004983611777424813
    },
    {
      "action_query_batch_std": 0.38946017920970916,
      "action_query_norm": 43.114186477661136,
      "action_relation_loss": 0.0048932964913547036,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.898971000935454,
      "flow_matching_loss": 0.03529320601373911,
      "grad_norm": 0.11341077089309692,
      "h_stock_valid_norm": 98.29655227661132,
      "invalid_tail_samples": 23.6,
      "learning_rate": 6.639647931789113e-05,
      "loss": 0.0402,
      "positive_teacher_pairs": 26895.6,
      "query_h_norm_ratio": 0.4386164307594299,
      "relation_loss": 0.0048932964913547036,
      "selected_pairs": 26895.6,
      "step": 12720,
      "student_distance_mean": 26.03762283325195,
      "teacher_distance_mean": 0.8578393399715424,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0048932964913547036,
      "weighted_relation_loss": 0.0048932964913547036
    },
    {
      "action_query_batch_std": 0.38923807442188263,
      "action_query_norm": 43.10767021179199,
      "action_relation_loss": 0.004912177985534072,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.908325537885874,
      "flow_matching_loss": 0.036116881668567656,
      "grad_norm": 0.10121864080429077,
      "h_stock_valid_norm": 98.45832595825195,
      "invalid_tail_samples": 22.7,
      "learning_rate": 6.634440148274713e-05,
      "loss": 0.041,
      "positive_teacher_pairs": 27104.5,
      "query_h_norm_ratio": 0.4378317266702652,
      "relation_loss": 0.004912177985534072,
      "selected_pairs": 27104.5,
      "step": 12730,
      "student_distance_mean": 25.84898109436035,
      "teacher_distance_mean": 0.8491608560085296,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.004912177985534072,
      "weighted_relation_loss": 0.004912177985534072
    },
    {
      "action_query_batch_std": 0.3924959063529968,
      "action_query_norm": 43.12055358886719,
      "action_relation_loss": 0.004790364298969507,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.917680074836296,
      "flow_matching_loss": 0.03618344981223345,
      "grad_norm": 0.10651486366987228,
      "h_stock_valid_norm": 99.28376083374023,
      "invalid_tail_samples": 23.8,
      "learning_rate": 6.629230378761415e-05,
      "loss": 0.041,
      "positive_teacher_pairs": 26848.5,
      "query_h_norm_ratio": 0.43432196974754333,
      "relation_loss": 0.004790364298969507,
      "selected_pairs": 26848.5,
      "step": 12740,
      "student_distance_mean": 26.0219388961792,
      "teacher_distance_mean": 0.8555347263813019,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.004790364298969507,
      "weighted_relation_loss": 0.004790364298969507
    },
    {
      "action_query_batch_std": 0.39285233318805696,
      "action_query_norm": 43.11502647399902,
      "action_relation_loss": 0.004831867991015315,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.927034611786716,
      "flow_matching_loss": 0.03418030608445406,
      "grad_norm": 0.08830320835113525,
      "h_stock_valid_norm": 98.89865188598633,
      "invalid_tail_samples": 25.0,
      "learning_rate": 6.624018629579581e-05,
      "loss": 0.039,
      "positive_teacher_pairs": 26568.8,
      "query_h_norm_ratio": 0.43596426844596864,
      "relation_loss": 0.004831867991015315,
      "selected_pairs": 26568.8,
      "step": 12750,
      "student_distance_mean": 26.066128540039063,
      "teacher_distance_mean": 0.8539229691028595,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.004831867991015315,
      "weighted_relation_loss": 0.004831867991015315
    },
    {
      "action_query_batch_std": 0.3970440536737442,
      "action_query_norm": 43.11936645507812,
      "action_relation_loss": 0.004772882582619787,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.936389148737138,
      "flow_matching_loss": 0.038384553790092465,
      "grad_norm": 0.12057249993085861,
      "h_stock_valid_norm": 96.89178009033203,
      "invalid_tail_samples": 23.2,
      "learning_rate": 6.618804907061976e-05,
      "loss": 0.0432,
      "positive_teacher_pairs": 26992.3,
      "query_h_norm_ratio": 0.4450359046459198,
      "relation_loss": 0.004772882582619787,
      "selected_pairs": 26992.3,
      "step": 12760,
      "student_distance_mean": 26.36066303253174,
      "teacher_distance_mean": 0.8688903987407685,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.004772882582619787,
      "weighted_relation_loss": 0.004772882582619787
    },
    {
      "action_query_batch_std": 0.3887550175189972,
      "action_query_norm": 43.08517990112305,
      "action_relation_loss": 0.004895453853532672,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.945743685687559,
      "flow_matching_loss": 0.03442481625825167,
      "grad_norm": 0.13737602531909943,
      "h_stock_valid_norm": 96.07852554321289,
      "invalid_tail_samples": 24.4,
      "learning_rate": 6.613589217543766e-05,
      "loss": 0.0393,
      "positive_teacher_pairs": 26709.7,
      "query_h_norm_ratio": 0.4484410762786865,
      "relation_loss": 0.004895453853532672,
      "selected_pairs": 26709.7,
      "step": 12770,
      "student_distance_mean": 25.84168109893799,
      "teacher_distance_mean": 0.8574809849262237,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.004895453853532672,
      "weighted_relation_loss": 0.004895453853532672
    },
    {
      "action_query_batch_std": 0.3875212430953979,
      "action_query_norm": 43.08170509338379,
      "action_relation_loss": 0.004816579818725586,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.955098222637979,
      "flow_matching_loss": 0.03408526387065649,
      "grad_norm": 0.10738559067249298,
      "h_stock_valid_norm": 97.05473861694335,
      "invalid_tail_samples": 25.4,
      "learning_rate": 6.608371567362504e-05,
      "loss": 0.0389,
      "positive_teacher_pairs": 26477.0,
      "query_h_norm_ratio": 0.4439032465219498,
      "relation_loss": 0.004816579818725586,
      "selected_pairs": 26477.0,
      "step": 12780,
      "student_distance_mean": 25.75792064666748,
      "teacher_distance_mean": 0.8557636618614197,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.004816579818725586,
      "weighted_relation_loss": 0.004816579818725586
    },
    {
      "action_query_batch_std": 0.3895437479019165,
      "action_query_norm": 43.102149200439456,
      "action_relation_loss": 0.004813782451674342,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.9644527595884,
      "flow_matching_loss": 0.035474493727087976,
      "grad_norm": 0.1348114162683487,
      "h_stock_valid_norm": 98.56656341552734,
      "invalid_tail_samples": 25.2,
      "learning_rate": 6.60315196285813e-05,
      "loss": 0.0403,
      "positive_teacher_pairs": 26526.1,
      "query_h_norm_ratio": 0.43729929327964784,
      "relation_loss": 0.004813782451674342,
      "selected_pairs": 26526.1,
      "step": 12790,
      "student_distance_mean": 25.94536762237549,
      "teacher_distance_mean": 0.8512098848819732,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.004813782451674342,
      "weighted_relation_loss": 0.004813782451674342
    },
    {
      "action_query_batch_std": 0.38910619020462034,
      "action_query_norm": 43.09660453796387,
      "action_relation_loss": 0.0047226157039403915,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.973807296538821,
      "flow_matching_loss": 0.03571509122848511,
      "grad_norm": 0.12972965836524963,
      "h_stock_valid_norm": 98.34891281127929,
      "invalid_tail_samples": 26.7,
      "learning_rate": 6.597930410372954e-05,
      "loss": 0.0404,
      "positive_teacher_pairs": 26178.8,
      "query_h_norm_ratio": 0.43820412158966066,
      "relation_loss": 0.0047226157039403915,
      "selected_pairs": 26178.8,
      "step": 12800,
      "student_distance_mean": 25.958534240722656,
      "teacher_distance_mean": 0.8484682381153107,
      "valid_action_samples": 229.3,
      "weighted_auxiliary_loss": 0.0047226157039403915,
      "weighted_relation_loss": 0.0047226157039403915
    },
    {
      "action_query_batch_std": 0.3899178057909012,
      "action_query_norm": 43.086927032470705,
      "action_relation_loss": 0.0048918696120381355,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.983161833489243,
      "flow_matching_loss": 0.03532699588686228,
      "grad_norm": 0.11342179030179977,
      "h_stock_valid_norm": 98.13753356933594,
      "invalid_tail_samples": 24.9,
      "learning_rate": 6.592706916251653e-05,
      "loss": 0.0402,
      "positive_teacher_pairs": 26595.4,
      "query_h_norm_ratio": 0.4390505880117416,
      "relation_loss": 0.0048918696120381355,
      "selected_pairs": 26595.4,
      "step": 12810,
      "student_distance_mean": 25.81594581604004,
      "teacher_distance_mean": 0.8555446863174438,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0048918696120381355,
      "weighted_relation_loss": 0.0048918696120381355
    },
    {
      "action_query_batch_std": 0.3953600287437439,
      "action_query_norm": 43.108120346069335,
      "action_relation_loss": 0.004778676992282271,
      "auxiliary_loss_weight": 1.0,
      "epoch": 11.992516370439663,
      "flow_matching_loss": 0.03597608916461468,
      "grad_norm": 0.10902908444404602,
      "h_stock_valid_norm": 98.09876861572266,
      "invalid_tail_samples": 20.6,
      "learning_rate": 6.587481486841267e-05,
      "loss": 0.0408,
      "positive_teacher_pairs": 27595.8,
      "query_h_norm_ratio": 0.4394394397735596,
      "relation_loss": 0.004778676992282271,
      "selected_pairs": 27595.8,
      "step": 12820,
      "student_distance_mean": 26.206142807006835,
      "teacher_distance_mean": 0.8645118772983551,
      "valid_action_samples": 235.4,
      "weighted_auxiliary_loss": 0.004778676992282271,
      "weighted_relation_loss": 0.004778676992282271
    },
    {
      "action_query_batch_std": 0.3927658021450043,
      "action_query_norm": 43.09030609130859,
      "action_relation_loss": 0.004903903976082802,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.001870907390083,
      "flow_matching_loss": 0.03568384163081646,
      "grad_norm": 0.13303802907466888,
      "h_stock_valid_norm": 98.33267440795899,
      "invalid_tail_samples": 21.0,
      "learning_rate": 6.582254128491184e-05,
      "loss": 0.0406,
      "positive_teacher_pairs": 24547.5,
      "query_h_norm_ratio": 0.43821717202663424,
      "relation_loss": 0.004903903976082802,
      "selected_pairs": 24547.5,
      "step": 12830,
      "student_distance_mean": 26.236202049255372,
      "teacher_distance_mean": 0.862897390127182,
      "valid_action_samples": 215.1,
      "weighted_auxiliary_loss": 0.004903903976082802,
      "weighted_relation_loss": 0.004903903976082802
    },
    {
      "action_query_batch_std": 0.3914261102676392,
      "action_query_norm": 43.08996429443359,
      "action_relation_loss": 0.004632934415712953,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.011225444340505,
      "flow_matching_loss": 0.033304892294108865,
      "grad_norm": 0.11303124576807022,
      "h_stock_valid_norm": 98.24591751098633,
      "invalid_tail_samples": 23.7,
      "learning_rate": 6.577024847553139e-05,
      "loss": 0.0379,
      "positive_teacher_pairs": 26880.0,
      "query_h_norm_ratio": 0.43860510885715487,
      "relation_loss": 0.004632934415712953,
      "selected_pairs": 26880.0,
      "step": 12840,
      "student_distance_mean": 26.07547664642334,
      "teacher_distance_mean": 0.8522614657878875,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.004632934415712953,
      "weighted_relation_loss": 0.004632934415712953
    },
    {
      "action_query_batch_std": 0.38763823807239534,
      "action_query_norm": 43.082113647460936,
      "action_relation_loss": 0.004640293354168534,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.020579981290926,
      "flow_matching_loss": 0.03323731180280447,
      "grad_norm": 0.14708532392978668,
      "h_stock_valid_norm": 97.99544067382813,
      "invalid_tail_samples": 22.2,
      "learning_rate": 6.571793650381201e-05,
      "loss": 0.0379,
      "positive_teacher_pairs": 27223.7,
      "query_h_norm_ratio": 0.43963616490364077,
      "relation_loss": 0.004640293354168534,
      "selected_pairs": 27223.7,
      "step": 12850,
      "student_distance_mean": 25.905486488342284,
      "teacher_distance_mean": 0.8572782814502716,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.004640293354168534,
      "weighted_relation_loss": 0.004640293354168534
    },
    {
      "action_query_batch_std": 0.38574332892894747,
      "action_query_norm": 43.05234565734863,
      "action_relation_loss": 0.004705397551879287,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.029934518241348,
      "flow_matching_loss": 0.03625137247145176,
      "grad_norm": 0.16089023649692535,
      "h_stock_valid_norm": 97.71552276611328,
      "invalid_tail_samples": 24.0,
      "learning_rate": 6.566560543331766e-05,
      "loss": 0.041,
      "positive_teacher_pairs": 26802.0,
      "query_h_norm_ratio": 0.440592086315155,
      "relation_loss": 0.004705397551879287,
      "selected_pairs": 26802.0,
      "step": 12860,
      "student_distance_mean": 25.707371330261232,
      "teacher_distance_mean": 0.8496082186698913,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.004705397551879287,
      "weighted_relation_loss": 0.004705397551879287
    },
    {
      "action_query_batch_std": 0.39597031772136687,
      "action_query_norm": 43.10544090270996,
      "action_relation_loss": 0.004465520335361362,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.039289055191768,
      "flow_matching_loss": 0.0335993517190218,
      "grad_norm": 0.11001808941364288,
      "h_stock_valid_norm": 98.05013961791992,
      "invalid_tail_samples": 23.7,
      "learning_rate": 6.561325532763554e-05,
      "loss": 0.0381,
      "positive_teacher_pairs": 26877.0,
      "query_h_norm_ratio": 0.4396324634552002,
      "relation_loss": 0.004465520335361362,
      "selected_pairs": 26877.0,
      "step": 12870,
      "student_distance_mean": 26.18362216949463,
      "teacher_distance_mean": 0.8593826591968536,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.004465520335361362,
      "weighted_relation_loss": 0.004465520335361362
    },
    {
      "action_query_batch_std": 0.39015225768089296,
      "action_query_norm": 43.07391014099121,
      "action_relation_loss": 0.004424053779803217,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.04864359214219,
      "flow_matching_loss": 0.03451375123113394,
      "grad_norm": 0.11747758090496063,
      "h_stock_valid_norm": 97.34886779785157,
      "invalid_tail_samples": 23.5,
      "learning_rate": 6.556088625037597e-05,
      "loss": 0.0389,
      "positive_teacher_pairs": 26921.9,
      "query_h_norm_ratio": 0.44247927069664,
      "relation_loss": 0.004424053779803217,
      "selected_pairs": 26921.9,
      "step": 12880,
      "student_distance_mean": 26.009869956970213,
      "teacher_distance_mean": 0.8600747942924499,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.004424053779803217,
      "weighted_relation_loss": 0.004424053779803217
    },
    {
      "action_query_batch_std": 0.38840130567550657,
      "action_query_norm": 43.06902084350586,
      "action_relation_loss": 0.004523078235797584,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.05799812909261,
      "flow_matching_loss": 0.032945392467081544,
      "grad_norm": 0.13291926681995392,
      "h_stock_valid_norm": 96.82344131469726,
      "invalid_tail_samples": 24.5,
      "learning_rate": 6.55084982651723e-05,
      "loss": 0.0375,
      "positive_teacher_pairs": 26688.2,
      "query_h_norm_ratio": 0.4448220133781433,
      "relation_loss": 0.004523078235797584,
      "selected_pairs": 26688.2,
      "step": 12890,
      "student_distance_mean": 25.802050399780274,
      "teacher_distance_mean": 0.8572936415672302,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.004523078235797584,
      "weighted_relation_loss": 0.004523078235797584
    },
    {
      "action_query_batch_std": 0.3860392361879349,
      "action_query_norm": 43.06435203552246,
      "action_relation_loss": 0.004588244296610355,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.06735266604303,
      "flow_matching_loss": 0.034005548991262916,
      "grad_norm": 0.10793916136026382,
      "h_stock_valid_norm": 96.7775161743164,
      "invalid_tail_samples": 22.2,
      "learning_rate": 6.54560914356809e-05,
      "loss": 0.0386,
      "positive_teacher_pairs": 27220.9,
      "query_h_norm_ratio": 0.44498770236968993,
      "relation_loss": 0.004588244296610355,
      "selected_pairs": 27220.9,
      "step": 12900,
      "student_distance_mean": 25.777581024169923,
      "teacher_distance_mean": 0.8473742187023163,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.004588244296610355,
      "weighted_relation_loss": 0.004588244296610355
    },
    {
      "action_query_batch_std": 0.38919552564620974,
      "action_query_norm": 43.062042236328125,
      "action_relation_loss": 0.0045366010162979364,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.076707202993452,
      "flow_matching_loss": 0.03343885485082865,
      "grad_norm": 0.13418525457382202,
      "h_stock_valid_norm": 98.36283950805664,
      "invalid_tail_samples": 25.4,
      "learning_rate": 6.540366582558097e-05,
      "loss": 0.038,
      "positive_teacher_pairs": 26476.0,
      "query_h_norm_ratio": 0.43779906034469607,
      "relation_loss": 0.0045366010162979364,
      "selected_pairs": 26476.0,
      "step": 12910,
      "student_distance_mean": 25.961476516723632,
      "teacher_distance_mean": 0.8546046793460846,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0045366010162979364,
      "weighted_relation_loss": 0.0045366010162979364
    },
    {
      "action_query_batch_std": 0.38654270470142366,
      "action_query_norm": 43.05124435424805,
      "action_relation_loss": 0.004793914826586842,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.086061739943872,
      "flow_matching_loss": 0.037527145072817805,
      "grad_norm": 0.16148799657821655,
      "h_stock_valid_norm": 98.92303466796875,
      "invalid_tail_samples": 25.0,
      "learning_rate": 6.53512214985746e-05,
      "loss": 0.0423,
      "positive_teacher_pairs": 26573.0,
      "query_h_norm_ratio": 0.43520056903362275,
      "relation_loss": 0.004793914826586842,
      "selected_pairs": 26573.0,
      "step": 12920,
      "student_distance_mean": 25.744994354248046,
      "teacher_distance_mean": 0.8514582991600037,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.004793914826586842,
      "weighted_relation_loss": 0.004793914826586842
    },
    {
      "action_query_batch_std": 0.38772954046726227,
      "action_query_norm": 43.059675979614255,
      "action_relation_loss": 0.004371252050623298,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.095416276894294,
      "flow_matching_loss": 0.03407405782490969,
      "grad_norm": 0.1430204063653946,
      "h_stock_valid_norm": 98.39852371215821,
      "invalid_tail_samples": 20.6,
      "learning_rate": 6.529875851838658e-05,
      "loss": 0.0384,
      "positive_teacher_pairs": 27602.7,
      "query_h_norm_ratio": 0.4376080632209778,
      "relation_loss": 0.004371252050623298,
      "selected_pairs": 27602.7,
      "step": 12930,
      "student_distance_mean": 25.8487340927124,
      "teacher_distance_mean": 0.8590914368629455,
      "valid_action_samples": 235.4,
      "weighted_auxiliary_loss": 0.004371252050623298,
      "weighted_relation_loss": 0.004371252050623298
    },
    {
      "action_query_batch_std": 0.391165092587471,
      "action_query_norm": 43.063427734375,
      "action_relation_loss": 0.004513913998380303,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.104770813844715,
      "flow_matching_loss": 0.03538392428308725,
      "grad_norm": 0.13891209661960602,
      "h_stock_valid_norm": 97.90703964233398,
      "invalid_tail_samples": 21.3,
      "learning_rate": 6.52462769487644e-05,
      "loss": 0.0399,
      "positive_teacher_pairs": 27436.1,
      "query_h_norm_ratio": 0.4398435831069946,
      "relation_loss": 0.004513913998380303,
      "selected_pairs": 27436.1,
      "step": 12940,
      "student_distance_mean": 26.049435424804688,
      "teacher_distance_mean": 0.8584594368934632,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.004513913998380303,
      "weighted_relation_loss": 0.004513913998380303
    },
    {
      "action_query_batch_std": 0.3925072580575943,
      "action_query_norm": 43.07596969604492,
      "action_relation_loss": 0.00450907894410193,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.114125350795135,
      "flow_matching_loss": 0.032759391330182554,
      "grad_norm": 0.12446650117635727,
      "h_stock_valid_norm": 97.2540657043457,
      "invalid_tail_samples": 21.7,
      "learning_rate": 6.519377685347807e-05,
      "loss": 0.0373,
      "positive_teacher_pairs": 27340.1,
      "query_h_norm_ratio": 0.4429262697696686,
      "relation_loss": 0.00450907894410193,
      "selected_pairs": 27340.1,
      "step": 12950,
      "student_distance_mean": 26.18620147705078,
      "teacher_distance_mean": 0.8601427674293518,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.00450907894410193,
      "weighted_relation_loss": 0.00450907894410193
    },
    {
      "action_query_batch_std": 0.3839141815900803,
      "action_query_norm": 43.030907821655276,
      "action_relation_loss": 0.004887546133249998,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.123479887745557,
      "flow_matching_loss": 0.032861160859465596,
      "grad_norm": 0.10435204952955246,
      "h_stock_valid_norm": 97.52993087768554,
      "invalid_tail_samples": 26.4,
      "learning_rate": 6.514125829632021e-05,
      "loss": 0.0377,
      "positive_teacher_pairs": 26253.3,
      "query_h_norm_ratio": 0.44121108651161195,
      "relation_loss": 0.004887546133249998,
      "selected_pairs": 26253.3,
      "step": 12960,
      "student_distance_mean": 25.50841236114502,
      "teacher_distance_mean": 0.8406302034854889,
      "valid_action_samples": 229.6,
      "weighted_auxiliary_loss": 0.004887546133249998,
      "weighted_relation_loss": 0.004887546133249998
    },
    {
      "action_query_batch_std": 0.39087808430194854,
      "action_query_norm": 43.0548641204834,
      "action_relation_loss": 0.0043486255221068856,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.132834424695977,
      "flow_matching_loss": 0.03335586115717888,
      "grad_norm": 0.1255587935447693,
      "h_stock_valid_norm": 98.17084121704102,
      "invalid_tail_samples": 24.3,
      "learning_rate": 6.508872134110577e-05,
      "loss": 0.0377,
      "positive_teacher_pairs": 26734.1,
      "query_h_norm_ratio": 0.4385747253894806,
      "relation_loss": 0.0043486255221068856,
      "selected_pairs": 26734.1,
      "step": 12970,
      "student_distance_mean": 25.85351676940918,
      "teacher_distance_mean": 0.8559429943561554,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0043486255221068856,
      "weighted_relation_loss": 0.0043486255221068856
    },
    {
      "action_query_batch_std": 0.3792172521352768,
      "action_query_norm": 43.00577011108398,
      "action_relation_loss": 0.004445897461846471,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.142188961646399,
      "flow_matching_loss": 0.034247572720050815,
      "grad_norm": 0.13777199387550354,
      "h_stock_valid_norm": 98.15150833129883,
      "invalid_tail_samples": 22.0,
      "learning_rate": 6.503616605167213e-05,
      "loss": 0.0387,
      "positive_teacher_pairs": 27276.7,
      "query_h_norm_ratio": 0.4381611704826355,
      "relation_loss": 0.004445897461846471,
      "selected_pairs": 27276.7,
      "step": 12980,
      "student_distance_mean": 25.21569538116455,
      "teacher_distance_mean": 0.8398965895175934,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.004445897461846471,
      "weighted_relation_loss": 0.004445897461846471
    },
    {
      "action_query_batch_std": 0.3913216024637222,
      "action_query_norm": 43.04832038879395,
      "action_relation_loss": 0.0045460777590051295,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.15154349859682,
      "flow_matching_loss": 0.03291612397879362,
      "grad_norm": 0.08958608657121658,
      "h_stock_valid_norm": 98.15847396850586,
      "invalid_tail_samples": 23.3,
      "learning_rate": 6.498359249187893e-05,
      "loss": 0.0375,
      "positive_teacher_pairs": 26964.9,
      "query_h_norm_ratio": 0.43856562077999117,
      "relation_loss": 0.0045460777590051295,
      "selected_pairs": 26964.9,
      "step": 12990,
      "student_distance_mean": 26.062273597717287,
      "teacher_distance_mean": 0.8635129392147064,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0045460777590051295,
      "weighted_relation_loss": 0.0045460777590051295
    },
    {
      "action_query_batch_std": 0.38660354912281036,
      "action_query_norm": 43.03868789672852,
      "action_relation_loss": 0.004461273737251759,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.160898035547241,
      "flow_matching_loss": 0.0330227468162775,
      "grad_norm": 0.10549329221248627,
      "h_stock_valid_norm": 97.8126121520996,
      "invalid_tail_samples": 23.0,
      "learning_rate": 6.493100072560799e-05,
      "loss": 0.0375,
      "positive_teacher_pairs": 27036.2,
      "query_h_norm_ratio": 0.44001637399196625,
      "relation_loss": 0.004461273737251759,
      "selected_pairs": 27036.2,
      "step": 13000,
      "student_distance_mean": 25.757325744628908,
      "teacher_distance_mean": 0.8504546046257019,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.004461273737251759,
      "weighted_relation_loss": 0.004461273737251759
    },
    {
      "action_query_batch_std": 0.3803598076105118,
      "action_query_norm": 43.0117488861084,
      "action_relation_loss": 0.004756215959787369,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.170252572497661,
      "flow_matching_loss": 0.032740050181746486,
      "grad_norm": 0.08833510428667068,
      "h_stock_valid_norm": 97.96546325683593,
      "invalid_tail_samples": 21.8,
      "learning_rate": 6.487839081676327e-05,
      "loss": 0.0375,
      "positive_teacher_pairs": 27311.5,
      "query_h_norm_ratio": 0.4390546590089798,
      "relation_loss": 0.004756215959787369,
      "selected_pairs": 27311.5,
      "step": 13010,
      "student_distance_mean": 25.33536376953125,
      "teacher_distance_mean": 0.8400427341461182,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.004756215959787369,
      "weighted_relation_loss": 0.004756215959787369
    },
    {
      "action_query_batch_std": 0.38572847843170166,
      "action_query_norm": 43.03151245117188,
      "action_relation_loss": 0.004663897585123778,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.179607109448082,
      "flow_matching_loss": 0.03300885893404484,
      "grad_norm": 0.11390700191259384,
      "h_stock_valid_norm": 98.04907608032227,
      "invalid_tail_samples": 25.4,
      "learning_rate": 6.482576282927076e-05,
      "loss": 0.0377,
      "positive_teacher_pairs": 26485.3,
      "query_h_norm_ratio": 0.4388818234205246,
      "relation_loss": 0.004663897585123778,
      "selected_pairs": 26485.3,
      "step": 13020,
      "student_distance_mean": 25.69640693664551,
      "teacher_distance_mean": 0.8485021829605103,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.004663897585123778,
      "weighted_relation_loss": 0.004663897585123778
    },
    {
      "action_query_batch_std": 0.3846651613712311,
      "action_query_norm": 43.01998519897461,
      "action_relation_loss": 0.004340130882337689,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.188961646398504,
      "flow_matching_loss": 0.03220382798463106,
      "grad_norm": 0.1154744103550911,
      "h_stock_valid_norm": 97.01491241455078,
      "invalid_tail_samples": 24.1,
      "learning_rate": 6.477311682707845e-05,
      "loss": 0.0365,
      "positive_teacher_pairs": 26778.8,
      "query_h_norm_ratio": 0.44344071447849276,
      "relation_loss": 0.004340130882337689,
      "selected_pairs": 26778.8,
      "step": 13030,
      "student_distance_mean": 25.679644203186037,
      "teacher_distance_mean": 0.8520699143409729,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.004340130882337689,
      "weighted_relation_loss": 0.004340130882337689
    },
    {
      "action_query_batch_std": 0.38441345691680906,
      "action_query_norm": 43.00221328735351,
      "action_relation_loss": 0.004305492760613561,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.198316183348924,
      "flow_matching_loss": 0.0340916445478797,
      "grad_norm": 0.11642374098300934,
      "h_stock_valid_norm": 96.95690612792968,
      "invalid_tail_samples": 24.2,
      "learning_rate": 6.472045287415616e-05,
      "loss": 0.0384,
      "positive_teacher_pairs": 26755.5,
      "query_h_norm_ratio": 0.4435215026140213,
      "relation_loss": 0.004305492760613561,
      "selected_pairs": 26755.5,
      "step": 13040,
      "student_distance_mean": 25.620769500732422,
      "teacher_distance_mean": 0.851539534330368,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.004305492760613561,
      "weighted_relation_loss": 0.004305492760613561
    },
    {
      "action_query_batch_std": 0.38695296347141267,
      "action_query_norm": 43.02407035827637,
      "action_relation_loss": 0.004349979246035219,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.207670720299346,
      "flow_matching_loss": 0.03197932131588459,
      "grad_norm": 0.15327124297618866,
      "h_stock_valid_norm": 97.39481811523437,
      "invalid_tail_samples": 22.3,
      "learning_rate": 6.466777103449558e-05,
      "loss": 0.0363,
      "positive_teacher_pairs": 27207.5,
      "query_h_norm_ratio": 0.44175481498241426,
      "relation_loss": 0.004349979246035219,
      "selected_pairs": 27207.5,
      "step": 13050,
      "student_distance_mean": 25.62776279449463,
      "teacher_distance_mean": 0.8498904824256897,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.004349979246035219,
      "weighted_relation_loss": 0.004349979246035219
    },
    {
      "action_query_batch_std": 0.3912080556154251,
      "action_query_norm": 43.0430549621582,
      "action_relation_loss": 0.004577582236379385,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.217025257249766,
      "flow_matching_loss": 0.03301437012851238,
      "grad_norm": 0.13155899941921234,
      "h_stock_valid_norm": 97.98351287841797,
      "invalid_tail_samples": 22.4,
      "learning_rate": 6.461507137211012e-05,
      "loss": 0.0376,
      "positive_teacher_pairs": 27175.2,
      "query_h_norm_ratio": 0.4392917811870575,
      "relation_loss": 0.004577582236379385,
      "selected_pairs": 27175.2,
      "step": 13060,
      "student_distance_mean": 26.008709335327147,
      "teacher_distance_mean": 0.8623003304004669,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.004577582236379385,
      "weighted_relation_loss": 0.004577582236379385
    },
    {
      "action_query_batch_std": 0.390797883272171,
      "action_query_norm": 43.03801307678223,
      "action_relation_loss": 0.0043219010112807155,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.226379794200188,
      "flow_matching_loss": 0.03370568137615919,
      "grad_norm": 0.1330965906381607,
      "h_stock_valid_norm": 98.13870849609376,
      "invalid_tail_samples": 22.1,
      "learning_rate": 6.456235395103483e-05,
      "loss": 0.038,
      "positive_teacher_pairs": 27244.8,
      "query_h_norm_ratio": 0.43854392766952516,
      "relation_loss": 0.0043219010112807155,
      "selected_pairs": 27244.8,
      "step": 13070,
      "student_distance_mean": 25.908281135559083,
      "teacher_distance_mean": 0.8652443170547486,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0043219010112807155,
      "weighted_relation_loss": 0.0043219010112807155
    },
    {
      "action_query_batch_std": 0.38103427588939665,
      "action_query_norm": 42.99397315979004,
      "action_relation_loss": 0.004437827970832586,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.235734331150608,
      "flow_matching_loss": 0.03257111310958862,
      "grad_norm": 0.11763652414083481,
      "h_stock_valid_norm": 98.58574295043945,
      "invalid_tail_samples": 24.7,
      "learning_rate": 6.450961883532634e-05,
      "loss": 0.037,
      "positive_teacher_pairs": 26650.8,
      "query_h_norm_ratio": 0.4361120820045471,
      "relation_loss": 0.004437827970832586,
      "selected_pairs": 26650.8,
      "step": 13080,
      "student_distance_mean": 25.400679779052734,
      "teacher_distance_mean": 0.8446080625057221,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.004437827970832586,
      "weighted_relation_loss": 0.004437827970832586
    },
    {
      "action_query_batch_std": 0.3871916264295578,
      "action_query_norm": 43.00908584594727,
      "action_relation_loss": 0.0044966421090066435,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.245088868101028,
      "flow_matching_loss": 0.03406145162880421,
      "grad_norm": 0.1487172394990921,
      "h_stock_valid_norm": 98.61822128295898,
      "invalid_tail_samples": 26.1,
      "learning_rate": 6.445686608906283e-05,
      "loss": 0.0386,
      "positive_teacher_pairs": 26325.9,
      "query_h_norm_ratio": 0.43612011671066286,
      "relation_loss": 0.0044966421090066435,
      "selected_pairs": 26325.9,
      "step": 13090,
      "student_distance_mean": 25.927468681335448,
      "teacher_distance_mean": 0.8583576202392578,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.0044966421090066435,
      "weighted_relation_loss": 0.0044966421090066435
    },
    {
      "action_query_batch_std": 0.38057086169719695,
      "action_query_norm": 42.995185470581056,
      "action_relation_loss": 0.004541440587490797,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.25444340505145,
      "flow_matching_loss": 0.034050907380878924,
      "grad_norm": 0.1357424408197403,
      "h_stock_valid_norm": 98.09894104003907,
      "invalid_tail_samples": 23.0,
      "learning_rate": 6.44040957763438e-05,
      "loss": 0.0386,
      "positive_teacher_pairs": 27033.9,
      "query_h_norm_ratio": 0.43828721940517423,
      "relation_loss": 0.004541440587490797,
      "selected_pairs": 27033.9,
      "step": 13100,
      "student_distance_mean": 25.262907028198242,
      "teacher_distance_mean": 0.8408303737640381,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.004541440587490797,
      "weighted_relation_loss": 0.004541440587490797
    },
    {
      "action_query_batch_std": 0.3830608129501343,
      "action_query_norm": 43.01238174438477,
      "action_relation_loss": 0.004573693964630365,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.26379794200187,
      "flow_matching_loss": 0.035525650903582576,
      "grad_norm": 0.1776445508003235,
      "h_stock_valid_norm": 97.2274673461914,
      "invalid_tail_samples": 25.6,
      "learning_rate": 6.435130796129018e-05,
      "loss": 0.0401,
      "positive_teacher_pairs": 26436.6,
      "query_h_norm_ratio": 0.4424062520265579,
      "relation_loss": 0.004573693964630365,
      "selected_pairs": 26436.6,
      "step": 13110,
      "student_distance_mean": 25.630516624450685,
      "teacher_distance_mean": 0.8560683727264404,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.004573693964630365,
      "weighted_relation_loss": 0.004573693964630365
    },
    {
      "action_query_batch_std": 0.38805329501628877,
      "action_query_norm": 43.02774505615234,
      "action_relation_loss": 0.004458918632008135,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.273152478952293,
      "flow_matching_loss": 0.03572888020426035,
      "grad_norm": 0.11633066833019257,
      "h_stock_valid_norm": 97.1144416809082,
      "invalid_tail_samples": 25.3,
      "learning_rate": 6.429850270804416e-05,
      "loss": 0.0402,
      "positive_teacher_pairs": 26513.1,
      "query_h_norm_ratio": 0.4430682212114334,
      "relation_loss": 0.004458918632008135,
      "selected_pairs": 26513.1,
      "step": 13120,
      "student_distance_mean": 25.889844512939455,
      "teacher_distance_mean": 0.8559604167938233,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.004458918632008135,
      "weighted_relation_loss": 0.004458918632008135
    },
    {
      "action_query_batch_std": 0.39110131859779357,
      "action_query_norm": 43.027306365966794,
      "action_relation_loss": 0.0043465138878673315,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.282507015902713,
      "flow_matching_loss": 0.0349451769143343,
      "grad_norm": 0.11656657606363297,
      "h_stock_valid_norm": 97.96557159423828,
      "invalid_tail_samples": 22.7,
      "learning_rate": 6.424568008076909e-05,
      "loss": 0.0393,
      "positive_teacher_pairs": 27102.7,
      "query_h_norm_ratio": 0.43921553492546084,
      "relation_loss": 0.0043465138878673315,
      "selected_pairs": 27102.7,
      "step": 13130,
      "student_distance_mean": 26.031383895874022,
      "teacher_distance_mean": 0.8612931966781616,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0043465138878673315,
      "weighted_relation_loss": 0.0043465138878673315
    },
    {
      "action_query_batch_std": 0.3880890399217606,
      "action_query_norm": 43.01159591674805,
      "action_relation_loss": 0.0043630422558635475,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.291861552853133,
      "flow_matching_loss": 0.0327279219403863,
      "grad_norm": 0.13050112128257751,
      "h_stock_valid_norm": 98.3382682800293,
      "invalid_tail_samples": 24.1,
      "learning_rate": 6.419284014364943e-05,
      "loss": 0.0371,
      "positive_teacher_pairs": 26779.5,
      "query_h_norm_ratio": 0.43738765716552735,
      "relation_loss": 0.0043630422558635475,
      "selected_pairs": 26779.5,
      "step": 13140,
      "student_distance_mean": 25.75140495300293,
      "teacher_distance_mean": 0.8564020574092865,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0043630422558635475,
      "weighted_relation_loss": 0.0043630422558635475
    },
    {
      "action_query_batch_std": 0.3911836683750153,
      "action_query_norm": 43.022307586669925,
      "action_relation_loss": 0.004523788904771208,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.301216089803555,
      "flow_matching_loss": 0.03378591798245907,
      "grad_norm": 0.11085306853055954,
      "h_stock_valid_norm": 98.08989028930664,
      "invalid_tail_samples": 24.2,
      "learning_rate": 6.41399829608907e-05,
      "loss": 0.0383,
      "positive_teacher_pairs": 26763.6,
      "query_h_norm_ratio": 0.43860720098018646,
      "relation_loss": 0.004523788904771208,
      "selected_pairs": 26763.6,
      "step": 13150,
      "student_distance_mean": 25.99953269958496,
      "teacher_distance_mean": 0.8663534820079803,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.004523788904771208,
      "weighted_relation_loss": 0.004523788904771208
    },
    {
      "action_query_batch_std": 0.3891300082206726,
      "action_query_norm": 43.01082572937012,
      "action_relation_loss": 0.004536872962489724,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.310570626753975,
      "flow_matching_loss": 0.03439366910606623,
      "grad_norm": 0.1098749190568924,
      "h_stock_valid_norm": 98.46726150512696,
      "invalid_tail_samples": 21.5,
      "learning_rate": 6.408710859671938e-05,
      "loss": 0.0389,
      "positive_teacher_pairs": 27382.1,
      "query_h_norm_ratio": 0.4368066281080246,
      "relation_loss": 0.004536872962489724,
      "selected_pairs": 27382.1,
      "step": 13160,
      "student_distance_mean": 25.94047451019287,
      "teacher_distance_mean": 0.8547034144401551,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.004536872962489724,
      "weighted_relation_loss": 0.004536872962489724
    },
    {
      "action_query_batch_std": 0.3927482932806015,
      "action_query_norm": 43.02109909057617,
      "action_relation_loss": 0.0046305210329592224,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.319925163704397,
      "flow_matching_loss": 0.03547908905893564,
      "grad_norm": 0.12499572336673737,
      "h_stock_valid_norm": 98.1554786682129,
      "invalid_tail_samples": 23.0,
      "learning_rate": 6.403421711538278e-05,
      "loss": 0.0401,
      "positive_teacher_pairs": 27035.4,
      "query_h_norm_ratio": 0.438297364115715,
      "relation_loss": 0.0046305210329592224,
      "selected_pairs": 27035.4,
      "step": 13170,
      "student_distance_mean": 26.030504608154295,
      "teacher_distance_mean": 0.8515196740627289,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0046305210329592224,
      "weighted_relation_loss": 0.0046305210329592224
    },
    {
      "action_query_batch_std": 0.38722616732120513,
      "action_query_norm": 43.00313758850098,
      "action_relation_loss": 0.004525921819731593,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.329279700654817,
      "flow_matching_loss": 0.03718099892139435,
      "grad_norm": 0.1123390942811966,
      "h_stock_valid_norm": 98.14929428100587,
      "invalid_tail_samples": 23.5,
      "learning_rate": 6.398130858114903e-05,
      "loss": 0.0417,
      "positive_teacher_pairs": 26919.8,
      "query_h_norm_ratio": 0.4381419986486435,
      "relation_loss": 0.004525921819731593,
      "selected_pairs": 26919.8,
      "step": 13180,
      "student_distance_mean": 25.88489360809326,
      "teacher_distance_mean": 0.8583630204200745,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.004525921819731593,
      "weighted_relation_loss": 0.004525921819731593
    },
    {
      "action_query_batch_std": 0.3911367803812027,
      "action_query_norm": 43.01832580566406,
      "action_relation_loss": 0.00456056222319603,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.33863423760524,
      "flow_matching_loss": 0.0340559670701623,
      "grad_norm": 0.10407084226608276,
      "h_stock_valid_norm": 97.70042877197265,
      "invalid_tail_samples": 23.9,
      "learning_rate": 6.392838305830701e-05,
      "loss": 0.0386,
      "positive_teacher_pairs": 26835.9,
      "query_h_norm_ratio": 0.44031154215335844,
      "relation_loss": 0.00456056222319603,
      "selected_pairs": 26835.9,
      "step": 13190,
      "student_distance_mean": 26.015539360046386,
      "teacher_distance_mean": 0.8657506465911865,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.00456056222319603,
      "weighted_relation_loss": 0.00456056222319603
    },
    {
      "action_query_batch_std": 0.38674943447113036,
      "action_query_norm": 42.99322319030762,
      "action_relation_loss": 0.0047344652935862545,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.34798877455566,
      "flow_matching_loss": 0.03331939335912466,
      "grad_norm": 0.11455553770065308,
      "h_stock_valid_norm": 97.6848258972168,
      "invalid_tail_samples": 23.6,
      "learning_rate": 6.387544061116621e-05,
      "loss": 0.0381,
      "positive_teacher_pairs": 26901.6,
      "query_h_norm_ratio": 0.4401254773139954,
      "relation_loss": 0.0047344652935862545,
      "selected_pairs": 26901.6,
      "step": 13200,
      "student_distance_mean": 25.71343078613281,
      "teacher_distance_mean": 0.8476172685623169,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0047344652935862545,
      "weighted_relation_loss": 0.0047344652935862545
    },
    {
      "action_query_batch_std": 0.3826545476913452,
      "action_query_norm": 42.96904182434082,
      "action_relation_loss": 0.004422352695837617,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.35734331150608,
      "flow_matching_loss": 0.03256053663790226,
      "grad_norm": 0.12375988811254501,
      "h_stock_valid_norm": 97.53464126586914,
      "invalid_tail_samples": 24.1,
      "learning_rate": 6.382248130405671e-05,
      "loss": 0.037,
      "positive_teacher_pairs": 26781.2,
      "query_h_norm_ratio": 0.4405532658100128,
      "relation_loss": 0.004422352695837617,
      "selected_pairs": 26781.2,
      "step": 13210,
      "student_distance_mean": 25.575059127807616,
      "teacher_distance_mean": 0.8472536385059357,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.004422352695837617,
      "weighted_relation_loss": 0.004422352695837617
    },
    {
      "action_query_batch_std": 0.3856045871973038,
      "action_query_norm": 42.97628479003906,
      "action_relation_loss": 0.00463278426323086,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.366697848456502,
      "flow_matching_loss": 0.03335020057857037,
      "grad_norm": 0.11164511740207672,
      "h_stock_valid_norm": 97.45890579223632,
      "invalid_tail_samples": 27.3,
      "learning_rate": 6.376950520132905e-05,
      "loss": 0.038,
      "positive_teacher_pairs": 26045.0,
      "query_h_norm_ratio": 0.44097378849983215,
      "relation_loss": 0.00463278426323086,
      "selected_pairs": 26045.0,
      "step": 13220,
      "student_distance_mean": 25.63389663696289,
      "teacher_distance_mean": 0.8525663673877716,
      "valid_action_samples": 228.7,
      "weighted_auxiliary_loss": 0.00463278426323086,
      "weighted_relation_loss": 0.00463278426323086
    },
    {
      "action_query_batch_std": 0.38868297934532164,
      "action_query_norm": 42.987925338745114,
      "action_relation_loss": 0.0043122201226651665,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.376052385406922,
      "flow_matching_loss": 0.03318491019308567,
      "grad_norm": 0.10902968049049377,
      "h_stock_valid_norm": 97.13631591796874,
      "invalid_tail_samples": 21.3,
      "learning_rate": 6.371651236735418e-05,
      "loss": 0.0375,
      "positive_teacher_pairs": 27433.1,
      "query_h_norm_ratio": 0.442556431889534,
      "relation_loss": 0.0043122201226651665,
      "selected_pairs": 27433.1,
      "step": 13230,
      "student_distance_mean": 25.944349479675292,
      "teacher_distance_mean": 0.8561501860618591,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.0043122201226651665,
      "weighted_relation_loss": 0.0043122201226651665
    },
    {
      "action_query_batch_std": 0.393282887339592,
      "action_query_norm": 42.99527931213379,
      "action_relation_loss": 0.004235824034549296,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.385406922357344,
      "flow_matching_loss": 0.0325489591807127,
      "grad_norm": 0.10520252585411072,
      "h_stock_valid_norm": 97.26089935302734,
      "invalid_tail_samples": 27.0,
      "learning_rate": 6.366350286652341e-05,
      "loss": 0.0368,
      "positive_teacher_pairs": 26117.2,
      "query_h_norm_ratio": 0.44206740856170657,
      "relation_loss": 0.004235824034549296,
      "selected_pairs": 26117.2,
      "step": 13240,
      "student_distance_mean": 26.003007316589354,
      "teacher_distance_mean": 0.8596866190433502,
      "valid_action_samples": 229.0,
      "weighted_auxiliary_loss": 0.004235824034549296,
      "weighted_relation_loss": 0.004235824034549296
    },
    {
      "action_query_batch_std": 0.3908356994390488,
      "action_query_norm": 42.99690933227539,
      "action_relation_loss": 0.004588130209594965,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.394761459307764,
      "flow_matching_loss": 0.034267901070415975,
      "grad_norm": 0.10083898901939392,
      "h_stock_valid_norm": 97.48456573486328,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.361047676324827e-05,
      "loss": 0.0389,
      "positive_teacher_pairs": 26666.4,
      "query_h_norm_ratio": 0.44106648564338685,
      "relation_loss": 0.004588130209594965,
      "selected_pairs": 26666.4,
      "step": 13250,
      "student_distance_mean": 25.969461822509764,
      "teacher_distance_mean": 0.8610926687717437,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.004588130209594965,
      "weighted_relation_loss": 0.004588130209594965
    },
    {
      "action_query_batch_std": 0.38900211453437805,
      "action_query_norm": 42.96649932861328,
      "action_relation_loss": 0.004475840833038092,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.404115996258184,
      "flow_matching_loss": 0.03212092891335487,
      "grad_norm": 0.12292687594890594,
      "h_stock_valid_norm": 97.72400207519532,
      "invalid_tail_samples": 25.5,
      "learning_rate": 6.355743412196047e-05,
      "loss": 0.0366,
      "positive_teacher_pairs": 26467.0,
      "query_h_norm_ratio": 0.43967601358890535,
      "relation_loss": 0.004475840833038092,
      "selected_pairs": 26467.0,
      "step": 13260,
      "student_distance_mean": 25.923002433776855,
      "teacher_distance_mean": 0.8642558217048645,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.004475840833038092,
      "weighted_relation_loss": 0.004475840833038092
    },
    {
      "action_query_batch_std": 0.3858731657266617,
      "action_query_norm": 42.96665573120117,
      "action_relation_loss": 0.004400137066841126,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.413470533208606,
      "flow_matching_loss": 0.03377783428877592,
      "grad_norm": 0.10993053764104843,
      "h_stock_valid_norm": 97.69271774291992,
      "invalid_tail_samples": 25.1,
      "learning_rate": 6.350437500711184e-05,
      "loss": 0.0382,
      "positive_teacher_pairs": 26553.9,
      "query_h_norm_ratio": 0.43982061743736267,
      "relation_loss": 0.004400137066841126,
      "selected_pairs": 26553.9,
      "step": 13270,
      "student_distance_mean": 25.663371658325197,
      "teacher_distance_mean": 0.8589188873767852,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.004400137066841126,
      "weighted_relation_loss": 0.004400137066841126
    },
    {
      "action_query_batch_std": 0.3836019843816757,
      "action_query_norm": 42.95604972839355,
      "action_relation_loss": 0.00468595540151,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.422825070159027,
      "flow_matching_loss": 0.03205263800919056,
      "grad_norm": 0.09471907466650009,
      "h_stock_valid_norm": 97.45218505859376,
      "invalid_tail_samples": 23.0,
      "learning_rate": 6.345129948317418e-05,
      "loss": 0.0367,
      "positive_teacher_pairs": 27041.1,
      "query_h_norm_ratio": 0.4407923072576523,
      "relation_loss": 0.00468595540151,
      "selected_pairs": 27041.1,
      "step": 13280,
      "student_distance_mean": 25.40033302307129,
      "teacher_distance_mean": 0.852505213022232,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.00468595540151,
      "weighted_relation_loss": 0.00468595540151
    },
    {
      "action_query_batch_std": 0.3907670140266418,
      "action_query_norm": 42.986202239990234,
      "action_relation_loss": 0.0044198079034686085,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.432179607109449,
      "flow_matching_loss": 0.032668995670974255,
      "grad_norm": 0.1099792867898941,
      "h_stock_valid_norm": 97.98518829345703,
      "invalid_tail_samples": 25.1,
      "learning_rate": 6.33982076146393e-05,
      "loss": 0.0371,
      "positive_teacher_pairs": 26554.9,
      "query_h_norm_ratio": 0.43870455026626587,
      "relation_loss": 0.0044198079034686085,
      "selected_pairs": 26554.9,
      "step": 13290,
      "student_distance_mean": 26.08322696685791,
      "teacher_distance_mean": 0.8592599868774414,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0044198079034686085,
      "weighted_relation_loss": 0.0044198079034686085
    },
    {
      "action_query_batch_std": 0.38849411606788636,
      "action_query_norm": 42.967716979980466,
      "action_relation_loss": 0.004405098967254162,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.441534144059869,
      "flow_matching_loss": 0.0329485023394227,
      "grad_norm": 0.10513227432966232,
      "h_stock_valid_norm": 98.12916946411133,
      "invalid_tail_samples": 23.0,
      "learning_rate": 6.334509946601879e-05,
      "loss": 0.0374,
      "positive_teacher_pairs": 27034.5,
      "query_h_norm_ratio": 0.4378725677728653,
      "relation_loss": 0.004405098967254162,
      "selected_pairs": 27034.5,
      "step": 13300,
      "student_distance_mean": 25.7216459274292,
      "teacher_distance_mean": 0.8551376640796662,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.004405098967254162,
      "weighted_relation_loss": 0.004405098967254162
    },
    {
      "action_query_batch_std": 0.38309510946273806,
      "action_query_norm": 42.943318939208986,
      "action_relation_loss": 0.004467064514756203,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.45088868101029,
      "flow_matching_loss": 0.03396035600453615,
      "grad_norm": 0.10643676668405533,
      "h_stock_valid_norm": 97.69378433227538,
      "invalid_tail_samples": 24.5,
      "learning_rate": 6.329197510184406e-05,
      "loss": 0.0384,
      "positive_teacher_pairs": 26698.5,
      "query_h_norm_ratio": 0.439578241109848,
      "relation_loss": 0.004467064514756203,
      "selected_pairs": 26698.5,
      "step": 13310,
      "student_distance_mean": 25.4277042388916,
      "teacher_distance_mean": 0.8459551870822907,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.004467064514756203,
      "weighted_relation_loss": 0.004467064514756203
    },
    {
      "action_query_batch_std": 0.38889231979846955,
      "action_query_norm": 42.974788665771484,
      "action_relation_loss": 0.004375076410360635,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.460243217960711,
      "flow_matching_loss": 0.034895900078117846,
      "grad_norm": 0.10022526979446411,
      "h_stock_valid_norm": 96.86330947875976,
      "invalid_tail_samples": 25.6,
      "learning_rate": 6.323883458666624e-05,
      "loss": 0.0393,
      "positive_teacher_pairs": 26439.3,
      "query_h_norm_ratio": 0.44367403388023374,
      "relation_loss": 0.004375076410360635,
      "selected_pairs": 26439.3,
      "step": 13320,
      "student_distance_mean": 25.893926048278807,
      "teacher_distance_mean": 0.8561292827129364,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.004375076410360635,
      "weighted_relation_loss": 0.004375076410360635
    },
    {
      "action_query_batch_std": 0.38682118952274325,
      "action_query_norm": 42.97469635009766,
      "action_relation_loss": 0.004359534103423357,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.469597754911131,
      "flow_matching_loss": 0.032985628582537174,
      "grad_norm": 0.09055539965629578,
      "h_stock_valid_norm": 96.82872161865234,
      "invalid_tail_samples": 24.5,
      "learning_rate": 6.318567798505605e-05,
      "loss": 0.0373,
      "positive_teacher_pairs": 26685.4,
      "query_h_norm_ratio": 0.4438256651163101,
      "relation_loss": 0.004359534103423357,
      "selected_pairs": 26685.4,
      "step": 13330,
      "student_distance_mean": 25.738562202453615,
      "teacher_distance_mean": 0.8550348818302155,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.004359534103423357,
      "weighted_relation_loss": 0.004359534103423357
    },
    {
      "action_query_batch_std": 0.39314462542533873,
      "action_query_norm": 42.980991744995116,
      "action_relation_loss": 0.00449895157944411,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.478952291861553,
      "flow_matching_loss": 0.03560729045420885,
      "grad_norm": 0.13003228604793549,
      "h_stock_valid_norm": 97.93388900756835,
      "invalid_tail_samples": 24.8,
      "learning_rate": 6.313250536160378e-05,
      "loss": 0.0401,
      "positive_teacher_pairs": 26621.0,
      "query_h_norm_ratio": 0.4388866305351257,
      "relation_loss": 0.00449895157944411,
      "selected_pairs": 26621.0,
      "step": 13340,
      "student_distance_mean": 26.01904354095459,
      "teacher_distance_mean": 0.8640237271785736,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.00449895157944411,
      "weighted_relation_loss": 0.00449895157944411
    },
    {
      "action_query_batch_std": 0.3874568849802017,
      "action_query_norm": 42.96033401489258,
      "action_relation_loss": 0.004248652164824307,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.488306828811973,
      "flow_matching_loss": 0.03452477604150772,
      "grad_norm": 0.1240900382399559,
      "h_stock_valid_norm": 98.39599838256837,
      "invalid_tail_samples": 22.4,
      "learning_rate": 6.307931678091917e-05,
      "loss": 0.0388,
      "positive_teacher_pairs": 27183.3,
      "query_h_norm_ratio": 0.4366099268198013,
      "relation_loss": 0.004248652164824307,
      "selected_pairs": 27183.3,
      "step": 13350,
      "student_distance_mean": 25.812343788146972,
      "teacher_distance_mean": 0.859494811296463,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.004248652164824307,
      "weighted_relation_loss": 0.004248652164824307
    },
    {
      "action_query_batch_std": 0.3780455976724625,
      "action_query_norm": 42.92089424133301,
      "action_relation_loss": 0.004530009557493031,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.497661365762395,
      "flow_matching_loss": 0.03291382379829884,
      "grad_norm": 0.10464265197515488,
      "h_stock_valid_norm": 98.10551071166992,
      "invalid_tail_samples": 23.7,
      "learning_rate": 6.302611230763138e-05,
      "loss": 0.0374,
      "positive_teacher_pairs": 26871.9,
      "query_h_norm_ratio": 0.43750016391277313,
      "relation_loss": 0.004530009557493031,
      "selected_pairs": 26871.9,
      "step": 13360,
      "student_distance_mean": 25.236062812805176,
      "teacher_distance_mean": 0.8424903929233551,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.004530009557493031,
      "weighted_relation_loss": 0.004530009557493031
    },
    {
      "action_query_batch_std": 0.38296292424201966,
      "action_query_norm": 42.953857421875,
      "action_relation_loss": 0.0043505854671821,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.507015902712816,
      "flow_matching_loss": 0.03323595654219389,
      "grad_norm": 0.11069197207689285,
      "h_stock_valid_norm": 97.93322677612305,
      "invalid_tail_samples": 23.4,
      "learning_rate": 6.297289200638887e-05,
      "loss": 0.0376,
      "positive_teacher_pairs": 26944.6,
      "query_h_norm_ratio": 0.4386082351207733,
      "relation_loss": 0.0043505854671821,
      "selected_pairs": 26944.6,
      "step": 13370,
      "student_distance_mean": 25.57488269805908,
      "teacher_distance_mean": 0.8555254220962525,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0043505854671821,
      "weighted_relation_loss": 0.0043505854671821
    },
    {
      "action_query_batch_std": 0.38620065748691557,
      "action_query_norm": 42.956189727783205,
      "action_relation_loss": 0.004479770339094103,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.516370439663238,
      "flow_matching_loss": 0.032406662590801716,
      "grad_norm": 0.11232072860002518,
      "h_stock_valid_norm": 97.79031219482422,
      "invalid_tail_samples": 21.6,
      "learning_rate": 6.29196559418593e-05,
      "loss": 0.0369,
      "positive_teacher_pairs": 27368.6,
      "query_h_norm_ratio": 0.439270481467247,
      "relation_loss": 0.004479770339094103,
      "selected_pairs": 27368.6,
      "step": 13380,
      "student_distance_mean": 25.820515632629395,
      "teacher_distance_mean": 0.855071097612381,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.004479770339094103,
      "weighted_relation_loss": 0.004479770339094103
    },
    {
      "action_query_batch_std": 0.38298891186714173,
      "action_query_norm": 42.935736083984374,
      "action_relation_loss": 0.0045884082093834875,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.525724976613658,
      "flow_matching_loss": 0.033396968990564345,
      "grad_norm": 0.11690757423639297,
      "h_stock_valid_norm": 97.66751251220703,
      "invalid_tail_samples": 24.2,
      "learning_rate": 6.286640417872951e-05,
      "loss": 0.038,
      "positive_teacher_pairs": 26754.9,
      "query_h_norm_ratio": 0.43961432874202727,
      "relation_loss": 0.0045884082093834875,
      "selected_pairs": 26754.9,
      "step": 13390,
      "student_distance_mean": 25.56835060119629,
      "teacher_distance_mean": 0.848039311170578,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0045884082093834875,
      "weighted_relation_loss": 0.0045884082093834875
    },
    {
      "action_query_batch_std": 0.38638463616371155,
      "action_query_norm": 42.947684860229494,
      "action_relation_loss": 0.004503429494798183,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.535079513564078,
      "flow_matching_loss": 0.03352229371666908,
      "grad_norm": 0.14091016352176666,
      "h_stock_valid_norm": 97.02054367065429,
      "invalid_tail_samples": 23.3,
      "learning_rate": 6.281313678170542e-05,
      "loss": 0.038,
      "positive_teacher_pairs": 26960.9,
      "query_h_norm_ratio": 0.44267250001430514,
      "relation_loss": 0.004503429494798183,
      "selected_pairs": 26960.9,
      "step": 13400,
      "student_distance_mean": 25.64460849761963,
      "teacher_distance_mean": 0.8513788700103759,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.004503429494798183,
      "weighted_relation_loss": 0.004503429494798183
    },
    {
      "action_query_batch_std": 0.38765900433063505,
      "action_query_norm": 42.943128204345705,
      "action_relation_loss": 0.004424836626276374,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.5444340505145,
      "flow_matching_loss": 0.03503690250217915,
      "grad_norm": 0.14972878992557526,
      "h_stock_valid_norm": 97.08434982299805,
      "invalid_tail_samples": 22.7,
      "learning_rate": 6.275985381551194e-05,
      "loss": 0.0395,
      "positive_teacher_pairs": 27107.3,
      "query_h_norm_ratio": 0.44233474135398865,
      "relation_loss": 0.004424836626276374,
      "selected_pairs": 27107.3,
      "step": 13410,
      "student_distance_mean": 25.728470611572266,
      "teacher_distance_mean": 0.8592681586742401,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.004424836626276374,
      "weighted_relation_loss": 0.004424836626276374
    },
    {
      "action_query_batch_std": 0.38490705788135526,
      "action_query_norm": 42.93362541198731,
      "action_relation_loss": 0.004409138695336878,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.55378858746492,
      "flow_matching_loss": 0.032989143021404745,
      "grad_norm": 0.1437787562608719,
      "h_stock_valid_norm": 96.74861450195313,
      "invalid_tail_samples": 23.6,
      "learning_rate": 6.270655534489292e-05,
      "loss": 0.0374,
      "positive_teacher_pairs": 26902.8,
      "query_h_norm_ratio": 0.4437690615653992,
      "relation_loss": 0.004409138695336878,
      "selected_pairs": 26902.8,
      "step": 13420,
      "student_distance_mean": 25.63631076812744,
      "teacher_distance_mean": 0.8535465657711029,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.004409138695336878,
      "weighted_relation_loss": 0.004409138695336878
    },
    {
      "action_query_batch_std": 0.3838429540395737,
      "action_query_norm": 42.943246078491214,
      "action_relation_loss": 0.004645770322531462,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.563143124415342,
      "flow_matching_loss": 0.035042164102196696,
      "grad_norm": 0.12630553543567657,
      "h_stock_valid_norm": 96.7281364440918,
      "invalid_tail_samples": 21.5,
      "learning_rate": 6.265324143461098e-05,
      "loss": 0.0397,
      "positive_teacher_pairs": 27383.8,
      "query_h_norm_ratio": 0.4439625471830368,
      "relation_loss": 0.004645770322531462,
      "selected_pairs": 27383.8,
      "step": 13430,
      "student_distance_mean": 25.573924255371093,
      "teacher_distance_mean": 0.8558097302913665,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.004645770322531462,
      "weighted_relation_loss": 0.004645770322531462
    },
    {
      "action_query_batch_std": 0.3853984475135803,
      "action_query_norm": 42.92519645690918,
      "action_relation_loss": 0.004371100012212992,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.572497661365762,
      "flow_matching_loss": 0.033795030787587166,
      "grad_norm": 0.11958993226289749,
      "h_stock_valid_norm": 97.49907836914062,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.259991214944758e-05,
      "loss": 0.0382,
      "positive_teacher_pairs": 26664.5,
      "query_h_norm_ratio": 0.4402684956789017,
      "relation_loss": 0.004371100012212992,
      "selected_pairs": 26664.5,
      "step": 13440,
      "student_distance_mean": 25.62693862915039,
      "teacher_distance_mean": 0.8584349989891052,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.004371100012212992,
      "weighted_relation_loss": 0.004371100012212992
    },
    {
      "action_query_batch_std": 0.38077654838562014,
      "action_query_norm": 42.91150016784668,
      "action_relation_loss": 0.004845098033547402,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.581852198316183,
      "flow_matching_loss": 0.03314705453813076,
      "grad_norm": 0.14142979681491852,
      "h_stock_valid_norm": 97.82489700317383,
      "invalid_tail_samples": 24.5,
      "learning_rate": 6.254656755420283e-05,
      "loss": 0.038,
      "positive_teacher_pairs": 26691.5,
      "query_h_norm_ratio": 0.43865928053855896,
      "relation_loss": 0.004845098033547402,
      "selected_pairs": 26691.5,
      "step": 13450,
      "student_distance_mean": 25.281015396118164,
      "teacher_distance_mean": 0.8460691034793854,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.004845098033547402,
      "weighted_relation_loss": 0.004845098033547402
    },
    {
      "action_query_batch_std": 0.3879021376371384,
      "action_query_norm": 42.93889389038086,
      "action_relation_loss": 0.0044189922511577604,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.591206735266605,
      "flow_matching_loss": 0.034366970881819725,
      "grad_norm": 0.11948257684707642,
      "h_stock_valid_norm": 97.58707580566406,
      "invalid_tail_samples": 23.3,
      "learning_rate": 6.249320771369545e-05,
      "loss": 0.0388,
      "positive_teacher_pairs": 26973.5,
      "query_h_norm_ratio": 0.440012189745903,
      "relation_loss": 0.0044189922511577604,
      "selected_pairs": 26973.5,
      "step": 13460,
      "student_distance_mean": 25.708088302612303,
      "teacher_distance_mean": 0.8615334749221801,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0044189922511577604,
      "weighted_relation_loss": 0.0044189922511577604
    },
    {
      "action_query_batch_std": 0.38945446312427523,
      "action_query_norm": 42.93208274841309,
      "action_relation_loss": 0.004329662700183689,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.600561272217025,
      "flow_matching_loss": 0.03310008905827999,
      "grad_norm": 0.12236495316028595,
      "h_stock_valid_norm": 97.57565155029297,
      "invalid_tail_samples": 26.8,
      "learning_rate": 6.243983269276263e-05,
      "loss": 0.0374,
      "positive_teacher_pairs": 26165.8,
      "query_h_norm_ratio": 0.43998901844024657,
      "relation_loss": 0.004329662700183689,
      "selected_pairs": 26165.8,
      "step": 13470,
      "student_distance_mean": 25.943479919433592,
      "teacher_distance_mean": 0.8636618852615356,
      "valid_action_samples": 229.2,
      "weighted_auxiliary_loss": 0.004329662700183689,
      "weighted_relation_loss": 0.004329662700183689
    },
    {
      "action_query_batch_std": 0.3848234087228775,
      "action_query_norm": 42.91211624145508,
      "action_relation_loss": 0.004523379076272249,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.609915809167447,
      "flow_matching_loss": 0.03457635883241892,
      "grad_norm": 0.10470407456159592,
      "h_stock_valid_norm": 97.82453308105468,
      "invalid_tail_samples": 24.9,
      "learning_rate": 6.238644255626012e-05,
      "loss": 0.0391,
      "positive_teacher_pairs": 26592.1,
      "query_h_norm_ratio": 0.43867045640945435,
      "relation_loss": 0.004523379076272249,
      "selected_pairs": 26592.1,
      "step": 13480,
      "student_distance_mean": 25.53254337310791,
      "teacher_distance_mean": 0.8511119842529297,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.004523379076272249,
      "weighted_relation_loss": 0.004523379076272249
    },
    {
      "action_query_batch_std": 0.3856538563966751,
      "action_query_norm": 42.918741607666014,
      "action_relation_loss": 0.004324200330302119,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.619270346117867,
      "flow_matching_loss": 0.03462781347334385,
      "grad_norm": 0.12439552694559097,
      "h_stock_valid_norm": 97.56714553833008,
      "invalid_tail_samples": 22.9,
      "learning_rate": 6.233303736906193e-05,
      "loss": 0.039,
      "positive_teacher_pairs": 27061.4,
      "query_h_norm_ratio": 0.439895886182785,
      "relation_loss": 0.004324200330302119,
      "selected_pairs": 27061.4,
      "step": 13490,
      "student_distance_mean": 25.550766944885254,
      "teacher_distance_mean": 0.8544594764709472,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.004324200330302119,
      "weighted_relation_loss": 0.004324200330302119
    },
    {
      "action_query_batch_std": 0.3776727855205536,
      "action_query_norm": 42.893483352661136,
      "action_relation_loss": 0.004581791907548904,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.628624883068287,
      "flow_matching_loss": 0.034767828322947024,
      "grad_norm": 0.14442144334316254,
      "h_stock_valid_norm": 97.05257720947266,
      "invalid_tail_samples": 22.4,
      "learning_rate": 6.227961719606039e-05,
      "loss": 0.0393,
      "positive_teacher_pairs": 27171.5,
      "query_h_norm_ratio": 0.4419646352529526,
      "relation_loss": 0.004581791907548904,
      "selected_pairs": 27171.5,
      "step": 13500,
      "student_distance_mean": 25.286033058166502,
      "teacher_distance_mean": 0.8488042652606964,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.004581791907548904,
      "weighted_relation_loss": 0.004581791907548904
    },
    {
      "action_query_batch_std": 0.3840842604637146,
      "action_query_norm": 42.9171329498291,
      "action_relation_loss": 0.004639260889962316,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.63797942001871,
      "flow_matching_loss": 0.035385407879948615,
      "grad_norm": 0.12397748231887817,
      "h_stock_valid_norm": 97.52848739624024,
      "invalid_tail_samples": 25.1,
      "learning_rate": 6.222618210216608e-05,
      "loss": 0.04,
      "positive_teacher_pairs": 26555.0,
      "query_h_norm_ratio": 0.4400540918111801,
      "relation_loss": 0.004639260889962316,
      "selected_pairs": 26555.0,
      "step": 13510,
      "student_distance_mean": 25.602913856506348,
      "teacher_distance_mean": 0.8607168555259704,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.004639260889962316,
      "weighted_relation_loss": 0.004639260889962316
    },
    {
      "action_query_batch_std": 0.387912729382515,
      "action_query_norm": 42.93383369445801,
      "action_relation_loss": 0.004245844436809421,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.64733395696913,
      "flow_matching_loss": 0.03295196164399385,
      "grad_norm": 0.09406086802482605,
      "h_stock_valid_norm": 98.10119857788087,
      "invalid_tail_samples": 24.0,
      "learning_rate": 6.217273215230767e-05,
      "loss": 0.0372,
      "positive_teacher_pairs": 26808.8,
      "query_h_norm_ratio": 0.4376536995172501,
      "relation_loss": 0.004245844436809421,
      "selected_pairs": 26808.8,
      "step": 13520,
      "student_distance_mean": 25.85923194885254,
      "teacher_distance_mean": 0.8677144885063172,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.004245844436809421,
      "weighted_relation_loss": 0.004245844436809421
    },
    {
      "action_query_batch_std": 0.38415142595767976,
      "action_query_norm": 42.91267204284668,
      "action_relation_loss": 0.004499532142654061,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.656688493919551,
      "flow_matching_loss": 0.033077636547386643,
      "grad_norm": 0.11463518440723419,
      "h_stock_valid_norm": 98.00830154418945,
      "invalid_tail_samples": 22.6,
      "learning_rate": 6.211926741143188e-05,
      "loss": 0.0376,
      "positive_teacher_pairs": 27130.9,
      "query_h_norm_ratio": 0.43784962594509125,
      "relation_loss": 0.004499532142654061,
      "selected_pairs": 27130.9,
      "step": 13530,
      "student_distance_mean": 25.464859580993654,
      "teacher_distance_mean": 0.8526810586452485,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.004499532142654061,
      "weighted_relation_loss": 0.004499532142654061
    },
    {
      "action_query_batch_std": 0.38576348721981046,
      "action_query_norm": 42.910820388793944,
      "action_relation_loss": 0.004485007794573903,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.666043030869972,
      "flow_matching_loss": 0.03440592009574175,
      "grad_norm": 0.12464963644742966,
      "h_stock_valid_norm": 97.92037506103516,
      "invalid_tail_samples": 23.2,
      "learning_rate": 6.20657879445034e-05,
      "loss": 0.0389,
      "positive_teacher_pairs": 26985.8,
      "query_h_norm_ratio": 0.43822515308856963,
      "relation_loss": 0.004485007794573903,
      "selected_pairs": 26985.8,
      "step": 13540,
      "student_distance_mean": 25.734481430053712,
      "teacher_distance_mean": 0.8644776582717896,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.004485007794573903,
      "weighted_relation_loss": 0.004485007794573903
    },
    {
      "action_query_batch_std": 0.376139834523201,
      "action_query_norm": 42.87168312072754,
      "action_relation_loss": 0.004331628582440317,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.675397567820394,
      "flow_matching_loss": 0.03359673358500004,
      "grad_norm": 0.12730106711387634,
      "h_stock_valid_norm": 98.0699562072754,
      "invalid_tail_samples": 24.0,
      "learning_rate": 6.201229381650485e-05,
      "loss": 0.0379,
      "positive_teacher_pairs": 26808.9,
      "query_h_norm_ratio": 0.4371562868356705,
      "relation_loss": 0.004331628582440317,
      "selected_pairs": 26808.9,
      "step": 13550,
      "student_distance_mean": 25.102888107299805,
      "teacher_distance_mean": 0.8409919083118439,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.004331628582440317,
      "weighted_relation_loss": 0.004331628582440317
    },
    {
      "action_query_batch_std": 0.38658730685710907,
      "action_query_norm": 42.9072265625,
      "action_relation_loss": 0.004202657542191446,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.684752104770814,
      "flow_matching_loss": 0.032292125001549724,
      "grad_norm": 0.12095431238412857,
      "h_stock_valid_norm": 98.04512405395508,
      "invalid_tail_samples": 23.2,
      "learning_rate": 6.195878509243661e-05,
      "loss": 0.0365,
      "positive_teacher_pairs": 26989.8,
      "query_h_norm_ratio": 0.4376292616128922,
      "relation_loss": 0.004202657542191446,
      "selected_pairs": 26989.8,
      "step": 13560,
      "student_distance_mean": 25.58802261352539,
      "teacher_distance_mean": 0.8592975199222564,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.004202657542191446,
      "weighted_relation_loss": 0.004202657542191446
    },
    {
      "action_query_batch_std": 0.3857176035642624,
      "action_query_norm": 42.901855087280275,
      "action_relation_loss": 0.004530850425362587,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.694106641721234,
      "flow_matching_loss": 0.03551481943577528,
      "grad_norm": 0.1225956380367279,
      "h_stock_valid_norm": 97.73855285644531,
      "invalid_tail_samples": 23.0,
      "learning_rate": 6.190526183731685e-05,
      "loss": 0.04,
      "positive_teacher_pairs": 27036.0,
      "query_h_norm_ratio": 0.4389469176530838,
      "relation_loss": 0.004530850425362587,
      "selected_pairs": 27036.0,
      "step": 13570,
      "student_distance_mean": 25.704624366760253,
      "teacher_distance_mean": 0.8591034293174744,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.004530850425362587,
      "weighted_relation_loss": 0.004530850425362587
    },
    {
      "action_query_batch_std": 0.3793456703424454,
      "action_query_norm": 42.879541015625,
      "action_relation_loss": 0.004447926068678498,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.703461178671656,
      "flow_matching_loss": 0.032628596760332584,
      "grad_norm": 0.09646161645650864,
      "h_stock_valid_norm": 97.46525497436524,
      "invalid_tail_samples": 22.3,
      "learning_rate": 6.185172411618138e-05,
      "loss": 0.0371,
      "positive_teacher_pairs": 27200.5,
      "query_h_norm_ratio": 0.4399529337882996,
      "relation_loss": 0.004447926068678498,
      "selected_pairs": 27200.5,
      "step": 13580,
      "student_distance_mean": 25.10926971435547,
      "teacher_distance_mean": 0.8422351717948914,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.004447926068678498,
      "weighted_relation_loss": 0.004447926068678498
    },
    {
      "action_query_batch_std": 0.3847436308860779,
      "action_query_norm": 42.89809989929199,
      "action_relation_loss": 0.004512772336602211,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.712815715622076,
      "flow_matching_loss": 0.032550984993577003,
      "grad_norm": 0.10113260895013809,
      "h_stock_valid_norm": 97.28374938964843,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.179817199408355e-05,
      "loss": 0.0371,
      "positive_teacher_pairs": 26666.7,
      "query_h_norm_ratio": 0.440962952375412,
      "relation_loss": 0.004512772336602211,
      "selected_pairs": 26666.7,
      "step": 13590,
      "student_distance_mean": 25.70287170410156,
      "teacher_distance_mean": 0.8635909795761109,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.004512772336602211,
      "weighted_relation_loss": 0.004512772336602211
    },
    {
      "action_query_batch_std": 0.3759575605392456,
      "action_query_norm": 42.861825180053714,
      "action_relation_loss": 0.004584990069270134,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.722170252572498,
      "flow_matching_loss": 0.03481077626347542,
      "grad_norm": 0.11002442985773087,
      "h_stock_valid_norm": 96.54838638305664,
      "invalid_tail_samples": 24.1,
      "learning_rate": 6.174460553609426e-05,
      "loss": 0.0394,
      "positive_teacher_pairs": 26779.6,
      "query_h_norm_ratio": 0.4439501821994781,
      "relation_loss": 0.004584990069270134,
      "selected_pairs": 26779.6,
      "step": 13600,
      "student_distance_mean": 25.263327407836915,
      "teacher_distance_mean": 0.8450697064399719,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.004584990069270134,
      "weighted_relation_loss": 0.004584990069270134
    },
    {
      "action_query_batch_std": 0.38447371423244475,
      "action_query_norm": 42.90027008056641,
      "action_relation_loss": 0.0043005113489925865,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.731524789522918,
      "flow_matching_loss": 0.032344466634094717,
      "grad_norm": 0.1283971220254898,
      "h_stock_valid_norm": 96.97162017822265,
      "invalid_tail_samples": 21.7,
      "learning_rate": 6.16910248073018e-05,
      "loss": 0.0366,
      "positive_teacher_pairs": 27351.9,
      "query_h_norm_ratio": 0.4424070239067078,
      "relation_loss": 0.0043005113489925865,
      "selected_pairs": 27351.9,
      "step": 13610,
      "student_distance_mean": 25.56002712249756,
      "teacher_distance_mean": 0.852106124162674,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0043005113489925865,
      "weighted_relation_loss": 0.0043005113489925865
    },
    {
      "action_query_batch_std": 0.3781412333250046,
      "action_query_norm": 42.86783981323242,
      "action_relation_loss": 0.0043269374640658494,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.74087932647334,
      "flow_matching_loss": 0.03307845517992973,
      "grad_norm": 0.11947763711214066,
      "h_stock_valid_norm": 97.6743881225586,
      "invalid_tail_samples": 22.8,
      "learning_rate": 6.16374298728118e-05,
      "loss": 0.0374,
      "positive_teacher_pairs": 27082.4,
      "query_h_norm_ratio": 0.4388860762119293,
      "relation_loss": 0.0043269374640658494,
      "selected_pairs": 27082.4,
      "step": 13620,
      "student_distance_mean": 25.113819122314453,
      "teacher_distance_mean": 0.8486816942691803,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0043269374640658494,
      "weighted_relation_loss": 0.0043269374640658494
    },
    {
      "action_query_batch_std": 0.38356355428695676,
      "action_query_norm": 42.88403129577637,
      "action_relation_loss": 0.004448277689516544,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.75023386342376,
      "flow_matching_loss": 0.034869834035635,
      "grad_norm": 0.11103982478380203,
      "h_stock_valid_norm": 97.74959869384766,
      "invalid_tail_samples": 24.4,
      "learning_rate": 6.158382079774716e-05,
      "loss": 0.0393,
      "positive_teacher_pairs": 26715.2,
      "query_h_norm_ratio": 0.4387227714061737,
      "relation_loss": 0.004448277689516544,
      "selected_pairs": 26715.2,
      "step": 13630,
      "student_distance_mean": 25.534104919433595,
      "teacher_distance_mean": 0.852534294128418,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.004448277689516544,
      "weighted_relation_loss": 0.004448277689516544
    },
    {
      "action_query_batch_std": 0.383871927857399,
      "action_query_norm": 42.88931579589844,
      "action_relation_loss": 0.004312377399764955,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.75958840037418,
      "flow_matching_loss": 0.03208737354725599,
      "grad_norm": 0.09100951254367828,
      "h_stock_valid_norm": 97.95874633789063,
      "invalid_tail_samples": 24.1,
      "learning_rate": 6.153019764724799e-05,
      "loss": 0.0364,
      "positive_teacher_pairs": 26780.5,
      "query_h_norm_ratio": 0.4378328263759613,
      "relation_loss": 0.004312377399764955,
      "selected_pairs": 26780.5,
      "step": 13640,
      "student_distance_mean": 25.49989433288574,
      "teacher_distance_mean": 0.8497792065143586,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.004312377399764955,
      "weighted_relation_loss": 0.004312377399764955
    },
    {
      "action_query_batch_std": 0.3866390585899353,
      "action_query_norm": 42.896583557128906,
      "action_relation_loss": 0.004304970242083072,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.768942937324603,
      "flow_matching_loss": 0.03266190607100725,
      "grad_norm": 0.13612563908100128,
      "h_stock_valid_norm": 97.95749435424804,
      "invalid_tail_samples": 22.5,
      "learning_rate": 6.147656048647144e-05,
      "loss": 0.037,
      "positive_teacher_pairs": 27152.0,
      "query_h_norm_ratio": 0.43791103959083555,
      "relation_loss": 0.004304970242083072,
      "selected_pairs": 27152.0,
      "step": 13650,
      "student_distance_mean": 25.68922290802002,
      "teacher_distance_mean": 0.8614619374275208,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.004304970242083072,
      "weighted_relation_loss": 0.004304970242083072
    },
    {
      "action_query_batch_std": 0.3846693396568298,
      "action_query_norm": 42.88062438964844,
      "action_relation_loss": 0.0046358815860003235,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.778297474275023,
      "flow_matching_loss": 0.03364855982363224,
      "grad_norm": 0.14185389876365662,
      "h_stock_valid_norm": 97.63799667358398,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.142290938059173e-05,
      "loss": 0.0383,
      "positive_teacher_pairs": 26666.9,
      "query_h_norm_ratio": 0.4391808420419693,
      "relation_loss": 0.0046358815860003235,
      "selected_pairs": 26666.9,
      "step": 13660,
      "student_distance_mean": 25.566114044189455,
      "teacher_distance_mean": 0.8615171134471893,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0046358815860003235,
      "weighted_relation_loss": 0.0046358815860003235
    },
    {
      "action_query_batch_std": 0.3804801285266876,
      "action_query_norm": 42.86127281188965,
      "action_relation_loss": 0.004236296215094626,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.787652011225445,
      "flow_matching_loss": 0.03335653077811003,
      "grad_norm": 0.1271250694990158,
      "h_stock_valid_norm": 97.49902877807617,
      "invalid_tail_samples": 25.2,
      "learning_rate": 6.136924439480001e-05,
      "loss": 0.0376,
      "positive_teacher_pairs": 26531.1,
      "query_h_norm_ratio": 0.43961456418037415,
      "relation_loss": 0.004236296215094626,
      "selected_pairs": 26531.1,
      "step": 13670,
      "student_distance_mean": 25.383771324157713,
      "teacher_distance_mean": 0.8486636221408844,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.004236296215094626,
      "weighted_relation_loss": 0.004236296215094626
    },
    {
      "action_query_batch_std": 0.38141991198062897,
      "action_query_norm": 42.87075080871582,
      "action_relation_loss": 0.004432589653879404,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.797006548175865,
      "flow_matching_loss": 0.03294353745877743,
      "grad_norm": 0.1205068826675415,
      "h_stock_valid_norm": 97.20081329345703,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.13155655943043e-05,
      "loss": 0.0374,
      "positive_teacher_pairs": 26679.4,
      "query_h_norm_ratio": 0.4410557597875595,
      "relation_loss": 0.004432589653879404,
      "selected_pairs": 26679.4,
      "step": 13680,
      "student_distance_mean": 25.37530517578125,
      "teacher_distance_mean": 0.8429974853992462,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.004432589653879404,
      "weighted_relation_loss": 0.004432589653879404
    },
    {
      "action_query_batch_std": 0.3829523175954819,
      "action_query_norm": 42.86612777709961,
      "action_relation_loss": 0.004181233001872897,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.806361085126285,
      "flow_matching_loss": 0.03143247589468956,
      "grad_norm": 0.09970524907112122,
      "h_stock_valid_norm": 97.38197631835938,
      "invalid_tail_samples": 23.6,
      "learning_rate": 6.126187304432941e-05,
      "loss": 0.0356,
      "positive_teacher_pairs": 26899.8,
      "query_h_norm_ratio": 0.44018911719322207,
      "relation_loss": 0.004181233001872897,
      "selected_pairs": 26899.8,
      "step": 13690,
      "student_distance_mean": 25.477152252197264,
      "teacher_distance_mean": 0.8521052122116088,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.004181233001872897,
      "weighted_relation_loss": 0.004181233001872897
    },
    {
      "action_query_batch_std": 0.390162056684494,
      "action_query_norm": 42.878954315185545,
      "action_relation_loss": 0.00423410814255476,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.815715622076707,
      "flow_matching_loss": 0.03465699795633555,
      "grad_norm": 0.12359575927257538,
      "h_stock_valid_norm": 97.95830993652343,
      "invalid_tail_samples": 23.7,
      "learning_rate": 6.120816681011684e-05,
      "loss": 0.0389,
      "positive_teacher_pairs": 26870.5,
      "query_h_norm_ratio": 0.4377288997173309,
      "relation_loss": 0.00423410814255476,
      "selected_pairs": 26870.5,
      "step": 13700,
      "student_distance_mean": 25.905462074279786,
      "teacher_distance_mean": 0.8689496517181396,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.00423410814255476,
      "weighted_relation_loss": 0.00423410814255476
    },
    {
      "action_query_batch_std": 0.3781547486782074,
      "action_query_norm": 42.83450775146484,
      "action_relation_loss": 0.004247993137687445,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.825070159027128,
      "flow_matching_loss": 0.03257289454340935,
      "grad_norm": 0.0970306247472763,
      "h_stock_valid_norm": 97.98866271972656,
      "invalid_tail_samples": 24.3,
      "learning_rate": 6.115444695692474e-05,
      "loss": 0.0368,
      "positive_teacher_pairs": 26735.2,
      "query_h_norm_ratio": 0.43714614808559416,
      "relation_loss": 0.004247993137687445,
      "selected_pairs": 26735.2,
      "step": 13710,
      "student_distance_mean": 25.208770561218262,
      "teacher_distance_mean": 0.8471210360527038,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.004247993137687445,
      "weighted_relation_loss": 0.004247993137687445
    },
    {
      "action_query_batch_std": 0.38240003287792207,
      "action_query_norm": 42.85670127868652,
      "action_relation_loss": 0.004352928651496768,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.83442469597755,
      "flow_matching_loss": 0.03455248679965735,
      "grad_norm": 0.10543236136436462,
      "h_stock_valid_norm": 98.30841827392578,
      "invalid_tail_samples": 22.4,
      "learning_rate": 6.110071355002779e-05,
      "loss": 0.0389,
      "positive_teacher_pairs": 27173.6,
      "query_h_norm_ratio": 0.43594665825366974,
      "relation_loss": 0.004352928651496768,
      "selected_pairs": 27173.6,
      "step": 13720,
      "student_distance_mean": 25.47682476043701,
      "teacher_distance_mean": 0.8563750088214874,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.004352928651496768,
      "weighted_relation_loss": 0.004352928651496768
    },
    {
      "action_query_batch_std": 0.3863616168498993,
      "action_query_norm": 42.86952857971191,
      "action_relation_loss": 0.004620464262552559,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.84377923292797,
      "flow_matching_loss": 0.03382963929325342,
      "grad_norm": 0.10757213085889816,
      "h_stock_valid_norm": 98.11010360717773,
      "invalid_tail_samples": 21.8,
      "learning_rate": 6.104696665471714e-05,
      "loss": 0.0385,
      "positive_teacher_pairs": 27318.0,
      "query_h_norm_ratio": 0.4369614362716675,
      "relation_loss": 0.004620464262552559,
      "selected_pairs": 27318.0,
      "step": 13730,
      "student_distance_mean": 25.606081008911133,
      "teacher_distance_mean": 0.858522641658783,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.004620464262552559,
      "weighted_relation_loss": 0.004620464262552559
    },
    {
      "action_query_batch_std": 0.3843374937772751,
      "action_query_norm": 42.8636417388916,
      "action_relation_loss": 0.00445950860157609,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.853133769878392,
      "flow_matching_loss": 0.034167818166315554,
      "grad_norm": 0.09725341200828552,
      "h_stock_valid_norm": 98.8794662475586,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.099320633630036e-05,
      "loss": 0.0386,
      "positive_teacher_pairs": 26666.7,
      "query_h_norm_ratio": 0.4334961801767349,
      "relation_loss": 0.00445950860157609,
      "selected_pairs": 26666.7,
      "step": 13740,
      "student_distance_mean": 25.54770908355713,
      "teacher_distance_mean": 0.8556840240955352,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.00445950860157609,
      "weighted_relation_loss": 0.00445950860157609
    },
    {
      "action_query_batch_std": 0.3848859041929245,
      "action_query_norm": 42.86045532226562,
      "action_relation_loss": 0.004500194918364286,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.862488306828812,
      "flow_matching_loss": 0.032742413505911824,
      "grad_norm": 0.10533922910690308,
      "h_stock_valid_norm": 98.8098159790039,
      "invalid_tail_samples": 23.5,
      "learning_rate": 6.093943266010128e-05,
      "loss": 0.0372,
      "positive_teacher_pairs": 26918.8,
      "query_h_norm_ratio": 0.4337719649076462,
      "relation_loss": 0.004500194918364286,
      "selected_pairs": 26918.8,
      "step": 13750,
      "student_distance_mean": 25.710360527038574,
      "teacher_distance_mean": 0.8598865926265716,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.004500194918364286,
      "weighted_relation_loss": 0.004500194918364286
    },
    {
      "action_query_batch_std": 0.38314709067344666,
      "action_query_norm": 42.86137771606445,
      "action_relation_loss": 0.00430068310815841,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.871842843779232,
      "flow_matching_loss": 0.03200148344039917,
      "grad_norm": 0.09639216214418411,
      "h_stock_valid_norm": 98.65741195678712,
      "invalid_tail_samples": 24.2,
      "learning_rate": 6.088564569146e-05,
      "loss": 0.0363,
      "positive_teacher_pairs": 26756.2,
      "query_h_norm_ratio": 0.43444951474666593,
      "relation_loss": 0.00430068310815841,
      "selected_pairs": 26756.2,
      "step": 13760,
      "student_distance_mean": 25.57596492767334,
      "teacher_distance_mean": 0.8562843859195709,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.00430068310815841,
      "weighted_relation_loss": 0.00430068310815841
    },
    {
      "action_query_batch_std": 0.38381128311157225,
      "action_query_norm": 42.86020774841309,
      "action_relation_loss": 0.00437639735173434,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.881197380729654,
      "flow_matching_loss": 0.03416660465300083,
      "grad_norm": 0.1156512200832367,
      "h_stock_valid_norm": 98.14150314331054,
      "invalid_tail_samples": 24.7,
      "learning_rate": 6.083184549573274e-05,
      "loss": 0.0385,
      "positive_teacher_pairs": 26640.2,
      "query_h_norm_ratio": 0.43671947717666626,
      "relation_loss": 0.00437639735173434,
      "selected_pairs": 26640.2,
      "step": 13770,
      "student_distance_mean": 25.504975891113283,
      "teacher_distance_mean": 0.8583155751228333,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.00437639735173434,
      "weighted_relation_loss": 0.00437639735173434
    },
    {
      "action_query_batch_std": 0.38096716105937956,
      "action_query_norm": 42.836585235595706,
      "action_relation_loss": 0.004380984976887703,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.890551917680074,
      "flow_matching_loss": 0.03323866184800863,
      "grad_norm": 0.10195530951023102,
      "h_stock_valid_norm": 97.56330718994141,
      "invalid_tail_samples": 24.1,
      "learning_rate": 6.077803213829184e-05,
      "loss": 0.0376,
      "positive_teacher_pairs": 26778.2,
      "query_h_norm_ratio": 0.4390679866075516,
      "relation_loss": 0.004380984976887703,
      "selected_pairs": 26778.2,
      "step": 13780,
      "student_distance_mean": 25.396373558044434,
      "teacher_distance_mean": 0.860277658700943,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.004380984976887703,
      "weighted_relation_loss": 0.004380984976887703
    },
    {
      "action_query_batch_std": 0.3782562047243118,
      "action_query_norm": 42.82138671875,
      "action_relation_loss": 0.00449431985616684,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.899906454630496,
      "flow_matching_loss": 0.033893265388906,
      "grad_norm": 0.14252521097660065,
      "h_stock_valid_norm": 97.08406448364258,
      "invalid_tail_samples": 24.0,
      "learning_rate": 6.0724205684525594e-05,
      "loss": 0.0384,
      "positive_teacher_pairs": 26803.8,
      "query_h_norm_ratio": 0.44107796549797057,
      "relation_loss": 0.00449431985616684,
      "selected_pairs": 26803.8,
      "step": 13790,
      "student_distance_mean": 25.100705337524413,
      "teacher_distance_mean": 0.8471764326095581,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.00449431985616684,
      "weighted_relation_loss": 0.00449431985616684
    },
    {
      "action_query_batch_std": 0.38218564391136167,
      "action_query_norm": 42.83312454223633,
      "action_relation_loss": 0.004451752407476306,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.909260991580917,
      "flow_matching_loss": 0.03028068710118532,
      "grad_norm": 0.12019399553537369,
      "h_stock_valid_norm": 97.98285293579102,
      "invalid_tail_samples": 25.7,
      "learning_rate": 6.0670366199838215e-05,
      "loss": 0.0347,
      "positive_teacher_pairs": 26410.9,
      "query_h_norm_ratio": 0.4371523976325989,
      "relation_loss": 0.004451752407476306,
      "selected_pairs": 26410.9,
      "step": 13800,
      "student_distance_mean": 25.41598129272461,
      "teacher_distance_mean": 0.8551220417022705,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.004451752407476306,
      "weighted_relation_loss": 0.004451752407476306
    },
    {
      "action_query_batch_std": 0.3802265405654907,
      "action_query_norm": 42.82779235839844,
      "action_relation_loss": 0.004239039309322834,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.918615528531337,
      "flow_matching_loss": 0.0348316615447402,
      "grad_norm": 0.12478228658437729,
      "h_stock_valid_norm": 98.26446533203125,
      "invalid_tail_samples": 22.4,
      "learning_rate": 6.061651374964974e-05,
      "loss": 0.0391,
      "positive_teacher_pairs": 27181.3,
      "query_h_norm_ratio": 0.43584645688533785,
      "relation_loss": 0.004239039309322834,
      "selected_pairs": 27181.3,
      "step": 13810,
      "student_distance_mean": 25.265884399414062,
      "teacher_distance_mean": 0.8506572365760803,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.004239039309322834,
      "weighted_relation_loss": 0.004239039309322834
    },
    {
      "action_query_batch_std": 0.38307978212833405,
      "action_query_norm": 42.829307556152344,
      "action_relation_loss": 0.0043103916104882956,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.927970065481759,
      "flow_matching_loss": 0.03203777130693197,
      "grad_norm": 0.12826859951019287,
      "h_stock_valid_norm": 98.30426483154297,
      "invalid_tail_samples": 25.5,
      "learning_rate": 6.0562648399396004e-05,
      "loss": 0.0363,
      "positive_teacher_pairs": 26455.7,
      "query_h_norm_ratio": 0.4356852859258652,
      "relation_loss": 0.0043103916104882956,
      "selected_pairs": 26455.7,
      "step": 13820,
      "student_distance_mean": 25.468237495422365,
      "teacher_distance_mean": 0.855028235912323,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0043103916104882956,
      "weighted_relation_loss": 0.0043103916104882956
    },
    {
      "action_query_batch_std": 0.3768830090761185,
      "action_query_norm": 42.802124404907225,
      "action_relation_loss": 0.004190773260779679,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.937324602432179,
      "flow_matching_loss": 0.03310348317027092,
      "grad_norm": 0.10232819616794586,
      "h_stock_valid_norm": 98.4213966369629,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.050877021452845e-05,
      "loss": 0.0373,
      "positive_teacher_pairs": 26668.8,
      "query_h_norm_ratio": 0.43489229381084443,
      "relation_loss": 0.004190773260779679,
      "selected_pairs": 26668.8,
      "step": 13830,
      "student_distance_mean": 25.13729190826416,
      "teacher_distance_mean": 0.8463753223419189,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.004190773260779679,
      "weighted_relation_loss": 0.004190773260779679
    },
    {
      "action_query_batch_std": 0.3842443495988846,
      "action_query_norm": 42.8337646484375,
      "action_relation_loss": 0.004488215269520879,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.946679139382601,
      "flow_matching_loss": 0.033060939982533455,
      "grad_norm": 0.11447609961032867,
      "h_stock_valid_norm": 98.2484634399414,
      "invalid_tail_samples": 26.4,
      "learning_rate": 6.04548792605142e-05,
      "loss": 0.0375,
      "positive_teacher_pairs": 26260.4,
      "query_h_norm_ratio": 0.43597880005836487,
      "relation_loss": 0.004488215269520879,
      "selected_pairs": 26260.4,
      "step": 13840,
      "student_distance_mean": 25.531402397155762,
      "teacher_distance_mean": 0.8557541906833649,
      "valid_action_samples": 229.6,
      "weighted_auxiliary_loss": 0.004488215269520879,
      "weighted_relation_loss": 0.004488215269520879
    },
    {
      "action_query_batch_std": 0.38802794814109803,
      "action_query_norm": 42.85242042541504,
      "action_relation_loss": 0.004249884281307459,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.956033676333021,
      "flow_matching_loss": 0.03243545722216368,
      "grad_norm": 0.09448640048503876,
      "h_stock_valid_norm": 97.46012573242187,
      "invalid_tail_samples": 22.2,
      "learning_rate": 6.04009756028358e-05,
      "loss": 0.0367,
      "positive_teacher_pairs": 27220.0,
      "query_h_norm_ratio": 0.4396972328424454,
      "relation_loss": 0.004249884281307459,
      "selected_pairs": 27220.0,
      "step": 13850,
      "student_distance_mean": 25.732538986206055,
      "teacher_distance_mean": 0.8620552659034729,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.004249884281307459,
      "weighted_relation_loss": 0.004249884281307459
    },
    {
      "action_query_batch_std": 0.3853968381881714,
      "action_query_norm": 42.832115173339844,
      "action_relation_loss": 0.004324481659568846,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.965388213283443,
      "flow_matching_loss": 0.034078439511358735,
      "grad_norm": 0.10464924573898315,
      "h_stock_valid_norm": 97.92128524780273,
      "invalid_tail_samples": 25.3,
      "learning_rate": 6.0347059306991294e-05,
      "loss": 0.0384,
      "positive_teacher_pairs": 26502.8,
      "query_h_norm_ratio": 0.4374198317527771,
      "relation_loss": 0.004324481659568846,
      "selected_pairs": 26502.8,
      "step": 13860,
      "student_distance_mean": 25.76369514465332,
      "teacher_distance_mean": 0.8638887107372284,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.004324481659568846,
      "weighted_relation_loss": 0.004324481659568846
    },
    {
      "action_query_batch_std": 0.381464746594429,
      "action_query_norm": 42.829674911499026,
      "action_relation_loss": 0.0042780309449881315,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.974742750233863,
      "flow_matching_loss": 0.03246240895241499,
      "grad_norm": 0.11533666402101517,
      "h_stock_valid_norm": 98.1376235961914,
      "invalid_tail_samples": 21.9,
      "learning_rate": 6.029313043849407e-05,
      "loss": 0.0367,
      "positive_teacher_pairs": 27290.1,
      "query_h_norm_ratio": 0.43642825484275816,
      "relation_loss": 0.0042780309449881315,
      "selected_pairs": 27290.1,
      "step": 13870,
      "student_distance_mean": 25.404516410827636,
      "teacher_distance_mean": 0.8541031181812286,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.0042780309449881315,
      "weighted_relation_loss": 0.0042780309449881315
    },
    {
      "action_query_batch_std": 0.38057767152786254,
      "action_query_norm": 42.811133193969724,
      "action_relation_loss": 0.004355580219998956,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.984097287184284,
      "flow_matching_loss": 0.031625812873244286,
      "grad_norm": 0.15295575559139252,
      "h_stock_valid_norm": 97.6474723815918,
      "invalid_tail_samples": 24.3,
      "learning_rate": 6.023918906287279e-05,
      "loss": 0.036,
      "positive_teacher_pairs": 26736.5,
      "query_h_norm_ratio": 0.4384272187948227,
      "relation_loss": 0.004355580219998956,
      "selected_pairs": 26736.5,
      "step": 13880,
      "student_distance_mean": 25.271393013000488,
      "teacher_distance_mean": 0.8522013068199158,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.004355580219998956,
      "weighted_relation_loss": 0.004355580219998956
    },
    {
      "action_query_batch_std": 0.3846351355314255,
      "action_query_norm": 42.817527770996094,
      "action_relation_loss": 0.004200878157280386,
      "auxiliary_loss_weight": 1.0,
      "epoch": 12.993451824134706,
      "flow_matching_loss": 0.032107621990144254,
      "grad_norm": 0.11809571087360382,
      "h_stock_valid_norm": 96.79698333740234,
      "invalid_tail_samples": 22.7,
      "learning_rate": 6.01852352456713e-05,
      "loss": 0.0363,
      "positive_teacher_pairs": 27106.3,
      "query_h_norm_ratio": 0.4423441499471664,
      "relation_loss": 0.004200878157280386,
      "selected_pairs": 27106.3,
      "step": 13890,
      "student_distance_mean": 25.633599090576173,
      "teacher_distance_mean": 0.8675262153148651,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.004200878157280386,
      "weighted_relation_loss": 0.004200878157280386
    },
    {
      "action_query_batch_std": 0.3799960970878601,
      "action_query_norm": 42.80333786010742,
      "action_relation_loss": 0.004139935318380594,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.002806361085126,
      "flow_matching_loss": 0.03179317843168974,
      "grad_norm": 0.14950096607208252,
      "h_stock_valid_norm": 96.86601486206055,
      "invalid_tail_samples": 22.2,
      "learning_rate": 6.013126905244858e-05,
      "loss": 0.0359,
      "positive_teacher_pairs": 24235.0,
      "query_h_norm_ratio": 0.44188560247421266,
      "relation_loss": 0.004139935318380594,
      "selected_pairs": 24235.0,
      "step": 13900,
      "student_distance_mean": 25.28439598083496,
      "teacher_distance_mean": 0.8519642889499665,
      "valid_action_samples": 213.9,
      "weighted_auxiliary_loss": 0.004139935318380594,
      "weighted_relation_loss": 0.004139935318380594
    },
    {
      "action_query_batch_std": 0.38272006809711456,
      "action_query_norm": 42.812394332885745,
      "action_relation_loss": 0.004235674790106714,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.012160898035548,
      "flow_matching_loss": 0.033464810065925124,
      "grad_norm": 0.13650640845298767,
      "h_stock_valid_norm": 97.09598159790039,
      "invalid_tail_samples": 23.3,
      "learning_rate": 6.007729054877864e-05,
      "loss": 0.0377,
      "positive_teacher_pairs": 26963.8,
      "query_h_norm_ratio": 0.44092977643013,
      "relation_loss": 0.004235674790106714,
      "selected_pairs": 26963.8,
      "step": 13910,
      "student_distance_mean": 25.48650703430176,
      "teacher_distance_mean": 0.8599145531654357,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.004235674790106714,
      "weighted_relation_loss": 0.004235674790106714
    },
    {
      "action_query_batch_std": 0.3868546485900879,
      "action_query_norm": 42.83275527954102,
      "action_relation_loss": 0.004246284253895283,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.021515434985968,
      "flow_matching_loss": 0.034617511741817,
      "grad_norm": 0.10587436705827713,
      "h_stock_valid_norm": 96.91873474121094,
      "invalid_tail_samples": 21.3,
      "learning_rate": 6.0023299800250466e-05,
      "loss": 0.0389,
      "positive_teacher_pairs": 27438.0,
      "query_h_norm_ratio": 0.4419477194547653,
      "relation_loss": 0.004246284253895283,
      "selected_pairs": 27438.0,
      "step": 13920,
      "student_distance_mean": 25.569794273376466,
      "teacher_distance_mean": 0.8572829246520997,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.004246284253895283,
      "weighted_relation_loss": 0.004246284253895283
    },
    {
      "action_query_batch_std": 0.3848778009414673,
      "action_query_norm": 42.82086868286133,
      "action_relation_loss": 0.0040225841803476214,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.030869971936388,
      "flow_matching_loss": 0.032760356552898884,
      "grad_norm": 0.1307702362537384,
      "h_stock_valid_norm": 97.42843093872071,
      "invalid_tail_samples": 20.3,
      "learning_rate": 5.9969296872467894e-05,
      "loss": 0.0368,
      "positive_teacher_pairs": 27668.2,
      "query_h_norm_ratio": 0.43952047526836396,
      "relation_loss": 0.0040225841803476214,
      "selected_pairs": 27668.2,
      "step": 13930,
      "student_distance_mean": 25.65617256164551,
      "teacher_distance_mean": 0.8580231130123138,
      "valid_action_samples": 235.7,
      "weighted_auxiliary_loss": 0.0040225841803476214,
      "weighted_relation_loss": 0.0040225841803476214
    },
    {
      "action_query_batch_std": 0.3827452152967453,
      "action_query_norm": 42.80761032104492,
      "action_relation_loss": 0.004008765215985477,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.04022450888681,
      "flow_matching_loss": 0.032314978539943695,
      "grad_norm": 0.13401073217391968,
      "h_stock_valid_norm": 97.94767456054687,
      "invalid_tail_samples": 22.6,
      "learning_rate": 5.9915281831049585e-05,
      "loss": 0.0363,
      "positive_teacher_pairs": 27129.6,
      "query_h_norm_ratio": 0.43705074191093446,
      "relation_loss": 0.004008765215985477,
      "selected_pairs": 27129.6,
      "step": 13940,
      "student_distance_mean": 25.495705795288085,
      "teacher_distance_mean": 0.8497498333454132,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.004008765215985477,
      "weighted_relation_loss": 0.004008765215985477
    },
    {
      "action_query_batch_std": 0.3803970217704773,
      "action_query_norm": 42.79438667297363,
      "action_relation_loss": 0.00414782150182873,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.04957904583723,
      "flow_matching_loss": 0.03409364074468613,
      "grad_norm": 0.11368585377931595,
      "h_stock_valid_norm": 98.51945495605469,
      "invalid_tail_samples": 22.5,
      "learning_rate": 5.98612547416289e-05,
      "loss": 0.0382,
      "positive_teacher_pairs": 27150.1,
      "query_h_norm_ratio": 0.4343768239021301,
      "relation_loss": 0.00414782150182873,
      "selected_pairs": 27150.1,
      "step": 13950,
      "student_distance_mean": 25.272950172424316,
      "teacher_distance_mean": 0.8522293031215668,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.00414782150182873,
      "weighted_relation_loss": 0.00414782150182873
    },
    {
      "action_query_batch_std": 0.37807280719280245,
      "action_query_norm": 42.79259490966797,
      "action_relation_loss": 0.004333947249688208,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.058933582787652,
      "flow_matching_loss": 0.032155930437147616,
      "grad_norm": 0.12975087761878967,
      "h_stock_valid_norm": 98.75348052978515,
      "invalid_tail_samples": 23.4,
      "learning_rate": 5.9807215669853855e-05,
      "loss": 0.0365,
      "positive_teacher_pairs": 26945.5,
      "query_h_norm_ratio": 0.43332905173301695,
      "relation_loss": 0.004333947249688208,
      "selected_pairs": 26945.5,
      "step": 13960,
      "student_distance_mean": 25.221990203857423,
      "teacher_distance_mean": 0.8515672624111176,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.004333947249688208,
      "weighted_relation_loss": 0.004333947249688208
    },
    {
      "action_query_batch_std": 0.3850203573703766,
      "action_query_norm": 42.818972778320315,
      "action_relation_loss": 0.004200907051563263,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.068288119738073,
      "flow_matching_loss": 0.03206219617277384,
      "grad_norm": 0.14702191948890686,
      "h_stock_valid_norm": 98.24392395019531,
      "invalid_tail_samples": 22.6,
      "learning_rate": 5.9753164681387e-05,
      "loss": 0.0363,
      "positive_teacher_pairs": 27126.6,
      "query_h_norm_ratio": 0.4358494967222214,
      "relation_loss": 0.004200907051563263,
      "selected_pairs": 27126.6,
      "step": 13970,
      "student_distance_mean": 25.621056938171385,
      "teacher_distance_mean": 0.85916708111763,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.004200907051563263,
      "weighted_relation_loss": 0.004200907051563263
    },
    {
      "action_query_batch_std": 0.38649747967720033,
      "action_query_norm": 42.82124137878418,
      "action_relation_loss": 0.004427014873363078,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.077642656688495,
      "flow_matching_loss": 0.0332539351657033,
      "grad_norm": 0.0917983204126358,
      "h_stock_valid_norm": 97.68650665283204,
      "invalid_tail_samples": 24.4,
      "learning_rate": 5.969910184190539e-05,
      "loss": 0.0377,
      "positive_teacher_pairs": 26726.3,
      "query_h_norm_ratio": 0.43835954964160917,
      "relation_loss": 0.004427014873363078,
      "selected_pairs": 26726.3,
      "step": 13980,
      "student_distance_mean": 25.76392879486084,
      "teacher_distance_mean": 0.8616910994052887,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.004427014873363078,
      "weighted_relation_loss": 0.004427014873363078
    },
    {
      "action_query_batch_std": 0.38266201615333556,
      "action_query_norm": 42.80489234924316,
      "action_relation_loss": 0.0040668938774615524,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.086997193638915,
      "flow_matching_loss": 0.03187273293733597,
      "grad_norm": 0.12219680100679398,
      "h_stock_valid_norm": 98.10760040283203,
      "invalid_tail_samples": 25.7,
      "learning_rate": 5.9645027217100477e-05,
      "loss": 0.0359,
      "positive_teacher_pairs": 26406.5,
      "query_h_norm_ratio": 0.4363080471754074,
      "relation_loss": 0.0040668938774615524,
      "selected_pairs": 26406.5,
      "step": 13990,
      "student_distance_mean": 25.474846839904785,
      "teacher_distance_mean": 0.8538379490375518,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0040668938774615524,
      "weighted_relation_loss": 0.0040668938774615524
    },
    {
      "action_query_batch_std": 0.3811198651790619,
      "action_query_norm": 42.79435005187988,
      "action_relation_loss": 0.00422023213468492,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.096351730589335,
      "flow_matching_loss": 0.03273336682468653,
      "grad_norm": 0.11613141000270844,
      "h_stock_valid_norm": 98.18367004394531,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.959094087267804e-05,
      "loss": 0.037,
      "positive_teacher_pairs": 26987.7,
      "query_h_norm_ratio": 0.4358653277158737,
      "relation_loss": 0.00422023213468492,
      "selected_pairs": 26987.7,
      "step": 14000,
      "student_distance_mean": 25.424061393737794,
      "teacher_distance_mean": 0.8550365090370178,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.00422023213468492,
      "weighted_relation_loss": 0.00422023213468492
    },
    {
      "action_query_batch_std": 0.3817324280738831,
      "action_query_norm": 42.78629341125488,
      "action_relation_loss": 0.00393673584330827,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.105706267539757,
      "flow_matching_loss": 0.032003282196819784,
      "grad_norm": 0.10953201353549957,
      "h_stock_valid_norm": 97.73670806884766,
      "invalid_tail_samples": 26.5,
      "learning_rate": 5.9536842874358065e-05,
      "loss": 0.0359,
      "positive_teacher_pairs": 26228.4,
      "query_h_norm_ratio": 0.4377748370170593,
      "relation_loss": 0.00393673584330827,
      "selected_pairs": 26228.4,
      "step": 14010,
      "student_distance_mean": 25.535710525512695,
      "teacher_distance_mean": 0.8603272318840027,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.00393673584330827,
      "weighted_relation_loss": 0.00393673584330827
    },
    {
      "action_query_batch_std": 0.38090287148952484,
      "action_query_norm": 42.78321876525879,
      "action_relation_loss": 0.004036115342751145,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.115060804490177,
      "flow_matching_loss": 0.03299188055098057,
      "grad_norm": 0.10821781307458878,
      "h_stock_valid_norm": 97.49703674316406,
      "invalid_tail_samples": 22.5,
      "learning_rate": 5.9482733287874734e-05,
      "loss": 0.037,
      "positive_teacher_pairs": 27153.3,
      "query_h_norm_ratio": 0.43882031440734864,
      "relation_loss": 0.004036115342751145,
      "selected_pairs": 27153.3,
      "step": 14020,
      "student_distance_mean": 25.325497817993163,
      "teacher_distance_mean": 0.8541046977043152,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.004036115342751145,
      "weighted_relation_loss": 0.004036115342751145
    },
    {
      "action_query_batch_std": 0.3777468532323837,
      "action_query_norm": 42.76594924926758,
      "action_relation_loss": 0.004237015848048031,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.1244153414406,
      "flow_matching_loss": 0.032707509398460385,
      "grad_norm": 0.10805992037057877,
      "h_stock_valid_norm": 97.7596923828125,
      "invalid_tail_samples": 23.3,
      "learning_rate": 5.942861217897631e-05,
      "loss": 0.0369,
      "positive_teacher_pairs": 26967.4,
      "query_h_norm_ratio": 0.4374676674604416,
      "relation_loss": 0.004237015848048031,
      "selected_pairs": 26967.4,
      "step": 14030,
      "student_distance_mean": 25.231593132019043,
      "teacher_distance_mean": 0.8553227722644806,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.004237015848048031,
      "weighted_relation_loss": 0.004237015848048031
    },
    {
      "action_query_batch_std": 0.38104006052017214,
      "action_query_norm": 42.77674179077148,
      "action_relation_loss": 0.004431136045604944,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.13376987839102,
      "flow_matching_loss": 0.03294124156236648,
      "grad_norm": 0.11913934350013733,
      "h_stock_valid_norm": 98.30612182617188,
      "invalid_tail_samples": 25.2,
      "learning_rate": 5.9374479613425e-05,
      "loss": 0.0374,
      "positive_teacher_pairs": 26530.9,
      "query_h_norm_ratio": 0.4351411938667297,
      "relation_loss": 0.004431136045604944,
      "selected_pairs": 26530.9,
      "step": 14040,
      "student_distance_mean": 25.353912544250488,
      "teacher_distance_mean": 0.858187985420227,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.004431136045604944,
      "weighted_relation_loss": 0.004431136045604944
    },
    {
      "action_query_batch_std": 0.38223282992839813,
      "action_query_norm": 42.789088821411134,
      "action_relation_loss": 0.004035785794258118,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.14312441534144,
      "flow_matching_loss": 0.03314325623214245,
      "grad_norm": 0.13522009551525116,
      "h_stock_valid_norm": 99.00991668701172,
      "invalid_tail_samples": 27.3,
      "learning_rate": 5.932033565699704e-05,
      "loss": 0.0372,
      "positive_teacher_pairs": 26044.5,
      "query_h_norm_ratio": 0.43217220306396487,
      "relation_loss": 0.004035785794258118,
      "selected_pairs": 26044.5,
      "step": 14050,
      "student_distance_mean": 25.493194580078125,
      "teacher_distance_mean": 0.8578296601772308,
      "valid_action_samples": 228.7,
      "weighted_auxiliary_loss": 0.004035785794258118,
      "weighted_relation_loss": 0.004035785794258118
    },
    {
      "action_query_batch_std": 0.38559211790561676,
      "action_query_norm": 42.790456771850586,
      "action_relation_loss": 0.004072371846996247,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.152478952291862,
      "flow_matching_loss": 0.03233886230736971,
      "grad_norm": 0.15359032154083252,
      "h_stock_valid_norm": 99.02014236450195,
      "invalid_tail_samples": 26.1,
      "learning_rate": 5.926618037548237e-05,
      "loss": 0.0364,
      "positive_teacher_pairs": 26323.2,
      "query_h_norm_ratio": 0.4321424633264542,
      "relation_loss": 0.004072371846996247,
      "selected_pairs": 26323.2,
      "step": 14060,
      "student_distance_mean": 25.563468742370606,
      "teacher_distance_mean": 0.8634540498256683,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.004072371846996247,
      "weighted_relation_loss": 0.004072371846996247
    },
    {
      "action_query_batch_std": 0.3768066942691803,
      "action_query_norm": 42.75435218811035,
      "action_relation_loss": 0.004026155872270465,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.161833489242282,
      "flow_matching_loss": 0.03286741450428963,
      "grad_norm": 0.12555235624313354,
      "h_stock_valid_norm": 98.21523361206054,
      "invalid_tail_samples": 25.4,
      "learning_rate": 5.9212013834684824e-05,
      "loss": 0.0369,
      "positive_teacher_pairs": 26480.9,
      "query_h_norm_ratio": 0.43531774282455443,
      "relation_loss": 0.004026155872270465,
      "selected_pairs": 26480.9,
      "step": 14070,
      "student_distance_mean": 25.01300754547119,
      "teacher_distance_mean": 0.8470171868801117,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.004026155872270465,
      "weighted_relation_loss": 0.004026155872270465
    },
    {
      "action_query_batch_std": 0.37603531777858734,
      "action_query_norm": 42.76179428100586,
      "action_relation_loss": 0.004312229598872364,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.171188026192704,
      "flow_matching_loss": 0.03280347529798746,
      "grad_norm": 0.09593292325735092,
      "h_stock_valid_norm": 97.37412567138672,
      "invalid_tail_samples": 23.4,
      "learning_rate": 5.9157836100421824e-05,
      "loss": 0.0371,
      "positive_teacher_pairs": 26952.2,
      "query_h_norm_ratio": 0.4391545295715332,
      "relation_loss": 0.004312229598872364,
      "selected_pairs": 26952.2,
      "step": 14080,
      "student_distance_mean": 25.0602819442749,
      "teacher_distance_mean": 0.845092260837555,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.004312229598872364,
      "weighted_relation_loss": 0.004312229598872364
    },
    {
      "action_query_batch_std": 0.386503142118454,
      "action_query_norm": 42.79654884338379,
      "action_relation_loss": 0.004099552659317851,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.180542563143124,
      "flow_matching_loss": 0.03426348064094782,
      "grad_norm": 0.13383078575134277,
      "h_stock_valid_norm": 96.36313400268554,
      "invalid_tail_samples": 23.6,
      "learning_rate": 5.910364723852444e-05,
      "loss": 0.0384,
      "positive_teacher_pairs": 26897.0,
      "query_h_norm_ratio": 0.44412451088428495,
      "relation_loss": 0.004099552659317851,
      "selected_pairs": 26897.0,
      "step": 14090,
      "student_distance_mean": 25.624645042419434,
      "teacher_distance_mean": 0.8552297234535218,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.004099552659317851,
      "weighted_relation_loss": 0.004099552659317851
    },
    {
      "action_query_batch_std": 0.3813787639141083,
      "action_query_norm": 42.76907501220703,
      "action_relation_loss": 0.004003221984021366,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.189897100093546,
      "flow_matching_loss": 0.0332915335893631,
      "grad_norm": 0.1054699569940567,
      "h_stock_valid_norm": 96.10514831542969,
      "invalid_tail_samples": 25.2,
      "learning_rate": 5.904944731483723e-05,
      "loss": 0.0373,
      "positive_teacher_pairs": 26528.1,
      "query_h_norm_ratio": 0.44502730667591095,
      "relation_loss": 0.004003221984021366,
      "selected_pairs": 26528.1,
      "step": 14100,
      "student_distance_mean": 25.27428436279297,
      "teacher_distance_mean": 0.8526037573814392,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.004003221984021366,
      "weighted_relation_loss": 0.004003221984021366
    },
    {
      "action_query_batch_std": 0.37868200838565824,
      "action_query_norm": 42.75496826171875,
      "action_relation_loss": 0.004257587413303554,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.199251637043966,
      "flow_matching_loss": 0.03333237320184708,
      "grad_norm": 0.10687543451786041,
      "h_stock_valid_norm": 96.91015625,
      "invalid_tail_samples": 24.5,
      "learning_rate": 5.899523639521825e-05,
      "loss": 0.0376,
      "positive_teacher_pairs": 26693.0,
      "query_h_norm_ratio": 0.44118438065052035,
      "relation_loss": 0.004257587413303554,
      "selected_pairs": 26693.0,
      "step": 14110,
      "student_distance_mean": 25.10465431213379,
      "teacher_distance_mean": 0.8532336533069611,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.004257587413303554,
      "weighted_relation_loss": 0.004257587413303554
    },
    {
      "action_query_batch_std": 0.38077756464481355,
      "action_query_norm": 42.76925392150879,
      "action_relation_loss": 0.0041658304864540694,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.208606173994387,
      "flow_matching_loss": 0.03245426006615162,
      "grad_norm": 0.10679878294467926,
      "h_stock_valid_norm": 97.59395904541016,
      "invalid_tail_samples": 26.2,
      "learning_rate": 5.894101454553883e-05,
      "loss": 0.0366,
      "positive_teacher_pairs": 26299.5,
      "query_h_norm_ratio": 0.4382497161626816,
      "relation_loss": 0.0041658304864540694,
      "selected_pairs": 26299.5,
      "step": 14120,
      "student_distance_mean": 25.459918594360353,
      "teacher_distance_mean": 0.8592589855194092,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.0041658304864540694,
      "weighted_relation_loss": 0.0041658304864540694
    },
    {
      "action_query_batch_std": 0.3838575303554535,
      "action_query_norm": 42.77010498046875,
      "action_relation_loss": 0.0040885205147787925,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.217960710944809,
      "flow_matching_loss": 0.03223901204764843,
      "grad_norm": 0.10472135245800018,
      "h_stock_valid_norm": 98.05609893798828,
      "invalid_tail_samples": 25.3,
      "learning_rate": 5.888678183168368e-05,
      "loss": 0.0363,
      "positive_teacher_pairs": 26503.4,
      "query_h_norm_ratio": 0.4361818015575409,
      "relation_loss": 0.0040885205147787925,
      "selected_pairs": 26503.4,
      "step": 14130,
      "student_distance_mean": 25.577071571350096,
      "teacher_distance_mean": 0.8589671432971955,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0040885205147787925,
      "weighted_relation_loss": 0.0040885205147787925
    },
    {
      "action_query_batch_std": 0.38337860107421873,
      "action_query_norm": 42.754859161376956,
      "action_relation_loss": 0.004118054127320647,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.227315247895229,
      "flow_matching_loss": 0.031033811159431934,
      "grad_norm": 0.10442625731229782,
      "h_stock_valid_norm": 97.98768768310546,
      "invalid_tail_samples": 21.4,
      "learning_rate": 5.883253831955061e-05,
      "loss": 0.0352,
      "positive_teacher_pairs": 27409.9,
      "query_h_norm_ratio": 0.4363295018672943,
      "relation_loss": 0.004118054127320647,
      "selected_pairs": 27409.9,
      "step": 14140,
      "student_distance_mean": 25.421866607666015,
      "teacher_distance_mean": 0.8579809427261352,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.004118054127320647,
      "weighted_relation_loss": 0.004118054127320647
    },
    {
      "action_query_batch_std": 0.3779803991317749,
      "action_query_norm": 42.7338623046875,
      "action_relation_loss": 0.003966820542700589,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.23666978484565,
      "flow_matching_loss": 0.03197895269840956,
      "grad_norm": 0.11208344995975494,
      "h_stock_valid_norm": 98.36881866455079,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.8778284075050625e-05,
      "loss": 0.0359,
      "positive_teacher_pairs": 26997.3,
      "query_h_norm_ratio": 0.4344289392232895,
      "relation_loss": 0.003966820542700589,
      "selected_pairs": 26997.3,
      "step": 14150,
      "student_distance_mean": 25.171656990051268,
      "teacher_distance_mean": 0.8586224734783172,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.003966820542700589,
      "weighted_relation_loss": 0.003966820542700589
    },
    {
      "action_query_batch_std": 0.3776886761188507,
      "action_query_norm": 42.7395736694336,
      "action_relation_loss": 0.004470288706943393,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.246024321796071,
      "flow_matching_loss": 0.03274149931967259,
      "grad_norm": 0.10639866441488266,
      "h_stock_valid_norm": 97.78124694824218,
      "invalid_tail_samples": 23.0,
      "learning_rate": 5.872401916410777e-05,
      "loss": 0.0372,
      "positive_teacher_pairs": 27043.7,
      "query_h_norm_ratio": 0.43709692656993865,
      "relation_loss": 0.004470288706943393,
      "selected_pairs": 27043.7,
      "step": 14160,
      "student_distance_mean": 25.049954986572267,
      "teacher_distance_mean": 0.8471593856811523,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.004470288706943393,
      "weighted_relation_loss": 0.004470288706943393
    },
    {
      "action_query_batch_std": 0.38502673208713534,
      "action_query_norm": 42.76173858642578,
      "action_relation_loss": 0.004009977495297789,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.255378858746493,
      "flow_matching_loss": 0.0316594410687685,
      "grad_norm": 0.1384436935186386,
      "h_stock_valid_norm": 97.99805068969727,
      "invalid_tail_samples": 22.0,
      "learning_rate": 5.8669743652659014e-05,
      "loss": 0.0357,
      "positive_teacher_pairs": 27269.7,
      "query_h_norm_ratio": 0.43636255860328677,
      "relation_loss": 0.004009977495297789,
      "selected_pairs": 27269.7,
      "step": 14170,
      "student_distance_mean": 25.634110260009766,
      "teacher_distance_mean": 0.8600100994110107,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.004009977495297789,
      "weighted_relation_loss": 0.004009977495297789
    },
    {
      "action_query_batch_std": 0.38096067905426023,
      "action_query_norm": 42.73626556396484,
      "action_relation_loss": 0.004017475433647633,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.264733395696913,
      "flow_matching_loss": 0.03140897229313851,
      "grad_norm": 0.14213702082633972,
      "h_stock_valid_norm": 98.42823791503906,
      "invalid_tail_samples": 24.0,
      "learning_rate": 5.86154576066542e-05,
      "loss": 0.0354,
      "positive_teacher_pairs": 26804.7,
      "query_h_norm_ratio": 0.4341930478811264,
      "relation_loss": 0.004017475433647633,
      "selected_pairs": 26804.7,
      "step": 14180,
      "student_distance_mean": 25.289954948425294,
      "teacher_distance_mean": 0.8556435346603394,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.004017475433647633,
      "weighted_relation_loss": 0.004017475433647633
    },
    {
      "action_query_batch_std": 0.3793018013238907,
      "action_query_norm": 42.71913795471191,
      "action_relation_loss": 0.00394107149913907,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.274087932647333,
      "flow_matching_loss": 0.029117189534008502,
      "grad_norm": 0.11626610904932022,
      "h_stock_valid_norm": 98.6286506652832,
      "invalid_tail_samples": 23.6,
      "learning_rate": 5.856116109205602e-05,
      "loss": 0.0331,
      "positive_teacher_pairs": 26897.6,
      "query_h_norm_ratio": 0.4331334620714188,
      "relation_loss": 0.00394107149913907,
      "selected_pairs": 26897.6,
      "step": 14190,
      "student_distance_mean": 25.097984313964844,
      "teacher_distance_mean": 0.8520578563213348,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.00394107149913907,
      "weighted_relation_loss": 0.00394107149913907
    },
    {
      "action_query_batch_std": 0.37503733932971955,
      "action_query_norm": 42.71057205200195,
      "action_relation_loss": 0.004146958515048027,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.283442469597755,
      "flow_matching_loss": 0.03134654480963946,
      "grad_norm": 0.12446996569633484,
      "h_stock_valid_norm": 98.00945205688477,
      "invalid_tail_samples": 23.1,
      "learning_rate": 5.850685417483983e-05,
      "loss": 0.0355,
      "positive_teacher_pairs": 27012.3,
      "query_h_norm_ratio": 0.4357862055301666,
      "relation_loss": 0.004146958515048027,
      "selected_pairs": 27012.3,
      "step": 14200,
      "student_distance_mean": 24.987438011169434,
      "teacher_distance_mean": 0.8457768499851227,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.004146958515048027,
      "weighted_relation_loss": 0.004146958515048027
    },
    {
      "action_query_batch_std": 0.38091819882392886,
      "action_query_norm": 42.74099617004394,
      "action_relation_loss": 0.004003863222897053,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.292797006548176,
      "flow_matching_loss": 0.030778749287128447,
      "grad_norm": 0.11883891373872757,
      "h_stock_valid_norm": 97.58223190307618,
      "invalid_tail_samples": 25.7,
      "learning_rate": 5.8452536920993684e-05,
      "loss": 0.0348,
      "positive_teacher_pairs": 26422.8,
      "query_h_norm_ratio": 0.4380041569471359,
      "relation_loss": 0.004003863222897053,
      "selected_pairs": 26422.8,
      "step": 14210,
      "student_distance_mean": 25.24767951965332,
      "teacher_distance_mean": 0.8572965562343597,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.004003863222897053,
      "weighted_relation_loss": 0.004003863222897053
    },
    {
      "action_query_batch_std": 0.3801877349615097,
      "action_query_norm": 42.74489860534668,
      "action_relation_loss": 0.00420987259130925,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.302151543498598,
      "flow_matching_loss": 0.0322278393432498,
      "grad_norm": 0.08719993382692337,
      "h_stock_valid_norm": 97.89405670166016,
      "invalid_tail_samples": 24.4,
      "learning_rate": 5.839820939651817e-05,
      "loss": 0.0364,
      "positive_teacher_pairs": 26709.5,
      "query_h_norm_ratio": 0.43664684891700745,
      "relation_loss": 0.00420987259130925,
      "selected_pairs": 26709.5,
      "step": 14220,
      "student_distance_mean": 25.229261207580567,
      "teacher_distance_mean": 0.852701872587204,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.00420987259130925,
      "weighted_relation_loss": 0.00420987259130925
    },
    {
      "action_query_batch_std": 0.37766164243221284,
      "action_query_norm": 42.73088111877441,
      "action_relation_loss": 0.003979831538163126,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.311506080449018,
      "flow_matching_loss": 0.032467514649033545,
      "grad_norm": 0.12443391978740692,
      "h_stock_valid_norm": 97.80407943725587,
      "invalid_tail_samples": 24.5,
      "learning_rate": 5.834387166742632e-05,
      "loss": 0.0364,
      "positive_teacher_pairs": 26697.4,
      "query_h_norm_ratio": 0.4369064956903458,
      "relation_loss": 0.003979831538163126,
      "selected_pairs": 26697.4,
      "step": 14230,
      "student_distance_mean": 25.12683982849121,
      "teacher_distance_mean": 0.850729787349701,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.003979831538163126,
      "weighted_relation_loss": 0.003979831538163126
    },
    {
      "action_query_batch_std": 0.37981181442737577,
      "action_query_norm": 42.72498893737793,
      "action_relation_loss": 0.003998309979215265,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.320860617399438,
      "flow_matching_loss": 0.030016250722110272,
      "grad_norm": 0.10045865178108215,
      "h_stock_valid_norm": 97.8248046875,
      "invalid_tail_samples": 24.7,
      "learning_rate": 5.828952379974364e-05,
      "loss": 0.034,
      "positive_teacher_pairs": 26657.5,
      "query_h_norm_ratio": 0.43675399422645567,
      "relation_loss": 0.003998309979215265,
      "selected_pairs": 26657.5,
      "step": 14240,
      "student_distance_mean": 25.30625686645508,
      "teacher_distance_mean": 0.8542163848876954,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.003998309979215265,
      "weighted_relation_loss": 0.003998309979215265
    },
    {
      "action_query_batch_std": 0.3819379866123199,
      "action_query_norm": 42.73861846923828,
      "action_relation_loss": 0.004075005208142102,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.33021515434986,
      "flow_matching_loss": 0.030907531082630158,
      "grad_norm": 0.10905834287405014,
      "h_stock_valid_norm": 98.00029525756835,
      "invalid_tail_samples": 23.0,
      "learning_rate": 5.8235165859507864e-05,
      "loss": 0.035,
      "positive_teacher_pairs": 27042.5,
      "query_h_norm_ratio": 0.4361088275909424,
      "relation_loss": 0.004075005208142102,
      "selected_pairs": 27042.5,
      "step": 14250,
      "student_distance_mean": 25.405790901184083,
      "teacher_distance_mean": 0.8556453645229339,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.004075005208142102,
      "weighted_relation_loss": 0.004075005208142102
    },
    {
      "action_query_batch_std": 0.3855930596590042,
      "action_query_norm": 42.73585586547851,
      "action_relation_loss": 0.0041948717553168535,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.33956969130028,
      "flow_matching_loss": 0.031395955197513106,
      "grad_norm": 0.10628332197666168,
      "h_stock_valid_norm": 97.70016403198242,
      "invalid_tail_samples": 24.4,
      "learning_rate": 5.818079791276907e-05,
      "loss": 0.0356,
      "positive_teacher_pairs": 26709.1,
      "query_h_norm_ratio": 0.43742395341396334,
      "relation_loss": 0.0041948717553168535,
      "selected_pairs": 26709.1,
      "step": 14260,
      "student_distance_mean": 25.578933906555175,
      "teacher_distance_mean": 0.8586938321590424,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0041948717553168535,
      "weighted_relation_loss": 0.0041948717553168535
    },
    {
      "action_query_batch_std": 0.37838282287120817,
      "action_query_norm": 42.70005645751953,
      "action_relation_loss": 0.004068640223704278,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.348924228250702,
      "flow_matching_loss": 0.031901080161333084,
      "grad_norm": 0.12275238335132599,
      "h_stock_valid_norm": 97.85372543334961,
      "invalid_tail_samples": 26.0,
      "learning_rate": 5.8126420025589415e-05,
      "loss": 0.036,
      "positive_teacher_pairs": 26344.3,
      "query_h_norm_ratio": 0.43637046217918396,
      "relation_loss": 0.004068640223704278,
      "selected_pairs": 26344.3,
      "step": 14270,
      "student_distance_mean": 25.206419372558592,
      "teacher_distance_mean": 0.8525192081928253,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.004068640223704278,
      "weighted_relation_loss": 0.004068640223704278
    },
    {
      "action_query_batch_std": 0.37729853093624116,
      "action_query_norm": 42.704101181030275,
      "action_relation_loss": 0.004022977873682975,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.358278765201122,
      "flow_matching_loss": 0.02923886589705944,
      "grad_norm": 0.13244327902793884,
      "h_stock_valid_norm": 98.22992706298828,
      "invalid_tail_samples": 24.6,
      "learning_rate": 5.807203226404313e-05,
      "loss": 0.0333,
      "positive_teacher_pairs": 26672.0,
      "query_h_norm_ratio": 0.43473970890045166,
      "relation_loss": 0.004022977873682975,
      "selected_pairs": 26672.0,
      "step": 14280,
      "student_distance_mean": 24.985169219970704,
      "teacher_distance_mean": 0.8462829411029815,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.004022977873682975,
      "weighted_relation_loss": 0.004022977873682975
    },
    {
      "action_query_batch_std": 0.3776136487722397,
      "action_query_norm": 42.70916519165039,
      "action_relation_loss": 0.004014997463673353,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.367633302151544,
      "flow_matching_loss": 0.0310683423653245,
      "grad_norm": 0.09624037891626358,
      "h_stock_valid_norm": 99.73563690185547,
      "invalid_tail_samples": 23.8,
      "learning_rate": 5.801763469421652e-05,
      "loss": 0.0351,
      "positive_teacher_pairs": 26849.9,
      "query_h_norm_ratio": 0.42822823524475095,
      "relation_loss": 0.004014997463673353,
      "selected_pairs": 26849.9,
      "step": 14290,
      "student_distance_mean": 25.082570266723632,
      "teacher_distance_mean": 0.8513631343841552,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.004014997463673353,
      "weighted_relation_loss": 0.004014997463673353
    },
    {
      "action_query_batch_std": 0.38134126663208007,
      "action_query_norm": 42.723541259765625,
      "action_relation_loss": 0.004468635842204094,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.376987839101965,
      "flow_matching_loss": 0.031368482112884524,
      "grad_norm": 0.10387951135635376,
      "h_stock_valid_norm": 99.93121337890625,
      "invalid_tail_samples": 23.1,
      "learning_rate": 5.796322738220774e-05,
      "loss": 0.0358,
      "positive_teacher_pairs": 27014.1,
      "query_h_norm_ratio": 0.4275321692228317,
      "relation_loss": 0.004468635842204094,
      "selected_pairs": 27014.1,
      "step": 14300,
      "student_distance_mean": 25.312576293945312,
      "teacher_distance_mean": 0.8548417091369629,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.004468635842204094,
      "weighted_relation_loss": 0.004468635842204094
    },
    {
      "action_query_batch_std": 0.3796809196472168,
      "action_query_norm": 42.712622833251956,
      "action_relation_loss": 0.004080538754351437,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.386342376052385,
      "flow_matching_loss": 0.03111608773469925,
      "grad_norm": 0.096648208796978,
      "h_stock_valid_norm": 99.19405975341797,
      "invalid_tail_samples": 24.6,
      "learning_rate": 5.79088103941268e-05,
      "loss": 0.0352,
      "positive_teacher_pairs": 26665.7,
      "query_h_norm_ratio": 0.43061363101005556,
      "relation_loss": 0.004080538754351437,
      "selected_pairs": 26665.7,
      "step": 14310,
      "student_distance_mean": 25.286020278930664,
      "teacher_distance_mean": 0.8520420730113983,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.004080538754351437,
      "weighted_relation_loss": 0.004080538754351437
    },
    {
      "action_query_batch_std": 0.37261622250080106,
      "action_query_norm": 42.684204483032225,
      "action_relation_loss": 0.004340703622438014,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.395696913002807,
      "flow_matching_loss": 0.03226691409945488,
      "grad_norm": 0.10015710443258286,
      "h_stock_valid_norm": 98.01531753540038,
      "invalid_tail_samples": 25.4,
      "learning_rate": 5.785438379609549e-05,
      "loss": 0.0366,
      "positive_teacher_pairs": 26482.2,
      "query_h_norm_ratio": 0.4354881078004837,
      "relation_loss": 0.004340703622438014,
      "selected_pairs": 26482.2,
      "step": 14320,
      "student_distance_mean": 24.849166870117188,
      "teacher_distance_mean": 0.839301061630249,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.004340703622438014,
      "weighted_relation_loss": 0.004340703622438014
    },
    {
      "action_query_batch_std": 0.3824788600206375,
      "action_query_norm": 42.708247375488284,
      "action_relation_loss": 0.003923977166414261,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.405051449953227,
      "flow_matching_loss": 0.03368337024003267,
      "grad_norm": 0.10617858171463013,
      "h_stock_valid_norm": 98.32524108886719,
      "invalid_tail_samples": 24.8,
      "learning_rate": 5.779994765424724e-05,
      "loss": 0.0376,
      "positive_teacher_pairs": 26630.1,
      "query_h_norm_ratio": 0.43436438739299776,
      "relation_loss": 0.003923977166414261,
      "selected_pairs": 26630.1,
      "step": 14330,
      "student_distance_mean": 25.440981674194337,
      "teacher_distance_mean": 0.8602768898010253,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.003923977166414261,
      "weighted_relation_loss": 0.003923977166414261
    },
    {
      "action_query_batch_std": 0.3865474373102188,
      "action_query_norm": 42.72093238830566,
      "action_relation_loss": 0.0038760997587814925,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.414405986903649,
      "flow_matching_loss": 0.03179781176149845,
      "grad_norm": 0.10314846783876419,
      "h_stock_valid_norm": 98.9732681274414,
      "invalid_tail_samples": 25.1,
      "learning_rate": 5.7745502034727114e-05,
      "loss": 0.0357,
      "positive_teacher_pairs": 26548.8,
      "query_h_norm_ratio": 0.4316464692354202,
      "relation_loss": 0.0038760997587814925,
      "selected_pairs": 26548.8,
      "step": 14340,
      "student_distance_mean": 25.57351360321045,
      "teacher_distance_mean": 0.8622537851333618,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0038760997587814925,
      "weighted_relation_loss": 0.0038760997587814925
    },
    {
      "action_query_batch_std": 0.37979419231414796,
      "action_query_norm": 42.69493141174316,
      "action_relation_loss": 0.004216834087856114,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.42376052385407,
      "flow_matching_loss": 0.034712385199964044,
      "grad_norm": 0.110391765832901,
      "h_stock_valid_norm": 98.51459274291992,
      "invalid_tail_samples": 23.8,
      "learning_rate": 5.769104700369164e-05,
      "loss": 0.0389,
      "positive_teacher_pairs": 26848.3,
      "query_h_norm_ratio": 0.4333968997001648,
      "relation_loss": 0.004216834087856114,
      "selected_pairs": 26848.3,
      "step": 14350,
      "student_distance_mean": 25.25127830505371,
      "teacher_distance_mean": 0.8480804145336152,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.004216834087856114,
      "weighted_relation_loss": 0.004216834087856114
    },
    {
      "action_query_batch_std": 0.3826272636651993,
      "action_query_norm": 42.70637512207031,
      "action_relation_loss": 0.004235544963739813,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.43311506080449,
      "flow_matching_loss": 0.03382822573184967,
      "grad_norm": 0.11889579892158508,
      "h_stock_valid_norm": 97.77967224121093,
      "invalid_tail_samples": 24.7,
      "learning_rate": 5.7636582627308854e-05,
      "loss": 0.0381,
      "positive_teacher_pairs": 26644.8,
      "query_h_norm_ratio": 0.43676767945289613,
      "relation_loss": 0.004235544963739813,
      "selected_pairs": 26644.8,
      "step": 14360,
      "student_distance_mean": 25.477202796936034,
      "teacher_distance_mean": 0.8597558081150055,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.004235544963739813,
      "weighted_relation_loss": 0.004235544963739813
    },
    {
      "action_query_batch_std": 0.3839752316474915,
      "action_query_norm": 42.7055248260498,
      "action_relation_loss": 0.004164799419231713,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.442469597754911,
      "flow_matching_loss": 0.03303837701678276,
      "grad_norm": 0.10501924902200699,
      "h_stock_valid_norm": 98.02240524291992,
      "invalid_tail_samples": 26.7,
      "learning_rate": 5.7582108971758095e-05,
      "loss": 0.0372,
      "positive_teacher_pairs": 26184.2,
      "query_h_norm_ratio": 0.4356755793094635,
      "relation_loss": 0.004164799419231713,
      "selected_pairs": 26184.2,
      "step": 14370,
      "student_distance_mean": 25.500958061218263,
      "teacher_distance_mean": 0.8574795246124267,
      "valid_action_samples": 229.3,
      "weighted_auxiliary_loss": 0.004164799419231713,
      "weighted_relation_loss": 0.004164799419231713
    },
    {
      "action_query_batch_std": 0.3795935809612274,
      "action_query_norm": 42.70065422058106,
      "action_relation_loss": 0.004111209115944803,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.451824134705332,
      "flow_matching_loss": 0.03267585299909115,
      "grad_norm": 0.09813182801008224,
      "h_stock_valid_norm": 98.53035430908203,
      "invalid_tail_samples": 24.0,
      "learning_rate": 5.752762610322999e-05,
      "loss": 0.0368,
      "positive_teacher_pairs": 26801.9,
      "query_h_norm_ratio": 0.4333785504102707,
      "relation_loss": 0.004111209115944803,
      "selected_pairs": 26801.9,
      "step": 14380,
      "student_distance_mean": 25.295620918273926,
      "teacher_distance_mean": 0.8519717991352082,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.004111209115944803,
      "weighted_relation_loss": 0.004111209115944803
    },
    {
      "action_query_batch_std": 0.37599282562732694,
      "action_query_norm": 42.67771110534668,
      "action_relation_loss": 0.004183194180950522,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.461178671655754,
      "flow_matching_loss": 0.030363604798913,
      "grad_norm": 0.11249705404043198,
      "h_stock_valid_norm": 98.98442077636719,
      "invalid_tail_samples": 24.3,
      "learning_rate": 5.747313408792636e-05,
      "loss": 0.0345,
      "positive_teacher_pairs": 26741.5,
      "query_h_norm_ratio": 0.4311590701341629,
      "relation_loss": 0.004183194180950522,
      "selected_pairs": 26741.5,
      "step": 14390,
      "student_distance_mean": 24.892357635498048,
      "teacher_distance_mean": 0.8407695353031158,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.004183194180950522,
      "weighted_relation_loss": 0.004183194180950522
    },
    {
      "action_query_batch_std": 0.37762676179409027,
      "action_query_norm": 42.682978820800784,
      "action_relation_loss": 0.004202794656157494,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.470533208606174,
      "flow_matching_loss": 0.03190798256546259,
      "grad_norm": 0.1257057785987854,
      "h_stock_valid_norm": 98.6835090637207,
      "invalid_tail_samples": 25.1,
      "learning_rate": 5.741863299206014e-05,
      "loss": 0.0361,
      "positive_teacher_pairs": 26550.0,
      "query_h_norm_ratio": 0.4325269490480423,
      "relation_loss": 0.004202794656157494,
      "selected_pairs": 26550.0,
      "step": 14400,
      "student_distance_mean": 25.171273040771485,
      "teacher_distance_mean": 0.8502983033657074,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.004202794656157494,
      "weighted_relation_loss": 0.004202794656157494
    },
    {
      "action_query_batch_std": 0.3814500838518143,
      "action_query_norm": 42.69129257202148,
      "action_relation_loss": 0.00409954315982759,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.479887745556596,
      "flow_matching_loss": 0.030109640769660474,
      "grad_norm": 0.11047481000423431,
      "h_stock_valid_norm": 98.32665252685547,
      "invalid_tail_samples": 24.2,
      "learning_rate": 5.73641228818553e-05,
      "loss": 0.0342,
      "positive_teacher_pairs": 26754.5,
      "query_h_norm_ratio": 0.4341865211725235,
      "relation_loss": 0.00409954315982759,
      "selected_pairs": 26754.5,
      "step": 14410,
      "student_distance_mean": 25.444074440002442,
      "teacher_distance_mean": 0.8600599825382232,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.00409954315982759,
      "weighted_relation_loss": 0.00409954315982759
    },
    {
      "action_query_batch_std": 0.3780610501766205,
      "action_query_norm": 42.68629531860351,
      "action_relation_loss": 0.004057821095921099,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.489242282507016,
      "flow_matching_loss": 0.031785582564771174,
      "grad_norm": 0.1210227832198143,
      "h_stock_valid_norm": 97.73602905273438,
      "invalid_tail_samples": 21.2,
      "learning_rate": 5.730960382354676e-05,
      "loss": 0.0358,
      "positive_teacher_pairs": 27454.1,
      "query_h_norm_ratio": 0.4367534339427948,
      "relation_loss": 0.004057821095921099,
      "selected_pairs": 27454.1,
      "step": 14420,
      "student_distance_mean": 25.128313636779787,
      "teacher_distance_mean": 0.8504867374897003,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.004057821095921099,
      "weighted_relation_loss": 0.004057821095921099
    },
    {
      "action_query_batch_std": 0.3821741908788681,
      "action_query_norm": 42.70196266174317,
      "action_relation_loss": 0.004074286343529821,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.498596819457436,
      "flow_matching_loss": 0.03143173363059759,
      "grad_norm": 0.15481668710708618,
      "h_stock_valid_norm": 97.79127960205078,
      "invalid_tail_samples": 21.6,
      "learning_rate": 5.725507588338035e-05,
      "loss": 0.0355,
      "positive_teacher_pairs": 27368.8,
      "query_h_norm_ratio": 0.4366692274808884,
      "relation_loss": 0.004074286343529821,
      "selected_pairs": 27368.8,
      "step": 14430,
      "student_distance_mean": 25.548038673400878,
      "teacher_distance_mean": 0.8623886644840241,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.004074286343529821,
      "weighted_relation_loss": 0.004074286343529821
    },
    {
      "action_query_batch_std": 0.3798131436109543,
      "action_query_norm": 42.683570098876956,
      "action_relation_loss": 0.003951015323400498,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.507951356407858,
      "flow_matching_loss": 0.029779767245054246,
      "grad_norm": 0.1037169098854065,
      "h_stock_valid_norm": 98.17978210449219,
      "invalid_tail_samples": 22.4,
      "learning_rate": 5.7200539127612604e-05,
      "loss": 0.0337,
      "positive_teacher_pairs": 27182.5,
      "query_h_norm_ratio": 0.43475296795368196,
      "relation_loss": 0.003951015323400498,
      "selected_pairs": 27182.5,
      "step": 14440,
      "student_distance_mean": 25.293903160095216,
      "teacher_distance_mean": 0.8531279981136322,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.003951015323400498,
      "weighted_relation_loss": 0.003951015323400498
    },
    {
      "action_query_batch_std": 0.3766514629125595,
      "action_query_norm": 42.67123680114746,
      "action_relation_loss": 0.003948879218660295,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.517305893358278,
      "flow_matching_loss": 0.0320184376090765,
      "grad_norm": 0.12151330709457397,
      "h_stock_valid_norm": 98.4142692565918,
      "invalid_tail_samples": 22.9,
      "learning_rate": 5.714599362251087e-05,
      "loss": 0.036,
      "positive_teacher_pairs": 27060.6,
      "query_h_norm_ratio": 0.4335918664932251,
      "relation_loss": 0.003948879218660295,
      "selected_pairs": 27060.6,
      "step": 14450,
      "student_distance_mean": 25.05566120147705,
      "teacher_distance_mean": 0.8528855621814728,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.003948879218660295,
      "weighted_relation_loss": 0.003948879218660295
    },
    {
      "action_query_batch_std": 0.3829198688268661,
      "action_query_norm": 42.68204956054687,
      "action_relation_loss": 0.0040169295622035864,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.5266604303087,
      "flow_matching_loss": 0.032448678463697436,
      "grad_norm": 0.10354266315698624,
      "h_stock_valid_norm": 98.44103546142578,
      "invalid_tail_samples": 22.7,
      "learning_rate": 5.709143943435307e-05,
      "loss": 0.0365,
      "positive_teacher_pairs": 27106.4,
      "query_h_norm_ratio": 0.43358350694179537,
      "relation_loss": 0.0040169295622035864,
      "selected_pairs": 27106.4,
      "step": 14460,
      "student_distance_mean": 25.422306060791016,
      "teacher_distance_mean": 0.8628775477409363,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0040169295622035864,
      "weighted_relation_loss": 0.0040169295622035864
    },
    {
      "action_query_batch_std": 0.3834658205509186,
      "action_query_norm": 42.682287979125974,
      "action_relation_loss": 0.0038932378869503737,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.53601496725912,
      "flow_matching_loss": 0.03007687013596296,
      "grad_norm": 0.10421936213970184,
      "h_stock_valid_norm": 98.36843032836914,
      "invalid_tail_samples": 24.7,
      "learning_rate": 5.7036876629427646e-05,
      "loss": 0.034,
      "positive_teacher_pairs": 26642.5,
      "query_h_norm_ratio": 0.4339043617248535,
      "relation_loss": 0.0038932378869503737,
      "selected_pairs": 26642.5,
      "step": 14470,
      "student_distance_mean": 25.492753410339354,
      "teacher_distance_mean": 0.8593600571155549,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0038932378869503737,
      "weighted_relation_loss": 0.0038932378869503737
    },
    {
      "action_query_batch_std": 0.38351736664772035,
      "action_query_norm": 42.67696342468262,
      "action_relation_loss": 0.0038440991193056107,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.545369504209543,
      "flow_matching_loss": 0.03014751337468624,
      "grad_norm": 0.1356639564037323,
      "h_stock_valid_norm": 98.39685668945313,
      "invalid_tail_samples": 23.7,
      "learning_rate": 5.698230527403361e-05,
      "loss": 0.034,
      "positive_teacher_pairs": 26872.1,
      "query_h_norm_ratio": 0.43372493982315063,
      "relation_loss": 0.0038440991193056107,
      "selected_pairs": 26872.1,
      "step": 14480,
      "student_distance_mean": 25.485217094421387,
      "teacher_distance_mean": 0.8605400860309601,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0038440991193056107,
      "weighted_relation_loss": 0.0038440991193056107
    },
    {
      "action_query_batch_std": 0.37463952600955963,
      "action_query_norm": 42.64250144958496,
      "action_relation_loss": 0.004076963709667325,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.554724041159963,
      "flow_matching_loss": 0.03210950512439013,
      "grad_norm": 0.11694351583719254,
      "h_stock_valid_norm": 98.84956665039063,
      "invalid_tail_samples": 25.4,
      "learning_rate": 5.6927725434480274e-05,
      "loss": 0.0362,
      "positive_teacher_pairs": 26482.2,
      "query_h_norm_ratio": 0.43139185905456545,
      "relation_loss": 0.004076963709667325,
      "selected_pairs": 26482.2,
      "step": 14490,
      "student_distance_mean": 24.947858238220213,
      "teacher_distance_mean": 0.8495809257030487,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.004076963709667325,
      "weighted_relation_loss": 0.004076963709667325
    },
    {
      "action_query_batch_std": 0.38157479763031005,
      "action_query_norm": 42.66764907836914,
      "action_relation_loss": 0.004160483949817717,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.564078578110383,
      "flow_matching_loss": 0.029674748331308363,
      "grad_norm": 0.1092299073934555,
      "h_stock_valid_norm": 98.92423477172852,
      "invalid_tail_samples": 25.2,
      "learning_rate": 5.687313717708728e-05,
      "loss": 0.0338,
      "positive_teacher_pairs": 26522.0,
      "query_h_norm_ratio": 0.4313198536634445,
      "relation_loss": 0.004160483949817717,
      "selected_pairs": 26522.0,
      "step": 14500,
      "student_distance_mean": 25.370611000061036,
      "teacher_distance_mean": 0.860386210680008,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.004160483949817717,
      "weighted_relation_loss": 0.004160483949817717
    },
    {
      "action_query_batch_std": 0.37932281494140624,
      "action_query_norm": 42.666849517822264,
      "action_relation_loss": 0.004165376536548138,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.573433115060805,
      "flow_matching_loss": 0.03020363375544548,
      "grad_norm": 0.13909998536109924,
      "h_stock_valid_norm": 98.46456451416016,
      "invalid_tail_samples": 23.5,
      "learning_rate": 5.681854056818453e-05,
      "loss": 0.0344,
      "positive_teacher_pairs": 26919.3,
      "query_h_norm_ratio": 0.4333244293928146,
      "relation_loss": 0.004165376536548138,
      "selected_pairs": 26919.3,
      "step": 14510,
      "student_distance_mean": 25.335009384155274,
      "teacher_distance_mean": 0.8555406391620636,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.004165376536548138,
      "weighted_relation_loss": 0.004165376536548138
    },
    {
      "action_query_batch_std": 0.3799301773309708,
      "action_query_norm": 42.65462455749512,
      "action_relation_loss": 0.004016415914520621,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.582787652011225,
      "flow_matching_loss": 0.031078390963375568,
      "grad_norm": 0.11400794982910156,
      "h_stock_valid_norm": 98.1261085510254,
      "invalid_tail_samples": 21.1,
      "learning_rate": 5.676393567411205e-05,
      "loss": 0.0351,
      "positive_teacher_pairs": 27484.0,
      "query_h_norm_ratio": 0.4346974313259125,
      "relation_loss": 0.004016415914520621,
      "selected_pairs": 27484.0,
      "step": 14520,
      "student_distance_mean": 25.166683387756347,
      "teacher_distance_mean": 0.85482457280159,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.004016415914520621,
      "weighted_relation_loss": 0.004016415914520621
    },
    {
      "action_query_batch_std": 0.37952964305877684,
      "action_query_norm": 42.65754280090332,
      "action_relation_loss": 0.00393968285061419,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.592142188961647,
      "flow_matching_loss": 0.03162710219621658,
      "grad_norm": 0.11953730881214142,
      "h_stock_valid_norm": 98.11834487915038,
      "invalid_tail_samples": 23.4,
      "learning_rate": 5.670932256121991e-05,
      "loss": 0.0356,
      "positive_teacher_pairs": 26943.1,
      "query_h_norm_ratio": 0.4347583055496216,
      "relation_loss": 0.00393968285061419,
      "selected_pairs": 26943.1,
      "step": 14530,
      "student_distance_mean": 25.13713550567627,
      "teacher_distance_mean": 0.8541157066822052,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.00393968285061419,
      "weighted_relation_loss": 0.00393968285061419
    },
    {
      "action_query_batch_std": 0.3790106028318405,
      "action_query_norm": 42.65063972473145,
      "action_relation_loss": 0.003866756893694401,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.601496725912067,
      "flow_matching_loss": 0.03185644615441561,
      "grad_norm": 0.1145930215716362,
      "h_stock_valid_norm": 97.69853591918945,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.6654701295868216e-05,
      "loss": 0.0357,
      "positive_teacher_pairs": 26988.0,
      "query_h_norm_ratio": 0.4365565270185471,
      "relation_loss": 0.003866756893694401,
      "selected_pairs": 26988.0,
      "step": 14540,
      "student_distance_mean": 25.18355484008789,
      "teacher_distance_mean": 0.8585104525089264,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.003866756893694401,
      "weighted_relation_loss": 0.003866756893694401
    },
    {
      "action_query_batch_std": 0.37690978646278384,
      "action_query_norm": 42.64102973937988,
      "action_relation_loss": 0.0038906377973034977,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.610851262862488,
      "flow_matching_loss": 0.031087880954146385,
      "grad_norm": 0.10554026812314987,
      "h_stock_valid_norm": 97.69451293945312,
      "invalid_tail_samples": 23.5,
      "learning_rate": 5.660007194442697e-05,
      "loss": 0.035,
      "positive_teacher_pairs": 26924.0,
      "query_h_norm_ratio": 0.43647532165050507,
      "relation_loss": 0.0038906377973034977,
      "selected_pairs": 26924.0,
      "step": 14550,
      "student_distance_mean": 24.92318935394287,
      "teacher_distance_mean": 0.8512810289859771,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0038906377973034977,
      "weighted_relation_loss": 0.0038906377973034977
    },
    {
      "action_query_batch_std": 0.3765942543745041,
      "action_query_norm": 42.6277214050293,
      "action_relation_loss": 0.004008318902924657,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.62020579981291,
      "flow_matching_loss": 0.0319825304672122,
      "grad_norm": 0.11054299771785736,
      "h_stock_valid_norm": 97.5965560913086,
      "invalid_tail_samples": 21.6,
      "learning_rate": 5.6545434573275946e-05,
      "loss": 0.036,
      "positive_teacher_pairs": 27361.7,
      "query_h_norm_ratio": 0.4367784678936005,
      "relation_loss": 0.004008318902924657,
      "selected_pairs": 27361.7,
      "step": 14560,
      "student_distance_mean": 24.84796600341797,
      "teacher_distance_mean": 0.8415048301219941,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.004008318902924657,
      "weighted_relation_loss": 0.004008318902924657
    },
    {
      "action_query_batch_std": 0.38142114579677583,
      "action_query_norm": 42.64916877746582,
      "action_relation_loss": 0.0039009607164189218,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.62956033676333,
      "flow_matching_loss": 0.03146727755665779,
      "grad_norm": 0.09810826182365417,
      "h_stock_valid_norm": 97.83719940185547,
      "invalid_tail_samples": 22.9,
      "learning_rate": 5.649078924880472e-05,
      "loss": 0.0354,
      "positive_teacher_pairs": 27062.5,
      "query_h_norm_ratio": 0.4359233617782593,
      "relation_loss": 0.0039009607164189218,
      "selected_pairs": 27062.5,
      "step": 14570,
      "student_distance_mean": 25.34578800201416,
      "teacher_distance_mean": 0.8600736439228058,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0039009607164189218,
      "weighted_relation_loss": 0.0039009607164189218
    },
    {
      "action_query_batch_std": 0.383009535074234,
      "action_query_norm": 42.653992080688475,
      "action_relation_loss": 0.003910104208625853,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.638914873713752,
      "flow_matching_loss": 0.02950845118612051,
      "grad_norm": 0.09993851184844971,
      "h_stock_valid_norm": 98.27425918579101,
      "invalid_tail_samples": 24.0,
      "learning_rate": 5.643613603741252e-05,
      "loss": 0.0334,
      "positive_teacher_pairs": 26802.0,
      "query_h_norm_ratio": 0.4340342223644257,
      "relation_loss": 0.003910104208625853,
      "selected_pairs": 26802.0,
      "step": 14580,
      "student_distance_mean": 25.38444881439209,
      "teacher_distance_mean": 0.8629280030727386,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.003910104208625853,
      "weighted_relation_loss": 0.003910104208625853
    },
    {
      "action_query_batch_std": 0.38184078633785246,
      "action_query_norm": 42.64446792602539,
      "action_relation_loss": 0.004019231698475778,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.648269410664172,
      "flow_matching_loss": 0.03277768604457378,
      "grad_norm": 0.09755215793848038,
      "h_stock_valid_norm": 98.4427490234375,
      "invalid_tail_samples": 24.5,
      "learning_rate": 5.6381475005508154e-05,
      "loss": 0.0368,
      "positive_teacher_pairs": 26687.4,
      "query_h_norm_ratio": 0.43319287002086637,
      "relation_loss": 0.004019231698475778,
      "selected_pairs": 26687.4,
      "step": 14590,
      "student_distance_mean": 25.263082695007324,
      "teacher_distance_mean": 0.8615807473659516,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.004019231698475778,
      "weighted_relation_loss": 0.004019231698475778
    },
    {
      "action_query_batch_std": 0.373286646604538,
      "action_query_norm": 42.60293083190918,
      "action_relation_loss": 0.004146040882915258,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.657623947614592,
      "flow_matching_loss": 0.030308989994227886,
      "grad_norm": 0.10432425886392593,
      "h_stock_valid_norm": 98.1406967163086,
      "invalid_tail_samples": 24.9,
      "learning_rate": 5.632680621950992e-05,
      "loss": 0.0345,
      "positive_teacher_pairs": 26598.0,
      "query_h_norm_ratio": 0.4341027945280075,
      "relation_loss": 0.004146040882915258,
      "selected_pairs": 26598.0,
      "step": 14600,
      "student_distance_mean": 24.8875,
      "teacher_distance_mean": 0.8482327103614807,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.004146040882915258,
      "weighted_relation_loss": 0.004146040882915258
    },
    {
      "action_query_batch_std": 0.3759860426187515,
      "action_query_norm": 42.61272621154785,
      "action_relation_loss": 0.003944731596857309,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.666978484565014,
      "flow_matching_loss": 0.030049571953713893,
      "grad_norm": 0.09735627472400665,
      "h_stock_valid_norm": 97.96486968994141,
      "invalid_tail_samples": 25.6,
      "learning_rate": 5.627212974584555e-05,
      "loss": 0.034,
      "positive_teacher_pairs": 26434.8,
      "query_h_norm_ratio": 0.4349824607372284,
      "relation_loss": 0.003944731596857309,
      "selected_pairs": 26434.8,
      "step": 14610,
      "student_distance_mean": 24.964314651489257,
      "teacher_distance_mean": 0.8542099952697754,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.003944731596857309,
      "weighted_relation_loss": 0.003944731596857309
    },
    {
      "action_query_batch_std": 0.37219436168670655,
      "action_query_norm": 42.59302406311035,
      "action_relation_loss": 0.004019083664752543,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.676333021515434,
      "flow_matching_loss": 0.030325566232204438,
      "grad_norm": 0.11117154359817505,
      "h_stock_valid_norm": 98.1593147277832,
      "invalid_tail_samples": 24.4,
      "learning_rate": 5.62174456509521e-05,
      "loss": 0.0343,
      "positive_teacher_pairs": 26718.9,
      "query_h_norm_ratio": 0.43392190635204314,
      "relation_loss": 0.004019083664752543,
      "selected_pairs": 26718.9,
      "step": 14620,
      "student_distance_mean": 24.763465881347656,
      "teacher_distance_mean": 0.8434093952178955,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.004019083664752543,
      "weighted_relation_loss": 0.004019083664752543
    },
    {
      "action_query_batch_std": 0.37571848928928375,
      "action_query_norm": 42.61058082580566,
      "action_relation_loss": 0.004104083520360291,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.685687558465856,
      "flow_matching_loss": 0.031726916506886485,
      "grad_norm": 0.09356170147657394,
      "h_stock_valid_norm": 97.78594131469727,
      "invalid_tail_samples": 24.1,
      "learning_rate": 5.616275400127594e-05,
      "loss": 0.0358,
      "positive_teacher_pairs": 26794.1,
      "query_h_norm_ratio": 0.4357563227415085,
      "relation_loss": 0.004104083520360291,
      "selected_pairs": 26794.1,
      "step": 14630,
      "student_distance_mean": 24.978462791442873,
      "teacher_distance_mean": 0.8515236794948577,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.004104083520360291,
      "weighted_relation_loss": 0.004104083520360291
    },
    {
      "action_query_batch_std": 0.37627215385437013,
      "action_query_norm": 42.60231513977051,
      "action_relation_loss": 0.00401496971026063,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.695042095416277,
      "flow_matching_loss": 0.029418940469622613,
      "grad_norm": 0.10300067812204361,
      "h_stock_valid_norm": 97.84515228271485,
      "invalid_tail_samples": 23.0,
      "learning_rate": 5.610805486327254e-05,
      "loss": 0.0334,
      "positive_teacher_pairs": 27032.2,
      "query_h_norm_ratio": 0.43540928363800047,
      "relation_loss": 0.00401496971026063,
      "selected_pairs": 27032.2,
      "step": 14640,
      "student_distance_mean": 24.994792556762697,
      "teacher_distance_mean": 0.8551907241344452,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.00401496971026063,
      "weighted_relation_loss": 0.00401496971026063
    },
    {
      "action_query_batch_std": 0.37932526469230654,
      "action_query_norm": 42.6189826965332,
      "action_relation_loss": 0.00405665363650769,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.704396632366699,
      "flow_matching_loss": 0.033020464889705184,
      "grad_norm": 0.1314639002084732,
      "h_stock_valid_norm": 98.41036834716797,
      "invalid_tail_samples": 21.0,
      "learning_rate": 5.6053348303406536e-05,
      "loss": 0.0371,
      "positive_teacher_pairs": 27504.2,
      "query_h_norm_ratio": 0.43307665884494784,
      "relation_loss": 0.00405665363650769,
      "selected_pairs": 27504.2,
      "step": 14650,
      "student_distance_mean": 25.20284538269043,
      "teacher_distance_mean": 0.861364483833313,
      "valid_action_samples": 235.0,
      "weighted_auxiliary_loss": 0.00405665363650769,
      "weighted_relation_loss": 0.00405665363650769
    },
    {
      "action_query_batch_std": 0.37175511121749877,
      "action_query_norm": 42.582458114624025,
      "action_relation_loss": 0.004049248597584665,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.713751169317119,
      "flow_matching_loss": 0.031564096361398696,
      "grad_norm": 0.15771202743053436,
      "h_stock_valid_norm": 98.43310546875,
      "invalid_tail_samples": 23.4,
      "learning_rate": 5.599863438815156e-05,
      "loss": 0.0356,
      "positive_teacher_pairs": 26939.2,
      "query_h_norm_ratio": 0.4326076239347458,
      "relation_loss": 0.004049248597584665,
      "selected_pairs": 26939.2,
      "step": 14660,
      "student_distance_mean": 24.69018383026123,
      "teacher_distance_mean": 0.840383893251419,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.004049248597584665,
      "weighted_relation_loss": 0.004049248597584665
    },
    {
      "action_query_batch_std": 0.3808782160282135,
      "action_query_norm": 42.62172660827637,
      "action_relation_loss": 0.004063699650578201,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.723105706267539,
      "flow_matching_loss": 0.029690783470869064,
      "grad_norm": 0.11626261472702026,
      "h_stock_valid_norm": 99.02395095825196,
      "invalid_tail_samples": 25.3,
      "learning_rate": 5.594391318399017e-05,
      "loss": 0.0338,
      "positive_teacher_pairs": 26504.1,
      "query_h_norm_ratio": 0.4304227650165558,
      "relation_loss": 0.004063699650578201,
      "selected_pairs": 26504.1,
      "step": 14670,
      "student_distance_mean": 25.30157470703125,
      "teacher_distance_mean": 0.8638604760169983,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.004063699650578201,
      "weighted_relation_loss": 0.004063699650578201
    },
    {
      "action_query_batch_std": 0.3790983557701111,
      "action_query_norm": 42.621184539794925,
      "action_relation_loss": 0.003950757184065878,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.732460243217961,
      "flow_matching_loss": 0.03215807303786278,
      "grad_norm": 0.1063510999083519,
      "h_stock_valid_norm": 98.53468170166016,
      "invalid_tail_samples": 21.8,
      "learning_rate": 5.588918475741377e-05,
      "loss": 0.0361,
      "positive_teacher_pairs": 27316.3,
      "query_h_norm_ratio": 0.43255439698696135,
      "relation_loss": 0.003950757184065878,
      "selected_pairs": 27316.3,
      "step": 14680,
      "student_distance_mean": 25.16148433685303,
      "teacher_distance_mean": 0.8576387822628021,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.003950757184065878,
      "weighted_relation_loss": 0.003950757184065878
    },
    {
      "action_query_batch_std": 0.3769588083028793,
      "action_query_norm": 42.60458679199219,
      "action_relation_loss": 0.003871488617733121,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.741814780168381,
      "flow_matching_loss": 0.03127317186444998,
      "grad_norm": 0.11357185989618301,
      "h_stock_valid_norm": 98.3705841064453,
      "invalid_tail_samples": 26.1,
      "learning_rate": 5.5834449174922586e-05,
      "loss": 0.0351,
      "positive_teacher_pairs": 26320.7,
      "query_h_norm_ratio": 0.43310761749744414,
      "relation_loss": 0.003871488617733121,
      "selected_pairs": 26320.7,
      "step": 14690,
      "student_distance_mean": 25.220956230163573,
      "teacher_distance_mean": 0.85689976811409,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.003871488617733121,
      "weighted_relation_loss": 0.003871488617733121
    },
    {
      "action_query_batch_std": 0.3815200746059418,
      "action_query_norm": 42.62214279174805,
      "action_relation_loss": 0.0037895274348556996,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.751169317118803,
      "flow_matching_loss": 0.029440901055932046,
      "grad_norm": 0.12246867269277573,
      "h_stock_valid_norm": 98.15327758789063,
      "invalid_tail_samples": 23.7,
      "learning_rate": 5.57797065030255e-05,
      "loss": 0.0332,
      "positive_teacher_pairs": 26871.1,
      "query_h_norm_ratio": 0.43424144089221955,
      "relation_loss": 0.0037895274348556996,
      "selected_pairs": 26871.1,
      "step": 14700,
      "student_distance_mean": 25.382252883911132,
      "teacher_distance_mean": 0.8647200524806976,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0037895274348556996,
      "weighted_relation_loss": 0.0037895274348556996
    },
    {
      "action_query_batch_std": 0.3721540838479996,
      "action_query_norm": 42.58729019165039,
      "action_relation_loss": 0.003952292213216424,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.760523854069223,
      "flow_matching_loss": 0.03267661016434431,
      "grad_norm": 0.11211976408958435,
      "h_stock_valid_norm": 98.22600402832032,
      "invalid_tail_samples": 22.5,
      "learning_rate": 5.5724956808240016e-05,
      "loss": 0.0366,
      "positive_teacher_pairs": 27149.8,
      "query_h_norm_ratio": 0.4335668355226517,
      "relation_loss": 0.003952292213216424,
      "selected_pairs": 27149.8,
      "step": 14710,
      "student_distance_mean": 24.779401969909667,
      "teacher_distance_mean": 0.8489523470401764,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.003952292213216424,
      "weighted_relation_loss": 0.003952292213216424
    },
    {
      "action_query_batch_std": 0.3759277492761612,
      "action_query_norm": 42.59279518127441,
      "action_relation_loss": 0.0038116085110232235,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.769878391019645,
      "flow_matching_loss": 0.030717653408646583,
      "grad_norm": 0.08346795290708542,
      "h_stock_valid_norm": 98.5160140991211,
      "invalid_tail_samples": 22.2,
      "learning_rate": 5.567020015709219e-05,
      "loss": 0.0345,
      "positive_teacher_pairs": 27228.4,
      "query_h_norm_ratio": 0.4323477476835251,
      "relation_loss": 0.0038116085110232235,
      "selected_pairs": 27228.4,
      "step": 14720,
      "student_distance_mean": 24.974997901916502,
      "teacher_distance_mean": 0.8592362225055694,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0038116085110232235,
      "weighted_relation_loss": 0.0038116085110232235
    },
    {
      "action_query_batch_std": 0.3796753823757172,
      "action_query_norm": 42.60310516357422,
      "action_relation_loss": 0.004057801514863968,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.779232927970066,
      "flow_matching_loss": 0.029841978475451468,
      "grad_norm": 0.07770033925771713,
      "h_stock_valid_norm": 98.55707626342773,
      "invalid_tail_samples": 20.2,
      "learning_rate": 5.561543661611649e-05,
      "loss": 0.0339,
      "positive_teacher_pairs": 27686.0,
      "query_h_norm_ratio": 0.4322720915079117,
      "relation_loss": 0.004057801514863968,
      "selected_pairs": 27686.0,
      "step": 14730,
      "student_distance_mean": 25.131147956848146,
      "teacher_distance_mean": 0.8588516235351562,
      "valid_action_samples": 235.8,
      "weighted_auxiliary_loss": 0.004057801514863968,
      "weighted_relation_loss": 0.004057801514863968
    },
    {
      "action_query_batch_std": 0.3740797996520996,
      "action_query_norm": 42.58620948791504,
      "action_relation_loss": 0.004044030955992639,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.788587464920486,
      "flow_matching_loss": 0.03149729706346989,
      "grad_norm": 0.09370347112417221,
      "h_stock_valid_norm": 98.34171829223632,
      "invalid_tail_samples": 21.2,
      "learning_rate": 5.556066625185583e-05,
      "loss": 0.0355,
      "positive_teacher_pairs": 27458.2,
      "query_h_norm_ratio": 0.43304415941238406,
      "relation_loss": 0.004044030955992639,
      "selected_pairs": 27458.2,
      "step": 14740,
      "student_distance_mean": 24.905390357971193,
      "teacher_distance_mean": 0.848674887418747,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.004044030955992639,
      "weighted_relation_loss": 0.004044030955992639
    },
    {
      "action_query_batch_std": 0.37883247137069703,
      "action_query_norm": 42.59980773925781,
      "action_relation_loss": 0.0038074547657743095,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.797942001870908,
      "flow_matching_loss": 0.03235530722886324,
      "grad_norm": 0.0991986021399498,
      "h_stock_valid_norm": 97.84059295654296,
      "invalid_tail_samples": 22.9,
      "learning_rate": 5.550588913086131e-05,
      "loss": 0.0362,
      "positive_teacher_pairs": 27069.3,
      "query_h_norm_ratio": 0.4354031324386597,
      "relation_loss": 0.0038074547657743095,
      "selected_pairs": 27069.3,
      "step": 14750,
      "student_distance_mean": 25.199827575683592,
      "teacher_distance_mean": 0.8612845838069916,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0038074547657743095,
      "weighted_relation_loss": 0.0038074547657743095
    },
    {
      "action_query_batch_std": 0.3810226798057556,
      "action_query_norm": 42.614342498779294,
      "action_relation_loss": 0.003751996858045459,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.807296538821328,
      "flow_matching_loss": 0.03016116227954626,
      "grad_norm": 0.10166782885789871,
      "h_stock_valid_norm": 97.74458236694336,
      "invalid_tail_samples": 21.2,
      "learning_rate": 5.545110531969234e-05,
      "loss": 0.0339,
      "positive_teacher_pairs": 27461.8,
      "query_h_norm_ratio": 0.4359781086444855,
      "relation_loss": 0.003751996858045459,
      "selected_pairs": 27461.8,
      "step": 14760,
      "student_distance_mean": 25.263583183288574,
      "teacher_distance_mean": 0.8621920943260193,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.003751996858045459,
      "weighted_relation_loss": 0.003751996858045459
    },
    {
      "action_query_batch_std": 0.37570456564426424,
      "action_query_norm": 42.57607231140137,
      "action_relation_loss": 0.004004336893558502,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.81665107577175,
      "flow_matching_loss": 0.031115610525012015,
      "grad_norm": 0.1389419436454773,
      "h_stock_valid_norm": 98.10414123535156,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.539631488491641e-05,
      "loss": 0.0351,
      "positive_teacher_pairs": 26993.7,
      "query_h_norm_ratio": 0.4339912384748459,
      "relation_loss": 0.004004336893558502,
      "selected_pairs": 26993.7,
      "step": 14770,
      "student_distance_mean": 24.979058456420898,
      "teacher_distance_mean": 0.8523908197879791,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.004004336893558502,
      "weighted_relation_loss": 0.004004336893558502
    },
    {
      "action_query_batch_std": 0.3802566736936569,
      "action_query_norm": 42.60743408203125,
      "action_relation_loss": 0.0038948662113398314,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.82600561272217,
      "flow_matching_loss": 0.030662106536328792,
      "grad_norm": 0.1275985985994339,
      "h_stock_valid_norm": 98.98101272583008,
      "invalid_tail_samples": 25.3,
      "learning_rate": 5.534151789310904e-05,
      "loss": 0.0346,
      "positive_teacher_pairs": 26510.3,
      "query_h_norm_ratio": 0.43046708405017853,
      "relation_loss": 0.0038948662113398314,
      "selected_pairs": 26510.3,
      "step": 14780,
      "student_distance_mean": 25.332583236694337,
      "teacher_distance_mean": 0.863664448261261,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0038948662113398314,
      "weighted_relation_loss": 0.0038948662113398314
    },
    {
      "action_query_batch_std": 0.37853777706623076,
      "action_query_norm": 42.59979133605957,
      "action_relation_loss": 0.0037488064263015985,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.83536014967259,
      "flow_matching_loss": 0.02756661232560873,
      "grad_norm": 0.09878414869308472,
      "h_stock_valid_norm": 98.94254913330079,
      "invalid_tail_samples": 23.4,
      "learning_rate": 5.528671441085376e-05,
      "loss": 0.0313,
      "positive_teacher_pairs": 26941.6,
      "query_h_norm_ratio": 0.43055263459682463,
      "relation_loss": 0.0037488064263015985,
      "selected_pairs": 26941.6,
      "step": 14790,
      "student_distance_mean": 25.21862335205078,
      "teacher_distance_mean": 0.8639601349830628,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0037488064263015985,
      "weighted_relation_loss": 0.0037488064263015985
    },
    {
      "action_query_batch_std": 0.3723936527967453,
      "action_query_norm": 42.561055755615236,
      "action_relation_loss": 0.003978298138827085,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.844714686623012,
      "flow_matching_loss": 0.031547082029283045,
      "grad_norm": 0.10471809655427933,
      "h_stock_valid_norm": 98.35230178833008,
      "invalid_tail_samples": 26.5,
      "learning_rate": 5.523190450474197e-05,
      "loss": 0.0355,
      "positive_teacher_pairs": 26224.5,
      "query_h_norm_ratio": 0.43274675607681273,
      "relation_loss": 0.003978298138827085,
      "selected_pairs": 26224.5,
      "step": 14800,
      "student_distance_mean": 24.821535873413087,
      "teacher_distance_mean": 0.8483813583850861,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.003978298138827085,
      "weighted_relation_loss": 0.003978298138827085
    },
    {
      "action_query_batch_std": 0.3723214894533157,
      "action_query_norm": 42.55824775695801,
      "action_relation_loss": 0.004019342921674251,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.854069223573433,
      "flow_matching_loss": 0.03427289333194494,
      "grad_norm": 0.11025163531303406,
      "h_stock_valid_norm": 97.74595108032227,
      "invalid_tail_samples": 22.5,
      "learning_rate": 5.5177088241372865e-05,
      "loss": 0.0383,
      "positive_teacher_pairs": 27159.1,
      "query_h_norm_ratio": 0.4353994369506836,
      "relation_loss": 0.004019342921674251,
      "selected_pairs": 27159.1,
      "step": 14810,
      "student_distance_mean": 24.758850288391113,
      "teacher_distance_mean": 0.848833966255188,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.004019342921674251,
      "weighted_relation_loss": 0.004019342921674251
    },
    {
      "action_query_batch_std": 0.37651177048683165,
      "action_query_norm": 42.57557907104492,
      "action_relation_loss": 0.003876620624214411,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.863423760523855,
      "flow_matching_loss": 0.029324555583298207,
      "grad_norm": 0.10211648792028427,
      "h_stock_valid_norm": 97.37941131591796,
      "invalid_tail_samples": 24.9,
      "learning_rate": 5.512226568735338e-05,
      "loss": 0.0332,
      "positive_teacher_pairs": 26596.8,
      "query_h_norm_ratio": 0.43721606731414797,
      "relation_loss": 0.003876620624214411,
      "selected_pairs": 26596.8,
      "step": 14820,
      "student_distance_mean": 24.96623992919922,
      "teacher_distance_mean": 0.8592183470726014,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.003876620624214411,
      "weighted_relation_loss": 0.003876620624214411
    },
    {
      "action_query_batch_std": 0.3771675914525986,
      "action_query_norm": 42.57594108581543,
      "action_relation_loss": 0.004052728624083101,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.872778297474275,
      "flow_matching_loss": 0.030926693230867386,
      "grad_norm": 0.09719010442495346,
      "h_stock_valid_norm": 97.44964370727538,
      "invalid_tail_samples": 24.9,
      "learning_rate": 5.506743690929809e-05,
      "loss": 0.035,
      "positive_teacher_pairs": 26601.5,
      "query_h_norm_ratio": 0.43690593242645265,
      "relation_loss": 0.004052728624083101,
      "selected_pairs": 26601.5,
      "step": 14830,
      "student_distance_mean": 25.126014518737794,
      "teacher_distance_mean": 0.8555382311344146,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.004052728624083101,
      "weighted_relation_loss": 0.004052728624083101
    },
    {
      "action_query_batch_std": 0.3777851939201355,
      "action_query_norm": 42.58783073425293,
      "action_relation_loss": 0.0038898608181625605,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.882132834424697,
      "flow_matching_loss": 0.03049596957862377,
      "grad_norm": 0.09041766077280045,
      "h_stock_valid_norm": 97.50041809082032,
      "invalid_tail_samples": 24.7,
      "learning_rate": 5.501260197382912e-05,
      "loss": 0.0344,
      "positive_teacher_pairs": 26644.9,
      "query_h_norm_ratio": 0.43679803013801577,
      "relation_loss": 0.0038898608181625605,
      "selected_pairs": 26644.9,
      "step": 14840,
      "student_distance_mean": 25.103372192382814,
      "teacher_distance_mean": 0.8518848836421966,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0038898608181625605,
      "weighted_relation_loss": 0.0038898608181625605
    },
    {
      "action_query_batch_std": 0.38020521998405454,
      "action_query_norm": 42.5976448059082,
      "action_relation_loss": 0.0037837188923731447,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.891487371375117,
      "flow_matching_loss": 0.030813945829868315,
      "grad_norm": 0.11483379453420639,
      "h_stock_valid_norm": 97.81019744873046,
      "invalid_tail_samples": 24.0,
      "learning_rate": 5.49577609475761e-05,
      "loss": 0.0346,
      "positive_teacher_pairs": 26811.4,
      "query_h_norm_ratio": 0.4355172097682953,
      "relation_loss": 0.0037837188923731447,
      "selected_pairs": 26811.4,
      "step": 14850,
      "student_distance_mean": 25.17602653503418,
      "teacher_distance_mean": 0.8579761922359467,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0037837188923731447,
      "weighted_relation_loss": 0.0037837188923731447
    },
    {
      "action_query_batch_std": 0.37624865770339966,
      "action_query_norm": 42.56570472717285,
      "action_relation_loss": 0.003840234945528209,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.900841908325537,
      "flow_matching_loss": 0.031201279535889626,
      "grad_norm": 0.12201821058988571,
      "h_stock_valid_norm": 97.54741058349609,
      "invalid_tail_samples": 23.3,
      "learning_rate": 5.490291389717603e-05,
      "loss": 0.035,
      "positive_teacher_pairs": 26966.9,
      "query_h_norm_ratio": 0.43636142909526826,
      "relation_loss": 0.003840234945528209,
      "selected_pairs": 26966.9,
      "step": 14860,
      "student_distance_mean": 24.958946990966798,
      "teacher_distance_mean": 0.8570633947849273,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.003840234945528209,
      "weighted_relation_loss": 0.003840234945528209
    },
    {
      "action_query_batch_std": 0.37794903814792635,
      "action_query_norm": 42.571528625488284,
      "action_relation_loss": 0.003810102050192654,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.91019644527596,
      "flow_matching_loss": 0.031564718671143056,
      "grad_norm": 0.08138415217399597,
      "h_stock_valid_norm": 97.73691329956054,
      "invalid_tail_samples": 23.1,
      "learning_rate": 5.484806088927329e-05,
      "loss": 0.0354,
      "positive_teacher_pairs": 27011.7,
      "query_h_norm_ratio": 0.43557703495025635,
      "relation_loss": 0.003810102050192654,
      "selected_pairs": 27011.7,
      "step": 14870,
      "student_distance_mean": 25.161488151550294,
      "teacher_distance_mean": 0.8609717428684235,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.003810102050192654,
      "weighted_relation_loss": 0.003810102050192654
    },
    {
      "action_query_batch_std": 0.3742016077041626,
      "action_query_norm": 42.57407341003418,
      "action_relation_loss": 0.003962473967112601,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.91955098222638,
      "flow_matching_loss": 0.02977119330316782,
      "grad_norm": 0.09284268319606781,
      "h_stock_valid_norm": 98.01181259155274,
      "invalid_tail_samples": 23.6,
      "learning_rate": 5.479320199051942e-05,
      "loss": 0.0337,
      "positive_teacher_pairs": 26893.4,
      "query_h_norm_ratio": 0.43438100516796113,
      "relation_loss": 0.003962473967112601,
      "selected_pairs": 26893.4,
      "step": 14880,
      "student_distance_mean": 24.871695709228515,
      "teacher_distance_mean": 0.8504015266895294,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.003962473967112601,
      "weighted_relation_loss": 0.003962473967112601
    },
    {
      "action_query_batch_std": 0.3740199595689774,
      "action_query_norm": 42.56204071044922,
      "action_relation_loss": 0.0036741736577823756,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.928905519176801,
      "flow_matching_loss": 0.030478683486580848,
      "grad_norm": 0.10931292176246643,
      "h_stock_valid_norm": 98.46355209350585,
      "invalid_tail_samples": 22.4,
      "learning_rate": 5.473833726757314e-05,
      "loss": 0.0342,
      "positive_teacher_pairs": 27177.7,
      "query_h_norm_ratio": 0.4322643756866455,
      "relation_loss": 0.0036741736577823756,
      "selected_pairs": 27177.7,
      "step": 14890,
      "student_distance_mean": 24.790549087524415,
      "teacher_distance_mean": 0.8515190124511719,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.0036741736577823756,
      "weighted_relation_loss": 0.0036741736577823756
    },
    {
      "action_query_batch_std": 0.3720157504081726,
      "action_query_norm": 42.5369369506836,
      "action_relation_loss": 0.0039154361234977845,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.938260056127222,
      "flow_matching_loss": 0.031179318390786647,
      "grad_norm": 0.10401971638202667,
      "h_stock_valid_norm": 98.54710693359375,
      "invalid_tail_samples": 24.3,
      "learning_rate": 5.4683466787100304e-05,
      "loss": 0.0351,
      "positive_teacher_pairs": 26752.6,
      "query_h_norm_ratio": 0.4316438674926758,
      "relation_loss": 0.0039154361234977845,
      "selected_pairs": 26752.6,
      "step": 14900,
      "student_distance_mean": 24.74944610595703,
      "teacher_distance_mean": 0.8472546100616455,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0039154361234977845,
      "weighted_relation_loss": 0.0039154361234977845
    },
    {
      "action_query_batch_std": 0.3705588668584824,
      "action_query_norm": 42.538977432250974,
      "action_relation_loss": 0.003814047994092107,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.947614593077642,
      "flow_matching_loss": 0.030428186431527136,
      "grad_norm": 0.08807220309972763,
      "h_stock_valid_norm": 98.06870727539062,
      "invalid_tail_samples": 25.5,
      "learning_rate": 5.462859061577369e-05,
      "loss": 0.0342,
      "positive_teacher_pairs": 26453.7,
      "query_h_norm_ratio": 0.4337756186723709,
      "relation_loss": 0.003814047994092107,
      "selected_pairs": 26453.7,
      "step": 14910,
      "student_distance_mean": 24.694543647766114,
      "teacher_distance_mean": 0.8468113243579865,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.003814047994092107,
      "weighted_relation_loss": 0.003814047994092107
    },
    {
      "action_query_batch_std": 0.37595263719558714,
      "action_query_norm": 42.548102951049806,
      "action_relation_loss": 0.004240418877452612,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.956969130028064,
      "flow_matching_loss": 0.03101610764861107,
      "grad_norm": 0.12188930809497833,
      "h_stock_valid_norm": 97.74362030029297,
      "invalid_tail_samples": 21.9,
      "learning_rate": 5.4573708820273026e-05,
      "loss": 0.0353,
      "positive_teacher_pairs": 27289.4,
      "query_h_norm_ratio": 0.435306054353714,
      "relation_loss": 0.004240418877452612,
      "selected_pairs": 27289.4,
      "step": 14920,
      "student_distance_mean": 24.929238319396973,
      "teacher_distance_mean": 0.8549724221229553,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.004240418877452612,
      "weighted_relation_loss": 0.004240418877452612
    },
    {
      "action_query_batch_std": 0.37222225964069366,
      "action_query_norm": 42.53827133178711,
      "action_relation_loss": 0.003944453969597817,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.966323666978484,
      "flow_matching_loss": 0.030501939170062543,
      "grad_norm": 0.13976144790649414,
      "h_stock_valid_norm": 97.95472106933593,
      "invalid_tail_samples": 21.0,
      "learning_rate": 5.451882146728488e-05,
      "loss": 0.0344,
      "positive_teacher_pairs": 27504.0,
      "query_h_norm_ratio": 0.4342680811882019,
      "relation_loss": 0.003944453969597817,
      "selected_pairs": 27504.0,
      "step": 14930,
      "student_distance_mean": 24.82031135559082,
      "teacher_distance_mean": 0.8494069576263428,
      "valid_action_samples": 235.0,
      "weighted_auxiliary_loss": 0.003944453969597817,
      "weighted_relation_loss": 0.003944453969597817
    },
    {
      "action_query_batch_std": 0.37565374076366426,
      "action_query_norm": 42.553781509399414,
      "action_relation_loss": 0.003958670003339648,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.975678203928906,
      "flow_matching_loss": 0.03016034197062254,
      "grad_norm": 0.12909547984600067,
      "h_stock_valid_norm": 98.37073364257813,
      "invalid_tail_samples": 24.5,
      "learning_rate": 5.446392862350255e-05,
      "loss": 0.0341,
      "positive_teacher_pairs": 26686.5,
      "query_h_norm_ratio": 0.4325913220643997,
      "relation_loss": 0.003958670003339648,
      "selected_pairs": 26686.5,
      "step": 14940,
      "student_distance_mean": 25.000238990783693,
      "teacher_distance_mean": 0.8533485233783722,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.003958670003339648,
      "weighted_relation_loss": 0.003958670003339648
    },
    {
      "action_query_batch_std": 0.3748855531215668,
      "action_query_norm": 42.54867515563965,
      "action_relation_loss": 0.0039670294616371395,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.985032740879326,
      "flow_matching_loss": 0.02967281062155962,
      "grad_norm": 0.1504238396883011,
      "h_stock_valid_norm": 98.44082336425781,
      "invalid_tail_samples": 23.1,
      "learning_rate": 5.440903035562603e-05,
      "loss": 0.0336,
      "positive_teacher_pairs": 27024.4,
      "query_h_norm_ratio": 0.4322284430265427,
      "relation_loss": 0.0039670294616371395,
      "selected_pairs": 27024.4,
      "step": 14950,
      "student_distance_mean": 24.97898998260498,
      "teacher_distance_mean": 0.8540684282779694,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0039670294616371395,
      "weighted_relation_loss": 0.0039670294616371395
    },
    {
      "action_query_batch_std": 0.3770455181598663,
      "action_query_norm": 42.54343681335449,
      "action_relation_loss": 0.0038892766926437615,
      "auxiliary_loss_weight": 1.0,
      "epoch": 13.994387277829748,
      "flow_matching_loss": 0.029525961354374887,
      "grad_norm": 0.11035040766000748,
      "h_stock_valid_norm": 98.24970703125,
      "invalid_tail_samples": 26.9,
      "learning_rate": 5.435412673036188e-05,
      "loss": 0.0334,
      "positive_teacher_pairs": 26136.1,
      "query_h_norm_ratio": 0.43301920890808104,
      "relation_loss": 0.0038892766926437615,
      "selected_pairs": 26136.1,
      "step": 14960,
      "student_distance_mean": 25.071602058410644,
      "teacher_distance_mean": 0.8583277404308319,
      "valid_action_samples": 229.1,
      "weighted_auxiliary_loss": 0.0038892766926437615,
      "weighted_relation_loss": 0.0038892766926437615
    },
    {
      "action_query_batch_std": 0.376680850982666,
      "action_query_norm": 42.54078483581543,
      "action_relation_loss": 0.0037100808694958687,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.003741814780168,
      "flow_matching_loss": 0.033919600583612916,
      "grad_norm": 0.1200130507349968,
      "h_stock_valid_norm": 97.76338424682618,
      "invalid_tail_samples": 21.8,
      "learning_rate": 5.429921781442318e-05,
      "loss": 0.0376,
      "positive_teacher_pairs": 24327.9,
      "query_h_norm_ratio": 0.43514797985553744,
      "relation_loss": 0.0037100808694958687,
      "selected_pairs": 24327.9,
      "step": 14970,
      "student_distance_mean": 25.171500396728515,
      "teacher_distance_mean": 0.8604766547679901,
      "valid_action_samples": 214.3,
      "weighted_auxiliary_loss": 0.0037100808694958687,
      "weighted_relation_loss": 0.0037100808694958687
    },
    {
      "action_query_batch_std": 0.3756432503461838,
      "action_query_norm": 42.54148979187012,
      "action_relation_loss": 0.0039058303693309425,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.013096351730589,
      "flow_matching_loss": 0.030336875654757022,
      "grad_norm": 0.11350594460964203,
      "h_stock_valid_norm": 97.75558624267578,
      "invalid_tail_samples": 22.8,
      "learning_rate": 5.424430367452945e-05,
      "loss": 0.0342,
      "positive_teacher_pairs": 27086.2,
      "query_h_norm_ratio": 0.43518705368041993,
      "relation_loss": 0.0039058303693309425,
      "selected_pairs": 27086.2,
      "step": 14980,
      "student_distance_mean": 24.989315605163576,
      "teacher_distance_mean": 0.8594993114471435,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0039058303693309425,
      "weighted_relation_loss": 0.0039058303693309425
    },
    {
      "action_query_batch_std": 0.38006907105445864,
      "action_query_norm": 42.54486923217773,
      "action_relation_loss": 0.0036920862970873714,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.02245088868101,
      "flow_matching_loss": 0.030045767314732075,
      "grad_norm": 0.12730935215950012,
      "h_stock_valid_norm": 98.20060653686524,
      "invalid_tail_samples": 25.3,
      "learning_rate": 5.418938437740655e-05,
      "loss": 0.0337,
      "positive_teacher_pairs": 26500.3,
      "query_h_norm_ratio": 0.43324752151966095,
      "relation_loss": 0.0036920862970873714,
      "selected_pairs": 26500.3,
      "step": 14990,
      "student_distance_mean": 25.233977127075196,
      "teacher_distance_mean": 0.8644815683364868,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0036920862970873714,
      "weighted_relation_loss": 0.0036920862970873714
    },
    {
      "action_query_batch_std": 0.3739170014858246,
      "action_query_norm": 42.524179458618164,
      "action_relation_loss": 0.003759219264611602,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.03180542563143,
      "flow_matching_loss": 0.029859143681824206,
      "grad_norm": 0.11052414029836655,
      "h_stock_valid_norm": 98.19807815551758,
      "invalid_tail_samples": 22.3,
      "learning_rate": 5.413445998978658e-05,
      "loss": 0.0336,
      "positive_teacher_pairs": 27195.5,
      "query_h_norm_ratio": 0.4330478340387344,
      "relation_loss": 0.003759219264611602,
      "selected_pairs": 27195.5,
      "step": 15000,
      "student_distance_mean": 24.827925300598146,
      "teacher_distance_mean": 0.8548851132392883,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.003759219264611602,
      "weighted_relation_loss": 0.003759219264611602
    },
    {
      "action_query_batch_std": 0.3755181074142456,
      "action_query_norm": 42.52245864868164,
      "action_relation_loss": 0.0036363441962748767,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.041159962581853,
      "flow_matching_loss": 0.029453159123659135,
      "grad_norm": 0.11313466727733612,
      "h_stock_valid_norm": 98.32489395141602,
      "invalid_tail_samples": 25.8,
      "learning_rate": 5.407953057840789e-05,
      "loss": 0.0331,
      "positive_teacher_pairs": 26386.7,
      "query_h_norm_ratio": 0.43247230648994445,
      "relation_loss": 0.0036363441962748767,
      "selected_pairs": 26386.7,
      "step": 15010,
      "student_distance_mean": 24.9149450302124,
      "teacher_distance_mean": 0.855722713470459,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0036363441962748767,
      "weighted_relation_loss": 0.0036363441962748767
    },
    {
      "action_query_batch_std": 0.3765689730644226,
      "action_query_norm": 42.52709426879883,
      "action_relation_loss": 0.003732904768548906,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.050514499532273,
      "flow_matching_loss": 0.030367513559758662,
      "grad_norm": 0.10290496796369553,
      "h_stock_valid_norm": 97.89462280273438,
      "invalid_tail_samples": 19.2,
      "learning_rate": 5.4024596210014853e-05,
      "loss": 0.0341,
      "positive_teacher_pairs": 27927.0,
      "query_h_norm_ratio": 0.4344186455011368,
      "relation_loss": 0.003732904768548906,
      "selected_pairs": 27927.0,
      "step": 15020,
      "student_distance_mean": 24.97439670562744,
      "teacher_distance_mean": 0.8574491441249847,
      "valid_action_samples": 236.8,
      "weighted_auxiliary_loss": 0.003732904768548906,
      "weighted_relation_loss": 0.003732904768548906
    },
    {
      "action_query_batch_std": 0.3748013526201248,
      "action_query_norm": 42.508005905151364,
      "action_relation_loss": 0.003685741848312318,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.059869036482693,
      "flow_matching_loss": 0.030090304836630822,
      "grad_norm": 0.1209062784910202,
      "h_stock_valid_norm": 98.00828399658204,
      "invalid_tail_samples": 25.8,
      "learning_rate": 5.3969656951357935e-05,
      "loss": 0.0338,
      "positive_teacher_pairs": 26389.8,
      "query_h_norm_ratio": 0.4337219953536987,
      "relation_loss": 0.003685741848312318,
      "selected_pairs": 26389.8,
      "step": 15030,
      "student_distance_mean": 24.93273220062256,
      "teacher_distance_mean": 0.8604185879230499,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.003685741848312318,
      "weighted_relation_loss": 0.003685741848312318
    },
    {
      "action_query_batch_std": 0.3753205895423889,
      "action_query_norm": 42.50975875854492,
      "action_relation_loss": 0.003678615181706846,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.069223573433115,
      "flow_matching_loss": 0.02821411695331335,
      "grad_norm": 0.09939111769199371,
      "h_stock_valid_norm": 98.43572998046875,
      "invalid_tail_samples": 24.7,
      "learning_rate": 5.3914712869193504e-05,
      "loss": 0.0319,
      "positive_teacher_pairs": 26648.8,
      "query_h_norm_ratio": 0.4318564236164093,
      "relation_loss": 0.003678615181706846,
      "selected_pairs": 26648.8,
      "step": 15040,
      "student_distance_mean": 24.93597297668457,
      "teacher_distance_mean": 0.858365786075592,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.003678615181706846,
      "weighted_relation_loss": 0.003678615181706846
    },
    {
      "action_query_batch_std": 0.37583383321762087,
      "action_query_norm": 42.52788200378418,
      "action_relation_loss": 0.0037125177681446076,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.078578110383535,
      "flow_matching_loss": 0.029175861552357674,
      "grad_norm": 0.12617607414722443,
      "h_stock_valid_norm": 98.13199310302734,
      "invalid_tail_samples": 23.5,
      "learning_rate": 5.385976403028381e-05,
      "loss": 0.0329,
      "positive_teacher_pairs": 26926.5,
      "query_h_norm_ratio": 0.4333779513835907,
      "relation_loss": 0.0037125177681446076,
      "selected_pairs": 26926.5,
      "step": 15050,
      "student_distance_mean": 25.03703212738037,
      "teacher_distance_mean": 0.8557230949401855,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0037125177681446076,
      "weighted_relation_loss": 0.0037125177681446076
    },
    {
      "action_query_batch_std": 0.37362549304962156,
      "action_query_norm": 42.50255851745605,
      "action_relation_loss": 0.0036812395555898547,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.087932647333957,
      "flow_matching_loss": 0.02916595097631216,
      "grad_norm": 0.08698942512273788,
      "h_stock_valid_norm": 98.19522857666016,
      "invalid_tail_samples": 24.8,
      "learning_rate": 5.3804810501396875e-05,
      "loss": 0.0328,
      "positive_teacher_pairs": 26620.7,
      "query_h_norm_ratio": 0.4328392595052719,
      "relation_loss": 0.0036812395555898547,
      "selected_pairs": 26620.7,
      "step": 15060,
      "student_distance_mean": 24.747950744628906,
      "teacher_distance_mean": 0.8519750237464905,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0036812395555898547,
      "weighted_relation_loss": 0.0036812395555898547
    },
    {
      "action_query_batch_std": 0.37285428047180175,
      "action_query_norm": 42.49185409545898,
      "action_relation_loss": 0.003695950750261545,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.097287184284378,
      "flow_matching_loss": 0.029182313941419125,
      "grad_norm": 0.11953462660312653,
      "h_stock_valid_norm": 98.12302780151367,
      "invalid_tail_samples": 25.3,
      "learning_rate": 5.374985234930639e-05,
      "loss": 0.0329,
      "positive_teacher_pairs": 26509.2,
      "query_h_norm_ratio": 0.43304887413978577,
      "relation_loss": 0.003695950750261545,
      "selected_pairs": 26509.2,
      "step": 15070,
      "student_distance_mean": 24.845431900024415,
      "teacher_distance_mean": 0.8573721289634705,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.003695950750261545,
      "weighted_relation_loss": 0.003695950750261545
    },
    {
      "action_query_batch_std": 0.3756678462028503,
      "action_query_norm": 42.51694450378418,
      "action_relation_loss": 0.003700069664046168,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.1066417212348,
      "flow_matching_loss": 0.03000192381441593,
      "grad_norm": 0.13292816281318665,
      "h_stock_valid_norm": 98.43999710083008,
      "invalid_tail_samples": 25.0,
      "learning_rate": 5.3694889640791724e-05,
      "loss": 0.0337,
      "positive_teacher_pairs": 26578.4,
      "query_h_norm_ratio": 0.43190876841545106,
      "relation_loss": 0.003700069664046168,
      "selected_pairs": 26578.4,
      "step": 15080,
      "student_distance_mean": 24.97909927368164,
      "teacher_distance_mean": 0.8606893301010132,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.003700069664046168,
      "weighted_relation_loss": 0.003700069664046168
    },
    {
      "action_query_batch_std": 0.37563385665416715,
      "action_query_norm": 42.50286293029785,
      "action_relation_loss": 0.0037148850969970226,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.11599625818522,
      "flow_matching_loss": 0.029129215702414512,
      "grad_norm": 0.12683911621570587,
      "h_stock_valid_norm": 98.52525329589844,
      "invalid_tail_samples": 23.0,
      "learning_rate": 5.3639922442637736e-05,
      "loss": 0.0328,
      "positive_teacher_pairs": 27047.6,
      "query_h_norm_ratio": 0.4313936561346054,
      "relation_loss": 0.0037148850969970226,
      "selected_pairs": 27047.6,
      "step": 15090,
      "student_distance_mean": 24.892814445495606,
      "teacher_distance_mean": 0.8589600384235382,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0037148850969970226,
      "weighted_relation_loss": 0.0037148850969970226
    },
    {
      "action_query_batch_std": 0.3709962278604507,
      "action_query_norm": 42.49259262084961,
      "action_relation_loss": 0.003911653789691627,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.12535079513564,
      "flow_matching_loss": 0.032385351695120336,
      "grad_norm": 0.15329940617084503,
      "h_stock_valid_norm": 98.86604690551758,
      "invalid_tail_samples": 22.0,
      "learning_rate": 5.358495082163476e-05,
      "loss": 0.0363,
      "positive_teacher_pairs": 27266.9,
      "query_h_norm_ratio": 0.42980419993400576,
      "relation_loss": 0.003911653789691627,
      "selected_pairs": 27266.9,
      "step": 15100,
      "student_distance_mean": 24.558317565917967,
      "teacher_distance_mean": 0.842194265127182,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.003911653789691627,
      "weighted_relation_loss": 0.003911653789691627
    },
    {
      "action_query_batch_std": 0.3759760230779648,
      "action_query_norm": 42.51373138427734,
      "action_relation_loss": 0.0039992894046008585,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.134705332086062,
      "flow_matching_loss": 0.030884252674877644,
      "grad_norm": 0.09622199833393097,
      "h_stock_valid_norm": 98.98568954467774,
      "invalid_tail_samples": 23.7,
      "learning_rate": 5.35299748445785e-05,
      "loss": 0.0349,
      "positive_teacher_pairs": 26875.8,
      "query_h_norm_ratio": 0.4294964760541916,
      "relation_loss": 0.0039992894046008585,
      "selected_pairs": 26875.8,
      "step": 15110,
      "student_distance_mean": 24.995268630981446,
      "teacher_distance_mean": 0.8581041753292084,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0039992894046008585,
      "weighted_relation_loss": 0.0039992894046008585
    },
    {
      "action_query_batch_std": 0.3775649040937424,
      "action_query_norm": 42.501440811157224,
      "action_relation_loss": 0.003653652430512011,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.144059869036482,
      "flow_matching_loss": 0.031908022612333296,
      "grad_norm": 0.11173137277364731,
      "h_stock_valid_norm": 99.14844360351563,
      "invalid_tail_samples": 22.8,
      "learning_rate": 5.3474994578269944e-05,
      "loss": 0.0356,
      "positive_teacher_pairs": 27082.0,
      "query_h_norm_ratio": 0.4286668449640274,
      "relation_loss": 0.003653652430512011,
      "selected_pairs": 27082.0,
      "step": 15120,
      "student_distance_mean": 25.012870025634765,
      "teacher_distance_mean": 0.8605375647544861,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.003653652430512011,
      "weighted_relation_loss": 0.003653652430512011
    },
    {
      "action_query_batch_std": 0.3722192168235779,
      "action_query_norm": 42.470166015625,
      "action_relation_loss": 0.003851889865472913,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.153414405986904,
      "flow_matching_loss": 0.03042133804410696,
      "grad_norm": 0.13097524642944336,
      "h_stock_valid_norm": 98.9355354309082,
      "invalid_tail_samples": 22.0,
      "learning_rate": 5.34200100895153e-05,
      "loss": 0.0343,
      "positive_teacher_pairs": 27270.8,
      "query_h_norm_ratio": 0.4292725384235382,
      "relation_loss": 0.003851889865472913,
      "selected_pairs": 27270.8,
      "step": 15130,
      "student_distance_mean": 24.73759937286377,
      "teacher_distance_mean": 0.8485803723335266,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.003851889865472913,
      "weighted_relation_loss": 0.003851889865472913
    },
    {
      "action_query_batch_std": 0.36728976368904115,
      "action_query_norm": 42.44356231689453,
      "action_relation_loss": 0.003847018536180258,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.162768942937324,
      "flow_matching_loss": 0.030389810912311077,
      "grad_norm": 0.10450127720832825,
      "h_stock_valid_norm": 98.51049728393555,
      "invalid_tail_samples": 25.6,
      "learning_rate": 5.3365021445125916e-05,
      "loss": 0.0342,
      "positive_teacher_pairs": 26434.4,
      "query_h_norm_ratio": 0.43085783123970034,
      "relation_loss": 0.003847018536180258,
      "selected_pairs": 26434.4,
      "step": 15140,
      "student_distance_mean": 24.518439292907715,
      "teacher_distance_mean": 0.8492766976356506,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.003847018536180258,
      "weighted_relation_loss": 0.003847018536180258
    },
    {
      "action_query_batch_std": 0.368030247092247,
      "action_query_norm": 42.45320014953613,
      "action_relation_loss": 0.003827829333022237,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.172123479887745,
      "flow_matching_loss": 0.029077771306037902,
      "grad_norm": 0.10202023386955261,
      "h_stock_valid_norm": 98.31294479370118,
      "invalid_tail_samples": 22.4,
      "learning_rate": 5.331002871191817e-05,
      "loss": 0.0329,
      "positive_teacher_pairs": 27185.7,
      "query_h_norm_ratio": 0.4318196654319763,
      "relation_loss": 0.003827829333022237,
      "selected_pairs": 27185.7,
      "step": 15150,
      "student_distance_mean": 24.338311767578126,
      "teacher_distance_mean": 0.8437113404273987,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.003827829333022237,
      "weighted_relation_loss": 0.003827829333022237
    },
    {
      "action_query_batch_std": 0.37892825305461886,
      "action_query_norm": 42.505506896972655,
      "action_relation_loss": 0.003793058916926384,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.181478016838167,
      "flow_matching_loss": 0.030117242597043514,
      "grad_norm": 0.11983779817819595,
      "h_stock_valid_norm": 97.8451919555664,
      "invalid_tail_samples": 24.3,
      "learning_rate": 5.3255031956713443e-05,
      "loss": 0.0339,
      "positive_teacher_pairs": 26733.3,
      "query_h_norm_ratio": 0.4344185084104538,
      "relation_loss": 0.003793058916926384,
      "selected_pairs": 26733.3,
      "step": 15160,
      "student_distance_mean": 25.072969245910645,
      "teacher_distance_mean": 0.8595671117305755,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.003793058916926384,
      "weighted_relation_loss": 0.003793058916926384
    },
    {
      "action_query_batch_std": 0.376348140835762,
      "action_query_norm": 42.49086952209473,
      "action_relation_loss": 0.003555525094270706,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.190832553788587,
      "flow_matching_loss": 0.03079427797347307,
      "grad_norm": 0.10553649812936783,
      "h_stock_valid_norm": 98.09929580688477,
      "invalid_tail_samples": 24.6,
      "learning_rate": 5.320003124633795e-05,
      "loss": 0.0343,
      "positive_teacher_pairs": 26667.5,
      "query_h_norm_ratio": 0.43314622938632963,
      "relation_loss": 0.003555525094270706,
      "selected_pairs": 26667.5,
      "step": 15170,
      "student_distance_mean": 25.07152042388916,
      "teacher_distance_mean": 0.8596644043922425,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.003555525094270706,
      "weighted_relation_loss": 0.003555525094270706
    },
    {
      "action_query_batch_std": 0.37427865862846377,
      "action_query_norm": 42.480267715454104,
      "action_relation_loss": 0.004000860010273754,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.200187090739009,
      "flow_matching_loss": 0.029599453136324883,
      "grad_norm": 0.09549204260110855,
      "h_stock_valid_norm": 97.75199966430664,
      "invalid_tail_samples": 21.4,
      "learning_rate": 5.314502664762274e-05,
      "loss": 0.0336,
      "positive_teacher_pairs": 27406.4,
      "query_h_norm_ratio": 0.4345748215913773,
      "relation_loss": 0.004000860010273754,
      "selected_pairs": 27406.4,
      "step": 15180,
      "student_distance_mean": 24.824599838256837,
      "teacher_distance_mean": 0.8527046084403992,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.004000860010273754,
      "weighted_relation_loss": 0.004000860010273754
    },
    {
      "action_query_batch_std": 0.37248323261737826,
      "action_query_norm": 42.46641235351562,
      "action_relation_loss": 0.003736144071444869,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.209541627689429,
      "flow_matching_loss": 0.028952422179281713,
      "grad_norm": 0.107155941426754,
      "h_stock_valid_norm": 97.99846725463867,
      "invalid_tail_samples": 25.0,
      "learning_rate": 5.3090018227403604e-05,
      "loss": 0.0327,
      "positive_teacher_pairs": 26570.3,
      "query_h_norm_ratio": 0.4333408921957016,
      "relation_loss": 0.003736144071444869,
      "selected_pairs": 26570.3,
      "step": 15190,
      "student_distance_mean": 24.803898048400878,
      "teacher_distance_mean": 0.8523151397705078,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.003736144071444869,
      "weighted_relation_loss": 0.003736144071444869
    },
    {
      "action_query_batch_std": 0.37974746227264405,
      "action_query_norm": 42.50699691772461,
      "action_relation_loss": 0.0037664219737052916,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.218896164639851,
      "flow_matching_loss": 0.030070160701870918,
      "grad_norm": 0.08898793905973434,
      "h_stock_valid_norm": 98.03930587768555,
      "invalid_tail_samples": 20.0,
      "learning_rate": 5.3035006052520955e-05,
      "loss": 0.0338,
      "positive_teacher_pairs": 27736.8,
      "query_h_norm_ratio": 0.43357369005680085,
      "relation_loss": 0.0037664219737052916,
      "selected_pairs": 27736.8,
      "step": 15200,
      "student_distance_mean": 25.178233528137206,
      "teacher_distance_mean": 0.863695603609085,
      "valid_action_samples": 236.0,
      "weighted_auxiliary_loss": 0.0037664219737052916,
      "weighted_relation_loss": 0.0037664219737052916
    },
    {
      "action_query_batch_std": 0.3725461333990097,
      "action_query_norm": 42.45961647033691,
      "action_relation_loss": 0.0037225082283839583,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.228250701590271,
      "flow_matching_loss": 0.03002084605395794,
      "grad_norm": 0.09580104053020477,
      "h_stock_valid_norm": 97.70195693969727,
      "invalid_tail_samples": 22.9,
      "learning_rate": 5.297999018981976e-05,
      "loss": 0.0337,
      "positive_teacher_pairs": 27057.3,
      "query_h_norm_ratio": 0.43458632826805116,
      "relation_loss": 0.0037225082283839583,
      "selected_pairs": 27057.3,
      "step": 15210,
      "student_distance_mean": 24.760349464416503,
      "teacher_distance_mean": 0.8492531776428223,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0037225082283839583,
      "weighted_relation_loss": 0.0037225082283839583
    },
    {
      "action_query_batch_std": 0.370932400226593,
      "action_query_norm": 42.44753074645996,
      "action_relation_loss": 0.0037805019179359078,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.237605238540691,
      "flow_matching_loss": 0.030179528146982194,
      "grad_norm": 0.14584246277809143,
      "h_stock_valid_norm": 98.01390762329102,
      "invalid_tail_samples": 27.1,
      "learning_rate": 5.2924970706149505e-05,
      "loss": 0.034,
      "positive_teacher_pairs": 26092.1,
      "query_h_norm_ratio": 0.4330794483423233,
      "relation_loss": 0.0037805019179359078,
      "selected_pairs": 26092.1,
      "step": 15220,
      "student_distance_mean": 24.611553573608397,
      "teacher_distance_mean": 0.8503003597259522,
      "valid_action_samples": 228.9,
      "weighted_auxiliary_loss": 0.0037805019179359078,
      "weighted_relation_loss": 0.0037805019179359078
    },
    {
      "action_query_batch_std": 0.37400983572006224,
      "action_query_norm": 42.47034873962402,
      "action_relation_loss": 0.0036674302071332933,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.246959775491113,
      "flow_matching_loss": 0.027876114659011363,
      "grad_norm": 0.09363166242837906,
      "h_stock_valid_norm": 98.97568893432617,
      "invalid_tail_samples": 25.1,
      "learning_rate": 5.2869947668364014e-05,
      "loss": 0.0315,
      "positive_teacher_pairs": 26551.4,
      "query_h_norm_ratio": 0.42910545468330386,
      "relation_loss": 0.0036674302071332933,
      "selected_pairs": 26551.4,
      "step": 15230,
      "student_distance_mean": 24.909165954589845,
      "teacher_distance_mean": 0.8571379959583283,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0036674302071332933,
      "weighted_relation_loss": 0.0036674302071332933
    },
    {
      "action_query_batch_std": 0.3799647271633148,
      "action_query_norm": 42.49531936645508,
      "action_relation_loss": 0.0035113129299134014,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.256314312441534,
      "flow_matching_loss": 0.029847870208323003,
      "grad_norm": 0.09614664316177368,
      "h_stock_valid_norm": 99.2104866027832,
      "invalid_tail_samples": 25.5,
      "learning_rate": 5.2814921143321504e-05,
      "loss": 0.0334,
      "positive_teacher_pairs": 26454.3,
      "query_h_norm_ratio": 0.42833676040172575,
      "relation_loss": 0.0035113129299134014,
      "selected_pairs": 26454.3,
      "step": 15240,
      "student_distance_mean": 25.227699089050294,
      "teacher_distance_mean": 0.8648015260696411,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0035113129299134014,
      "weighted_relation_loss": 0.0035113129299134014
    },
    {
      "action_query_batch_std": 0.37473940253257754,
      "action_query_norm": 42.46003608703613,
      "action_relation_loss": 0.0037057396955788137,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.265668849391956,
      "flow_matching_loss": 0.029735036380589008,
      "grad_norm": 0.11212548613548279,
      "h_stock_valid_norm": 98.85179748535157,
      "invalid_tail_samples": 25.9,
      "learning_rate": 5.275989119788436e-05,
      "loss": 0.0334,
      "positive_teacher_pairs": 26371.2,
      "query_h_norm_ratio": 0.4295355647802353,
      "relation_loss": 0.0037057396955788137,
      "selected_pairs": 26371.2,
      "step": 15250,
      "student_distance_mean": 24.864441680908204,
      "teacher_distance_mean": 0.8558477580547332,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.0037057396955788137,
      "weighted_relation_loss": 0.0037057396955788137
    },
    {
      "action_query_batch_std": 0.37557651102542877,
      "action_query_norm": 42.464616012573245,
      "action_relation_loss": 0.0036914109718054534,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.275023386342376,
      "flow_matching_loss": 0.028182129934430124,
      "grad_norm": 0.1309448927640915,
      "h_stock_valid_norm": 98.25094985961914,
      "invalid_tail_samples": 23.5,
      "learning_rate": 5.270485789891919e-05,
      "loss": 0.0319,
      "positive_teacher_pairs": 26929.4,
      "query_h_norm_ratio": 0.43221037089824677,
      "relation_loss": 0.0036914109718054534,
      "selected_pairs": 26929.4,
      "step": 15260,
      "student_distance_mean": 24.885042190551758,
      "teacher_distance_mean": 0.8596670627593994,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0036914109718054534,
      "weighted_relation_loss": 0.0036914109718054534
    },
    {
      "action_query_batch_std": 0.3739406198263168,
      "action_query_norm": 42.46692008972168,
      "action_relation_loss": 0.0036801609909161924,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.284377923292798,
      "flow_matching_loss": 0.030386986769735815,
      "grad_norm": 0.12512686848640442,
      "h_stock_valid_norm": 97.62222366333008,
      "invalid_tail_samples": 21.7,
      "learning_rate": 5.264982131329661e-05,
      "loss": 0.0341,
      "positive_teacher_pairs": 27337.0,
      "query_h_norm_ratio": 0.4350156903266907,
      "relation_loss": 0.0036801609909161924,
      "selected_pairs": 27337.0,
      "step": 15270,
      "student_distance_mean": 24.894422912597655,
      "teacher_distance_mean": 0.8563940942287445,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0036801609909161924,
      "weighted_relation_loss": 0.0036801609909161924
    },
    {
      "action_query_batch_std": 0.37594978511333466,
      "action_query_norm": 42.468115615844724,
      "action_relation_loss": 0.0036937919445335864,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.293732460243218,
      "flow_matching_loss": 0.02963186725974083,
      "grad_norm": 0.10591024905443192,
      "h_stock_valid_norm": 97.52255020141601,
      "invalid_tail_samples": 24.6,
      "learning_rate": 5.259478150789128e-05,
      "loss": 0.0333,
      "positive_teacher_pairs": 26663.3,
      "query_h_norm_ratio": 0.4354749858379364,
      "relation_loss": 0.0036937919445335864,
      "selected_pairs": 26663.3,
      "step": 15280,
      "student_distance_mean": 24.954902267456056,
      "teacher_distance_mean": 0.8610250532627106,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0036937919445335864,
      "weighted_relation_loss": 0.0036937919445335864
    },
    {
      "action_query_batch_std": 0.37449921667575836,
      "action_query_norm": 42.46001510620117,
      "action_relation_loss": 0.0038297903491184117,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.303086997193638,
      "flow_matching_loss": 0.030063740350306033,
      "grad_norm": 0.11998813599348068,
      "h_stock_valid_norm": 97.59796829223633,
      "invalid_tail_samples": 21.7,
      "learning_rate": 5.253973854958173e-05,
      "loss": 0.0339,
      "positive_teacher_pairs": 27337.3,
      "query_h_norm_ratio": 0.4350537836551666,
      "relation_loss": 0.0038297903491184117,
      "selected_pairs": 27337.3,
      "step": 15290,
      "student_distance_mean": 24.842397499084473,
      "teacher_distance_mean": 0.8534475445747376,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0038297903491184117,
      "weighted_relation_loss": 0.0038297903491184117
    },
    {
      "action_query_batch_std": 0.3743341565132141,
      "action_query_norm": 42.45458869934082,
      "action_relation_loss": 0.003827262111008167,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.31244153414406,
      "flow_matching_loss": 0.030114916525781153,
      "grad_norm": 0.09166161715984344,
      "h_stock_valid_norm": 97.62750778198242,
      "invalid_tail_samples": 23.6,
      "learning_rate": 5.2484692505250376e-05,
      "loss": 0.0339,
      "positive_teacher_pairs": 26891.5,
      "query_h_norm_ratio": 0.4348651558160782,
      "relation_loss": 0.003827262111008167,
      "selected_pairs": 26891.5,
      "step": 15300,
      "student_distance_mean": 24.961505699157716,
      "teacher_distance_mean": 0.8534066617488861,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.003827262111008167,
      "weighted_relation_loss": 0.003827262111008167
    },
    {
      "action_query_batch_std": 0.37257111966609957,
      "action_query_norm": 42.4434513092041,
      "action_relation_loss": 0.003661088040098548,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.32179607109448,
      "flow_matching_loss": 0.02887818180024624,
      "grad_norm": 0.11740708351135254,
      "h_stock_valid_norm": 97.74791717529297,
      "invalid_tail_samples": 22.6,
      "learning_rate": 5.242964344178333e-05,
      "loss": 0.0325,
      "positive_teacher_pairs": 27132.1,
      "query_h_norm_ratio": 0.4342160403728485,
      "relation_loss": 0.003661088040098548,
      "selected_pairs": 27132.1,
      "step": 15310,
      "student_distance_mean": 24.707057571411134,
      "teacher_distance_mean": 0.8489765822887421,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.003661088040098548,
      "weighted_relation_loss": 0.003661088040098548
    },
    {
      "action_query_batch_std": 0.37140152156352996,
      "action_query_norm": 42.42758712768555,
      "action_relation_loss": 0.0036927721463143825,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.331150608044902,
      "flow_matching_loss": 0.03073676135390997,
      "grad_norm": 0.12622085213661194,
      "h_stock_valid_norm": 97.06342010498047,
      "invalid_tail_samples": 21.3,
      "learning_rate": 5.237459142607041e-05,
      "loss": 0.0344,
      "positive_teacher_pairs": 27429.6,
      "query_h_norm_ratio": 0.43711828291416166,
      "relation_loss": 0.0036927721463143825,
      "selected_pairs": 27429.6,
      "step": 15320,
      "student_distance_mean": 24.68206329345703,
      "teacher_distance_mean": 0.8509755551815033,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.0036927721463143825,
      "weighted_relation_loss": 0.0036927721463143825
    },
    {
      "action_query_batch_std": 0.37467186748981474,
      "action_query_norm": 42.43667144775391,
      "action_relation_loss": 0.0036992947338148952,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.340505144995323,
      "flow_matching_loss": 0.027572999708354473,
      "grad_norm": 0.09318609535694122,
      "h_stock_valid_norm": 96.50784835815429,
      "invalid_tail_samples": 26.0,
      "learning_rate": 5.231953652500498e-05,
      "loss": 0.0313,
      "positive_teacher_pairs": 26352.8,
      "query_h_norm_ratio": 0.4397241324186325,
      "relation_loss": 0.0036992947338148952,
      "selected_pairs": 26352.8,
      "step": 15330,
      "student_distance_mean": 25.03202838897705,
      "teacher_distance_mean": 0.8609633266925811,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.0036992947338148952,
      "weighted_relation_loss": 0.0036992947338148952
    },
    {
      "action_query_batch_std": 0.37618423998355865,
      "action_query_norm": 42.44185218811035,
      "action_relation_loss": 0.003530406393110752,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.349859681945743,
      "flow_matching_loss": 0.028749843128025532,
      "grad_norm": 0.08940459787845612,
      "h_stock_valid_norm": 97.4839874267578,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.2264478805483976e-05,
      "loss": 0.0323,
      "positive_teacher_pairs": 26985.5,
      "query_h_norm_ratio": 0.43538748621940615,
      "relation_loss": 0.003530406393110752,
      "selected_pairs": 26985.5,
      "step": 15340,
      "student_distance_mean": 24.832780647277833,
      "teacher_distance_mean": 0.8551229894161224,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.003530406393110752,
      "weighted_relation_loss": 0.003530406393110752
    },
    {
      "action_query_batch_std": 0.3731899172067642,
      "action_query_norm": 42.43019485473633,
      "action_relation_loss": 0.003775607398711145,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.359214218896165,
      "flow_matching_loss": 0.030060698464512824,
      "grad_norm": 0.10144065320491791,
      "h_stock_valid_norm": 98.80543365478516,
      "invalid_tail_samples": 25.8,
      "learning_rate": 5.220941833440768e-05,
      "loss": 0.0338,
      "positive_teacher_pairs": 26387.7,
      "query_h_norm_ratio": 0.42943730056285856,
      "relation_loss": 0.003775607398711145,
      "selected_pairs": 26387.7,
      "step": 15350,
      "student_distance_mean": 24.823909568786622,
      "teacher_distance_mean": 0.8519615173339844,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.003775607398711145,
      "weighted_relation_loss": 0.003775607398711145
    },
    {
      "action_query_batch_std": 0.3730513364076614,
      "action_query_norm": 42.43108139038086,
      "action_relation_loss": 0.0034855526173487306,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.368568755846585,
      "flow_matching_loss": 0.027257194556295872,
      "grad_norm": 0.12181190401315689,
      "h_stock_valid_norm": 98.64711151123046,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.215435517867977e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 26984.3,
      "query_h_norm_ratio": 0.4301334172487259,
      "relation_loss": 0.0034855526173487306,
      "selected_pairs": 26984.3,
      "step": 15360,
      "student_distance_mean": 24.761963653564454,
      "teacher_distance_mean": 0.8544203460216522,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0034855526173487306,
      "weighted_relation_loss": 0.0034855526173487306
    },
    {
      "action_query_batch_std": 0.37366718649864195,
      "action_query_norm": 42.41997184753418,
      "action_relation_loss": 0.0037279156036674975,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.377923292797007,
      "flow_matching_loss": 0.029028613306581973,
      "grad_norm": 0.12467677891254425,
      "h_stock_valid_norm": 98.13405685424804,
      "invalid_tail_samples": 23.6,
      "learning_rate": 5.209928940520719e-05,
      "loss": 0.0328,
      "positive_teacher_pairs": 26897.5,
      "query_h_norm_ratio": 0.4322675406932831,
      "relation_loss": 0.0037279156036674975,
      "selected_pairs": 26897.5,
      "step": 15370,
      "student_distance_mean": 24.781251907348633,
      "teacher_distance_mean": 0.8574257612228393,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0037279156036674975,
      "weighted_relation_loss": 0.0037279156036674975
    },
    {
      "action_query_batch_std": 0.3676285743713379,
      "action_query_norm": 42.409537887573244,
      "action_relation_loss": 0.003791047469712794,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.387277829747427,
      "flow_matching_loss": 0.02934916839003563,
      "grad_norm": 0.10502218455076218,
      "h_stock_valid_norm": 98.26931762695312,
      "invalid_tail_samples": 23.8,
      "learning_rate": 5.204422108090004e-05,
      "loss": 0.0331,
      "positive_teacher_pairs": 26852.0,
      "query_h_norm_ratio": 0.4315676480531693,
      "relation_loss": 0.003791047469712794,
      "selected_pairs": 26852.0,
      "step": 15380,
      "student_distance_mean": 24.42991771697998,
      "teacher_distance_mean": 0.8417704403400421,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.003791047469712794,
      "weighted_relation_loss": 0.003791047469712794
    },
    {
      "action_query_batch_std": 0.3747389495372772,
      "action_query_norm": 42.44044570922851,
      "action_relation_loss": 0.0038367700297385456,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.39663236669785,
      "flow_matching_loss": 0.02902067992836237,
      "grad_norm": 0.1253296434879303,
      "h_stock_valid_norm": 98.29020614624024,
      "invalid_tail_samples": 22.2,
      "learning_rate": 5.19891502726715e-05,
      "loss": 0.0329,
      "positive_teacher_pairs": 27217.9,
      "query_h_norm_ratio": 0.4317911922931671,
      "relation_loss": 0.0038367700297385456,
      "selected_pairs": 27217.9,
      "step": 15390,
      "student_distance_mean": 24.938116836547852,
      "teacher_distance_mean": 0.8579927146434784,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0038367700297385456,
      "weighted_relation_loss": 0.0038367700297385456
    },
    {
      "action_query_batch_std": 0.3712500870227814,
      "action_query_norm": 42.41354560852051,
      "action_relation_loss": 0.003780838451348245,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.40598690364827,
      "flow_matching_loss": 0.030663329176604748,
      "grad_norm": 0.1162371039390564,
      "h_stock_valid_norm": 98.13814010620118,
      "invalid_tail_samples": 22.1,
      "learning_rate": 5.193407704743782e-05,
      "loss": 0.0344,
      "positive_teacher_pairs": 27246.0,
      "query_h_norm_ratio": 0.4321858108043671,
      "relation_loss": 0.003780838451348245,
      "selected_pairs": 27246.0,
      "step": 15400,
      "student_distance_mean": 24.62513484954834,
      "teacher_distance_mean": 0.8520141661167144,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.003780838451348245,
      "weighted_relation_loss": 0.003780838451348245
    },
    {
      "action_query_batch_std": 0.37005323469638823,
      "action_query_norm": 42.39577369689941,
      "action_relation_loss": 0.0036850353935733438,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.41534144059869,
      "flow_matching_loss": 0.028744969330728055,
      "grad_norm": 0.1195354238152504,
      "h_stock_valid_norm": 98.13900222778321,
      "invalid_tail_samples": 23.6,
      "learning_rate": 5.187900147211815e-05,
      "loss": 0.0324,
      "positive_teacher_pairs": 26906.2,
      "query_h_norm_ratio": 0.4320022940635681,
      "relation_loss": 0.0036850353935733438,
      "selected_pairs": 26906.2,
      "step": 15410,
      "student_distance_mean": 24.526418495178223,
      "teacher_distance_mean": 0.8500774741172791,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0036850353935733438,
      "weighted_relation_loss": 0.0036850353935733438
    },
    {
      "action_query_batch_std": 0.37522175908088684,
      "action_query_norm": 42.41340560913086,
      "action_relation_loss": 0.0037169101648032664,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.424695977549112,
      "flow_matching_loss": 0.027471602149307727,
      "grad_norm": 0.1290072649717331,
      "h_stock_valid_norm": 98.78582458496093,
      "invalid_tail_samples": 24.2,
      "learning_rate": 5.182392361363453e-05,
      "loss": 0.0312,
      "positive_teacher_pairs": 26757.4,
      "query_h_norm_ratio": 0.4293529063463211,
      "relation_loss": 0.0037169101648032664,
      "selected_pairs": 26757.4,
      "step": 15420,
      "student_distance_mean": 24.81419429779053,
      "teacher_distance_mean": 0.8558978259563446,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0037169101648032664,
      "weighted_relation_loss": 0.0037169101648032664
    },
    {
      "action_query_batch_std": 0.3730104029178619,
      "action_query_norm": 42.402349853515624,
      "action_relation_loss": 0.0036824532551690938,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.434050514499532,
      "flow_matching_loss": 0.02995434496551752,
      "grad_norm": 0.09405054897069931,
      "h_stock_valid_norm": 99.18306579589844,
      "invalid_tail_samples": 24.9,
      "learning_rate": 5.176884353891172e-05,
      "loss": 0.0336,
      "positive_teacher_pairs": 26598.6,
      "query_h_norm_ratio": 0.42751953303813933,
      "relation_loss": 0.0036824532551690938,
      "selected_pairs": 26598.6,
      "step": 15430,
      "student_distance_mean": 24.617340087890625,
      "teacher_distance_mean": 0.852651983499527,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0036824532551690938,
      "weighted_relation_loss": 0.0036824532551690938
    },
    {
      "action_query_batch_std": 0.3726255476474762,
      "action_query_norm": 42.40461311340332,
      "action_relation_loss": 0.003671251027844846,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.443405051449954,
      "flow_matching_loss": 0.02837416511029005,
      "grad_norm": 0.1368742138147354,
      "h_stock_valid_norm": 98.44178314208985,
      "invalid_tail_samples": 23.9,
      "learning_rate": 5.171376131487722e-05,
      "loss": 0.032,
      "positive_teacher_pairs": 26835.8,
      "query_h_norm_ratio": 0.4307622253894806,
      "relation_loss": 0.003671251027844846,
      "selected_pairs": 26835.8,
      "step": 15440,
      "student_distance_mean": 24.823475456237794,
      "teacher_distance_mean": 0.8555058360099792,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.003671251027844846,
      "weighted_relation_loss": 0.003671251027844846
    },
    {
      "action_query_batch_std": 0.3697706013917923,
      "action_query_norm": 42.39876022338867,
      "action_relation_loss": 0.0037331263069063426,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.452759588400374,
      "flow_matching_loss": 0.029691309295594694,
      "grad_norm": 0.10651452094316483,
      "h_stock_valid_norm": 97.71855316162109,
      "invalid_tail_samples": 25.1,
      "learning_rate": 5.165867700846113e-05,
      "loss": 0.0334,
      "positive_teacher_pairs": 26554.4,
      "query_h_norm_ratio": 0.4338876187801361,
      "relation_loss": 0.0037331263069063426,
      "selected_pairs": 26554.4,
      "step": 15450,
      "student_distance_mean": 24.60791435241699,
      "teacher_distance_mean": 0.8446851074695587,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0037331263069063426,
      "weighted_relation_loss": 0.0037331263069063426
    },
    {
      "action_query_batch_std": 0.37020882666110994,
      "action_query_norm": 42.40116424560547,
      "action_relation_loss": 0.003776049637235701,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.462114125350794,
      "flow_matching_loss": 0.02953505665063858,
      "grad_norm": 0.1288827806711197,
      "h_stock_valid_norm": 97.59255905151367,
      "invalid_tail_samples": 22.8,
      "learning_rate": 5.160359068659606e-05,
      "loss": 0.0333,
      "positive_teacher_pairs": 27078.4,
      "query_h_norm_ratio": 0.43447395265102384,
      "relation_loss": 0.003776049637235701,
      "selected_pairs": 27078.4,
      "step": 15460,
      "student_distance_mean": 24.535902214050292,
      "teacher_distance_mean": 0.8474067986011505,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.003776049637235701,
      "weighted_relation_loss": 0.003776049637235701
    },
    {
      "action_query_batch_std": 0.3740754574537277,
      "action_query_norm": 42.41625900268555,
      "action_relation_loss": 0.003540332126431167,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.471468662301216,
      "flow_matching_loss": 0.03179398663341999,
      "grad_norm": 0.11387016624212265,
      "h_stock_valid_norm": 97.7369613647461,
      "invalid_tail_samples": 24.0,
      "learning_rate": 5.154850241621712e-05,
      "loss": 0.0353,
      "positive_teacher_pairs": 26806.7,
      "query_h_norm_ratio": 0.43398756682872774,
      "relation_loss": 0.003540332126431167,
      "selected_pairs": 26806.7,
      "step": 15470,
      "student_distance_mean": 24.911101722717284,
      "teacher_distance_mean": 0.8628239750862121,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.003540332126431167,
      "weighted_relation_loss": 0.003540332126431167
    },
    {
      "action_query_batch_std": 0.3695653975009918,
      "action_query_norm": 42.373059844970705,
      "action_relation_loss": 0.003873888775706291,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.480823199251637,
      "flow_matching_loss": 0.03098526205867529,
      "grad_norm": 0.12090414017438889,
      "h_stock_valid_norm": 98.31093292236328,
      "invalid_tail_samples": 23.5,
      "learning_rate": 5.149341226426172e-05,
      "loss": 0.0349,
      "positive_teacher_pairs": 26925.2,
      "query_h_norm_ratio": 0.43101350367069247,
      "relation_loss": 0.003873888775706291,
      "selected_pairs": 26925.2,
      "step": 15480,
      "student_distance_mean": 24.59883518218994,
      "teacher_distance_mean": 0.8513385832309723,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.003873888775706291,
      "weighted_relation_loss": 0.003873888775706291
    },
    {
      "action_query_batch_std": 0.3725243598222733,
      "action_query_norm": 42.40548553466797,
      "action_relation_loss": 0.003670750791206956,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.490177736202059,
      "flow_matching_loss": 0.0280259869992733,
      "grad_norm": 0.10256997495889664,
      "h_stock_valid_norm": 98.56755447387695,
      "invalid_tail_samples": 22.4,
      "learning_rate": 5.14383202976696e-05,
      "loss": 0.0317,
      "positive_teacher_pairs": 27174.2,
      "query_h_norm_ratio": 0.43022241294384,
      "relation_loss": 0.003670750791206956,
      "selected_pairs": 27174.2,
      "step": 15490,
      "student_distance_mean": 24.723746299743652,
      "teacher_distance_mean": 0.853810453414917,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.003670750791206956,
      "weighted_relation_loss": 0.003670750791206956
    },
    {
      "action_query_batch_std": 0.36290763318538666,
      "action_query_norm": 42.34799919128418,
      "action_relation_loss": 0.003945765527896583,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.499532273152479,
      "flow_matching_loss": 0.030158449709415436,
      "grad_norm": 0.14764469861984253,
      "h_stock_valid_norm": 98.80977859497071,
      "invalid_tail_samples": 23.9,
      "learning_rate": 5.1383226583382684e-05,
      "loss": 0.0341,
      "positive_teacher_pairs": 26833.9,
      "query_h_norm_ratio": 0.42858628928661346,
      "relation_loss": 0.003945765527896583,
      "selected_pairs": 26833.9,
      "step": 15500,
      "student_distance_mean": 24.16801528930664,
      "teacher_distance_mean": 0.8353751242160797,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.003945765527896583,
      "weighted_relation_loss": 0.003945765527896583
    },
    {
      "action_query_batch_std": 0.3708302050828934,
      "action_query_norm": 42.38589096069336,
      "action_relation_loss": 0.0036546210991218685,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.5088868101029,
      "flow_matching_loss": 0.02786724381148815,
      "grad_norm": 0.141211599111557,
      "h_stock_valid_norm": 99.24975662231445,
      "invalid_tail_samples": 23.5,
      "learning_rate": 5.132813118834504e-05,
      "loss": 0.0315,
      "positive_teacher_pairs": 26923.3,
      "query_h_norm_ratio": 0.4270669221878052,
      "relation_loss": 0.0036546210991218685,
      "selected_pairs": 26923.3,
      "step": 15510,
      "student_distance_mean": 24.610996055603028,
      "teacher_distance_mean": 0.8494824826717376,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0036546210991218685,
      "weighted_relation_loss": 0.0036546210991218685
    },
    {
      "action_query_batch_std": 0.3698928564786911,
      "action_query_norm": 42.368426132202146,
      "action_relation_loss": 0.003729891777038574,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.518241347053321,
      "flow_matching_loss": 0.029066531360149382,
      "grad_norm": 0.15669724345207214,
      "h_stock_valid_norm": 98.90496215820312,
      "invalid_tail_samples": 23.5,
      "learning_rate": 5.1273034179502777e-05,
      "loss": 0.0328,
      "positive_teacher_pairs": 26917.3,
      "query_h_norm_ratio": 0.42837841212749483,
      "relation_loss": 0.003729891777038574,
      "selected_pairs": 26917.3,
      "step": 15520,
      "student_distance_mean": 24.67330837249756,
      "teacher_distance_mean": 0.8582186758518219,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.003729891777038574,
      "weighted_relation_loss": 0.003729891777038574
    },
    {
      "action_query_batch_std": 0.36925179958343507,
      "action_query_norm": 42.36948471069336,
      "action_relation_loss": 0.00375878494232893,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.527595884003741,
      "flow_matching_loss": 0.030176077224314212,
      "grad_norm": 0.1587390899658203,
      "h_stock_valid_norm": 98.50670547485352,
      "invalid_tail_samples": 22.3,
      "learning_rate": 5.1217935623803945e-05,
      "loss": 0.0339,
      "positive_teacher_pairs": 27210.1,
      "query_h_norm_ratio": 0.43012005686759947,
      "relation_loss": 0.00375878494232893,
      "selected_pairs": 27210.1,
      "step": 15530,
      "student_distance_mean": 24.613569068908692,
      "teacher_distance_mean": 0.8546329081058502,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.00375878494232893,
      "weighted_relation_loss": 0.00375878494232893
    },
    {
      "action_query_batch_std": 0.37288745045661925,
      "action_query_norm": 42.37666206359863,
      "action_relation_loss": 0.003525750548578799,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.536950420954163,
      "flow_matching_loss": 0.0275880079716444,
      "grad_norm": 0.1487564891576767,
      "h_stock_valid_norm": 98.39963912963867,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.1162835588198476e-05,
      "loss": 0.0311,
      "positive_teacher_pairs": 26991.9,
      "query_h_norm_ratio": 0.43066313564777375,
      "relation_loss": 0.003525750548578799,
      "selected_pairs": 26991.9,
      "step": 15540,
      "student_distance_mean": 24.739430236816407,
      "teacher_distance_mean": 0.8591838419437409,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.003525750548578799,
      "weighted_relation_loss": 0.003525750548578799
    },
    {
      "action_query_batch_std": 0.37434818148612975,
      "action_query_norm": 42.38428001403808,
      "action_relation_loss": 0.0036742457188665865,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.546304957904583,
      "flow_matching_loss": 0.028936280868947506,
      "grad_norm": 0.10169811546802521,
      "h_stock_valid_norm": 98.18374176025391,
      "invalid_tail_samples": 24.1,
      "learning_rate": 5.110773413963813e-05,
      "loss": 0.0326,
      "positive_teacher_pairs": 26785.5,
      "query_h_norm_ratio": 0.4316873222589493,
      "relation_loss": 0.0036742457188665865,
      "selected_pairs": 26785.5,
      "step": 15550,
      "student_distance_mean": 24.93576374053955,
      "teacher_distance_mean": 0.8605045199394226,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0036742457188665865,
      "weighted_relation_loss": 0.0036742457188665865
    },
    {
      "action_query_batch_std": 0.3701043039560318,
      "action_query_norm": 42.380730056762694,
      "action_relation_loss": 0.003674586769193411,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.555659494855005,
      "flow_matching_loss": 0.02950149793177843,
      "grad_norm": 0.11205391585826874,
      "h_stock_valid_norm": 98.35227432250977,
      "invalid_tail_samples": 24.0,
      "learning_rate": 5.105263134507636e-05,
      "loss": 0.0332,
      "positive_teacher_pairs": 26803.6,
      "query_h_norm_ratio": 0.4309107273817062,
      "relation_loss": 0.003674586769193411,
      "selected_pairs": 26803.6,
      "step": 15560,
      "student_distance_mean": 24.61150493621826,
      "teacher_distance_mean": 0.8454653859138489,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.003674586769193411,
      "weighted_relation_loss": 0.003674586769193411
    },
    {
      "action_query_batch_std": 0.37359789907932284,
      "action_query_norm": 42.38586769104004,
      "action_relation_loss": 0.003505449485965073,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.565014031805426,
      "flow_matching_loss": 0.029259452223777772,
      "grad_norm": 0.08321761339902878,
      "h_stock_valid_norm": 98.39439697265625,
      "invalid_tail_samples": 26.0,
      "learning_rate": 5.099752727146824e-05,
      "loss": 0.0328,
      "positive_teacher_pairs": 26345.7,
      "query_h_norm_ratio": 0.43077975809574126,
      "relation_loss": 0.003505449485965073,
      "selected_pairs": 26345.7,
      "step": 15570,
      "student_distance_mean": 24.794794273376464,
      "teacher_distance_mean": 0.8526494860649109,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.003505449485965073,
      "weighted_relation_loss": 0.003505449485965073
    },
    {
      "action_query_batch_std": 0.37116640210151675,
      "action_query_norm": 42.37406921386719,
      "action_relation_loss": 0.0037469657137989996,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.574368568755848,
      "flow_matching_loss": 0.03017585314810276,
      "grad_norm": 0.1084422767162323,
      "h_stock_valid_norm": 98.03330154418946,
      "invalid_tail_samples": 25.5,
      "learning_rate": 5.094242198577042e-05,
      "loss": 0.0339,
      "positive_teacher_pairs": 26460.5,
      "query_h_norm_ratio": 0.43224329352378843,
      "relation_loss": 0.0037469657137989996,
      "selected_pairs": 26460.5,
      "step": 15580,
      "student_distance_mean": 24.568563652038574,
      "teacher_distance_mean": 0.8516272962093353,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0037469657137989996,
      "weighted_relation_loss": 0.0037469657137989996
    },
    {
      "action_query_batch_std": 0.3681309372186661,
      "action_query_norm": 42.34681777954101,
      "action_relation_loss": 0.0035481780068948865,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.583723105706268,
      "flow_matching_loss": 0.02814320903271437,
      "grad_norm": 0.12562967836856842,
      "h_stock_valid_norm": 98.22825775146484,
      "invalid_tail_samples": 23.3,
      "learning_rate": 5.088731555494102e-05,
      "loss": 0.0317,
      "positive_teacher_pairs": 26974.0,
      "query_h_norm_ratio": 0.43110955953598024,
      "relation_loss": 0.0035481780068948865,
      "selected_pairs": 26974.0,
      "step": 15590,
      "student_distance_mean": 24.419295692443846,
      "teacher_distance_mean": 0.8484114646911621,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0035481780068948865,
      "weighted_relation_loss": 0.0035481780068948865
    },
    {
      "action_query_batch_std": 0.3721627742052078,
      "action_query_norm": 42.37628402709961,
      "action_relation_loss": 0.0036703610559925435,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.593077642656688,
      "flow_matching_loss": 0.03067453149706125,
      "grad_norm": 0.13853491842746735,
      "h_stock_valid_norm": 98.14215621948242,
      "invalid_tail_samples": 28.2,
      "learning_rate": 5.0832208045939556e-05,
      "loss": 0.0343,
      "positive_teacher_pairs": 25839.1,
      "query_h_norm_ratio": 0.4317866325378418,
      "relation_loss": 0.0036703610559925435,
      "selected_pairs": 25839.1,
      "step": 15600,
      "student_distance_mean": 24.766941642761232,
      "teacher_distance_mean": 0.8575399994850159,
      "valid_action_samples": 227.8,
      "weighted_auxiliary_loss": 0.0036703610559925435,
      "weighted_relation_loss": 0.0036703610559925435
    },
    {
      "action_query_batch_std": 0.36600582003593446,
      "action_query_norm": 42.34188995361328,
      "action_relation_loss": 0.003881173487752676,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.60243217960711,
      "flow_matching_loss": 0.029504073411226274,
      "grad_norm": 0.12062785029411316,
      "h_stock_valid_norm": 97.87339324951172,
      "invalid_tail_samples": 24.6,
      "learning_rate": 5.0777099525726844e-05,
      "loss": 0.0334,
      "positive_teacher_pairs": 26663.7,
      "query_h_norm_ratio": 0.43262088894844053,
      "relation_loss": 0.003881173487752676,
      "selected_pairs": 26663.7,
      "step": 15610,
      "student_distance_mean": 24.413290405273436,
      "teacher_distance_mean": 0.8426892936229706,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.003881173487752676,
      "weighted_relation_loss": 0.003881173487752676
    },
    {
      "action_query_batch_std": 0.37337093949317934,
      "action_query_norm": 42.364239883422854,
      "action_relation_loss": 0.0034968563122674823,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.61178671655753,
      "flow_matching_loss": 0.02939418088644743,
      "grad_norm": 0.10834847390651703,
      "h_stock_valid_norm": 97.56809463500977,
      "invalid_tail_samples": 24.0,
      "learning_rate": 5.072199006126494e-05,
      "loss": 0.0329,
      "positive_teacher_pairs": 26808.0,
      "query_h_norm_ratio": 0.43420656621456144,
      "relation_loss": 0.0034968563122674823,
      "selected_pairs": 26808.0,
      "step": 15620,
      "student_distance_mean": 24.904303359985352,
      "teacher_distance_mean": 0.8622660636901855,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0034968563122674823,
      "weighted_relation_loss": 0.0034968563122674823
    },
    {
      "action_query_batch_std": 0.37025967240333557,
      "action_query_norm": 42.34873809814453,
      "action_relation_loss": 0.003751941048540175,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.621141253507952,
      "flow_matching_loss": 0.02957259528338909,
      "grad_norm": 0.1352291703224182,
      "h_stock_valid_norm": 98.3463966369629,
      "invalid_tail_samples": 22.2,
      "learning_rate": 5.0666879719517026e-05,
      "loss": 0.0333,
      "positive_teacher_pairs": 27228.9,
      "query_h_norm_ratio": 0.4306102216243744,
      "relation_loss": 0.003751941048540175,
      "selected_pairs": 27228.9,
      "step": 15630,
      "student_distance_mean": 24.717840576171874,
      "teacher_distance_mean": 0.8538028061389923,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.003751941048540175,
      "weighted_relation_loss": 0.003751941048540175
    },
    {
      "action_query_batch_std": 0.37427995800971986,
      "action_query_norm": 42.37807693481445,
      "action_relation_loss": 0.003623617463745177,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.630495790458372,
      "flow_matching_loss": 0.029513028636574744,
      "grad_norm": 0.10283011943101883,
      "h_stock_valid_norm": 99.01475448608399,
      "invalid_tail_samples": 24.8,
      "learning_rate": 5.061176856744737e-05,
      "loss": 0.0331,
      "positive_teacher_pairs": 26619.6,
      "query_h_norm_ratio": 0.42800138592720033,
      "relation_loss": 0.003623617463745177,
      "selected_pairs": 26619.6,
      "step": 15640,
      "student_distance_mean": 24.901432037353516,
      "teacher_distance_mean": 0.8608981013298035,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.003623617463745177,
      "weighted_relation_loss": 0.003623617463745177
    },
    {
      "action_query_batch_std": 0.3767656147480011,
      "action_query_norm": 42.38651542663574,
      "action_relation_loss": 0.003568098903633654,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.639850327408793,
      "flow_matching_loss": 0.030237762071192265,
      "grad_norm": 0.1166568323969841,
      "h_stock_valid_norm": 99.2713119506836,
      "invalid_tail_samples": 23.1,
      "learning_rate": 5.0556656672021205e-05,
      "loss": 0.0338,
      "positive_teacher_pairs": 27014.6,
      "query_h_norm_ratio": 0.42697825133800504,
      "relation_loss": 0.003568098903633654,
      "selected_pairs": 27014.6,
      "step": 15650,
      "student_distance_mean": 24.944606590270997,
      "teacher_distance_mean": 0.8618495166301727,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.003568098903633654,
      "weighted_relation_loss": 0.003568098903633654
    },
    {
      "action_query_batch_std": 0.37294385135173796,
      "action_query_norm": 42.356943893432614,
      "action_relation_loss": 0.003465835703536868,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.649204864359215,
      "flow_matching_loss": 0.028475220315158367,
      "grad_norm": 0.11064067482948303,
      "h_stock_valid_norm": 98.76723403930664,
      "invalid_tail_samples": 24.9,
      "learning_rate": 5.050154410020472e-05,
      "loss": 0.0319,
      "positive_teacher_pairs": 26594.3,
      "query_h_norm_ratio": 0.4288613319396973,
      "relation_loss": 0.003465835703536868,
      "selected_pairs": 26594.3,
      "step": 15660,
      "student_distance_mean": 24.873197937011717,
      "teacher_distance_mean": 0.8626181781291962,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.003465835703536868,
      "weighted_relation_loss": 0.003465835703536868
    },
    {
      "action_query_batch_std": 0.37093630731105803,
      "action_query_norm": 42.35305557250977,
      "action_relation_loss": 0.0035292554646730425,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.658559401309635,
      "flow_matching_loss": 0.029490721598267555,
      "grad_norm": 0.1007685735821724,
      "h_stock_valid_norm": 98.17859649658203,
      "invalid_tail_samples": 24.4,
      "learning_rate": 5.044643091896485e-05,
      "loss": 0.033,
      "positive_teacher_pairs": 26709.2,
      "query_h_norm_ratio": 0.43139444589614867,
      "relation_loss": 0.0035292554646730425,
      "selected_pairs": 26709.2,
      "step": 15670,
      "student_distance_mean": 24.64678382873535,
      "teacher_distance_mean": 0.8551167488098145,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0035292554646730425,
      "weighted_relation_loss": 0.0035292554646730425
    },
    {
      "action_query_batch_std": 0.37752116918563844,
      "action_query_norm": 42.37049827575684,
      "action_relation_loss": 0.0037252626614645125,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.667913938260057,
      "flow_matching_loss": 0.02816001232713461,
      "grad_norm": 0.08215472847223282,
      "h_stock_valid_norm": 97.41510009765625,
      "invalid_tail_samples": 24.6,
      "learning_rate": 5.039131719526932e-05,
      "loss": 0.0319,
      "positive_teacher_pairs": 26670.7,
      "query_h_norm_ratio": 0.4349505066871643,
      "relation_loss": 0.0037252626614645125,
      "selected_pairs": 26670.7,
      "step": 15680,
      "student_distance_mean": 25.06179904937744,
      "teacher_distance_mean": 0.8671139359474183,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0037252626614645125,
      "weighted_relation_loss": 0.0037252626614645125
    },
    {
      "action_query_batch_std": 0.37473011314868926,
      "action_query_norm": 42.35666961669922,
      "action_relation_loss": 0.0034291304647922514,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.677268475210477,
      "flow_matching_loss": 0.027897888980805873,
      "grad_norm": 0.11683110147714615,
      "h_stock_valid_norm": 97.21035079956054,
      "invalid_tail_samples": 24.6,
      "learning_rate": 5.03362029960865e-05,
      "loss": 0.0313,
      "positive_teacher_pairs": 26666.3,
      "query_h_norm_ratio": 0.43572446405887605,
      "relation_loss": 0.0034291304647922514,
      "selected_pairs": 26666.3,
      "step": 15690,
      "student_distance_mean": 24.891888427734376,
      "teacher_distance_mean": 0.8589360237121582,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0034291304647922514,
      "weighted_relation_loss": 0.0034291304647922514
    },
    {
      "action_query_batch_std": 0.3759178429841995,
      "action_query_norm": 42.356143188476565,
      "action_relation_loss": 0.0037264886777848005,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.686623012160897,
      "flow_matching_loss": 0.03045390546321869,
      "grad_norm": 0.09555433690547943,
      "h_stock_valid_norm": 97.53523254394531,
      "invalid_tail_samples": 21.3,
      "learning_rate": 5.028108838838533e-05,
      "loss": 0.0342,
      "positive_teacher_pairs": 27431.7,
      "query_h_norm_ratio": 0.43426900506019595,
      "relation_loss": 0.0037264886777848005,
      "selected_pairs": 27431.7,
      "step": 15700,
      "student_distance_mean": 24.87064723968506,
      "teacher_distance_mean": 0.8608365833759308,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.0037264886777848005,
      "weighted_relation_loss": 0.0037264886777848005
    },
    {
      "action_query_batch_std": 0.3686880797147751,
      "action_query_norm": 42.3196533203125,
      "action_relation_loss": 0.0035161208361387253,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.69597754911132,
      "flow_matching_loss": 0.029171244241297245,
      "grad_norm": 0.10416896641254425,
      "h_stock_valid_norm": 98.34744415283203,
      "invalid_tail_samples": 24.2,
      "learning_rate": 5.0225973439135276e-05,
      "loss": 0.0327,
      "positive_teacher_pairs": 26760.9,
      "query_h_norm_ratio": 0.43031114935874937,
      "relation_loss": 0.0035161208361387253,
      "selected_pairs": 26760.9,
      "step": 15710,
      "student_distance_mean": 24.455336570739746,
      "teacher_distance_mean": 0.8481704771518708,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0035161208361387253,
      "weighted_relation_loss": 0.0035161208361387253
    },
    {
      "action_query_batch_std": 0.3704484462738037,
      "action_query_norm": 42.33992042541504,
      "action_relation_loss": 0.003588238498196006,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.70533208606174,
      "flow_matching_loss": 0.028731555119156836,
      "grad_norm": 0.10709061473608017,
      "h_stock_valid_norm": 98.45318298339843,
      "invalid_tail_samples": 22.8,
      "learning_rate": 5.017085821530617e-05,
      "loss": 0.0323,
      "positive_teacher_pairs": 27088.1,
      "query_h_norm_ratio": 0.4300557643175125,
      "relation_loss": 0.003588238498196006,
      "selected_pairs": 27088.1,
      "step": 15720,
      "student_distance_mean": 24.623211288452147,
      "teacher_distance_mean": 0.8559103906154633,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.003588238498196006,
      "weighted_relation_loss": 0.003588238498196006
    },
    {
      "action_query_batch_std": 0.3655076503753662,
      "action_query_norm": 42.30539817810059,
      "action_relation_loss": 0.003668817225843668,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.714686623012161,
      "flow_matching_loss": 0.02784516420215368,
      "grad_norm": 0.1131400540471077,
      "h_stock_valid_norm": 97.9702522277832,
      "invalid_tail_samples": 24.6,
      "learning_rate": 5.011574278386822e-05,
      "loss": 0.0315,
      "positive_teacher_pairs": 26668.7,
      "query_h_norm_ratio": 0.4318223655223846,
      "relation_loss": 0.003668817225843668,
      "selected_pairs": 26668.7,
      "step": 15730,
      "student_distance_mean": 24.257131004333495,
      "teacher_distance_mean": 0.8457450151443482,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.003668817225843668,
      "weighted_relation_loss": 0.003668817225843668
    },
    {
      "action_query_batch_std": 0.37681005597114564,
      "action_query_norm": 42.3536792755127,
      "action_relation_loss": 0.003550896281376481,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.724041159962582,
      "flow_matching_loss": 0.02995133325457573,
      "grad_norm": 0.11162639409303665,
      "h_stock_valid_norm": 97.67268447875976,
      "invalid_tail_samples": 25.0,
      "learning_rate": 5.006062721179189e-05,
      "loss": 0.0335,
      "positive_teacher_pairs": 26573.9,
      "query_h_norm_ratio": 0.43363105058670043,
      "relation_loss": 0.003550896281376481,
      "selected_pairs": 26573.9,
      "step": 15740,
      "student_distance_mean": 25.085634613037108,
      "teacher_distance_mean": 0.869992595911026,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.003550896281376481,
      "weighted_relation_loss": 0.003550896281376481
    },
    {
      "action_query_batch_std": 0.37197905480861665,
      "action_query_norm": 42.34881439208984,
      "action_relation_loss": 0.003578306012786925,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.733395696913004,
      "flow_matching_loss": 0.028551795706152915,
      "grad_norm": 0.10866119712591171,
      "h_stock_valid_norm": 98.0788673400879,
      "invalid_tail_samples": 22.1,
      "learning_rate": 5.000551156604777e-05,
      "loss": 0.0321,
      "positive_teacher_pairs": 27247.2,
      "query_h_norm_ratio": 0.43178440928459166,
      "relation_loss": 0.003578306012786925,
      "selected_pairs": 27247.2,
      "step": 15750,
      "student_distance_mean": 24.683410263061525,
      "teacher_distance_mean": 0.8509743630886077,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.003578306012786925,
      "weighted_relation_loss": 0.003578306012786925
    },
    {
      "action_query_batch_std": 0.3698918431997299,
      "action_query_norm": 42.31038475036621,
      "action_relation_loss": 0.003696925542317331,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.742750233863424,
      "flow_matching_loss": 0.02970473263412714,
      "grad_norm": 0.12500332295894623,
      "h_stock_valid_norm": 98.62340698242187,
      "invalid_tail_samples": 23.5,
      "learning_rate": 4.9950395913606594e-05,
      "loss": 0.0334,
      "positive_teacher_pairs": 26918.3,
      "query_h_norm_ratio": 0.4290119081735611,
      "relation_loss": 0.003696925542317331,
      "selected_pairs": 26918.3,
      "step": 15760,
      "student_distance_mean": 24.48413791656494,
      "teacher_distance_mean": 0.848172789812088,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.003696925542317331,
      "weighted_relation_loss": 0.003696925542317331
    },
    {
      "action_query_batch_std": 0.3669834166765213,
      "action_query_norm": 42.311279678344725,
      "action_relation_loss": 0.0036540297558531167,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.752104770813844,
      "flow_matching_loss": 0.02991618253290653,
      "grad_norm": 0.11980308592319489,
      "h_stock_valid_norm": 98.74666137695313,
      "invalid_tail_samples": 26.0,
      "learning_rate": 4.989528032143903e-05,
      "loss": 0.0336,
      "positive_teacher_pairs": 26343.4,
      "query_h_norm_ratio": 0.4284882783889771,
      "relation_loss": 0.0036540297558531167,
      "selected_pairs": 26343.4,
      "step": 15770,
      "student_distance_mean": 24.451297760009766,
      "teacher_distance_mean": 0.847814929485321,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.0036540297558531167,
      "weighted_relation_loss": 0.0036540297558531167
    },
    {
      "action_query_batch_std": 0.3702510803937912,
      "action_query_norm": 42.32181282043457,
      "action_relation_loss": 0.003680339828133583,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.761459307764266,
      "flow_matching_loss": 0.02819010466337204,
      "grad_norm": 0.11374745517969131,
      "h_stock_valid_norm": 98.29815216064453,
      "invalid_tail_samples": 25.3,
      "learning_rate": 4.984016485651578e-05,
      "loss": 0.0319,
      "positive_teacher_pairs": 26507.2,
      "query_h_norm_ratio": 0.43054933845996857,
      "relation_loss": 0.003680339828133583,
      "selected_pairs": 26507.2,
      "step": 15780,
      "student_distance_mean": 24.602228546142577,
      "teacher_distance_mean": 0.8494770228862762,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.003680339828133583,
      "weighted_relation_loss": 0.003680339828133583
    },
    {
      "action_query_batch_std": 0.37509176433086394,
      "action_query_norm": 42.346059799194336,
      "action_relation_loss": 0.0035446817521005867,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.770813844714686,
      "flow_matching_loss": 0.02908182516694069,
      "grad_norm": 0.18515460193157196,
      "h_stock_valid_norm": 97.88367233276367,
      "invalid_tail_samples": 20.7,
      "learning_rate": 4.9785049585807274e-05,
      "loss": 0.0326,
      "positive_teacher_pairs": 27568.6,
      "query_h_norm_ratio": 0.43261857628822326,
      "relation_loss": 0.0035446817521005867,
      "selected_pairs": 27568.6,
      "step": 15790,
      "student_distance_mean": 24.981078910827637,
      "teacher_distance_mean": 0.8693042159080505,
      "valid_action_samples": 235.3,
      "weighted_auxiliary_loss": 0.0035446817521005867,
      "weighted_relation_loss": 0.0035446817521005867
    },
    {
      "action_query_batch_std": 0.36906742453575136,
      "action_query_norm": 42.30775451660156,
      "action_relation_loss": 0.0037722609471529723,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.780168381665108,
      "flow_matching_loss": 0.030222304910421372,
      "grad_norm": 0.14539282023906708,
      "h_stock_valid_norm": 97.7489631652832,
      "invalid_tail_samples": 23.5,
      "learning_rate": 4.972993457628381e-05,
      "loss": 0.034,
      "positive_teacher_pairs": 26918.4,
      "query_h_norm_ratio": 0.43282175064086914,
      "relation_loss": 0.0037722609471529723,
      "selected_pairs": 26918.4,
      "step": 15800,
      "student_distance_mean": 24.473915672302248,
      "teacher_distance_mean": 0.8501485884189606,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0037722609471529723,
      "weighted_relation_loss": 0.0037722609471529723
    },
    {
      "action_query_batch_std": 0.3733552753925323,
      "action_query_norm": 42.32946090698242,
      "action_relation_loss": 0.0036832018522545695,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.789522918615528,
      "flow_matching_loss": 0.028224744275212287,
      "grad_norm": 0.12918931245803833,
      "h_stock_valid_norm": 97.77371520996094,
      "invalid_tail_samples": 23.7,
      "learning_rate": 4.9674819894915306e-05,
      "loss": 0.0319,
      "positive_teacher_pairs": 26868.7,
      "query_h_norm_ratio": 0.43293745517730714,
      "relation_loss": 0.0036832018522545695,
      "selected_pairs": 26868.7,
      "step": 15810,
      "student_distance_mean": 24.823918342590332,
      "teacher_distance_mean": 0.8562114775180817,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0036832018522545695,
      "weighted_relation_loss": 0.0036832018522545695
    },
    {
      "action_query_batch_std": 0.3761213421821594,
      "action_query_norm": 42.32887229919434,
      "action_relation_loss": 0.0034958488773554563,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.79887745556595,
      "flow_matching_loss": 0.030056909099221228,
      "grad_norm": 0.09933456778526306,
      "h_stock_valid_norm": 97.92922973632812,
      "invalid_tail_samples": 25.5,
      "learning_rate": 4.9619705608671265e-05,
      "loss": 0.0336,
      "positive_teacher_pairs": 26462.3,
      "query_h_norm_ratio": 0.43224388360977173,
      "relation_loss": 0.0034958488773554563,
      "selected_pairs": 26462.3,
      "step": 15820,
      "student_distance_mean": 24.98383159637451,
      "teacher_distance_mean": 0.8582633435726166,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0034958488773554563,
      "weighted_relation_loss": 0.0034958488773554563
    },
    {
      "action_query_batch_std": 0.3731614738702774,
      "action_query_norm": 42.319457626342775,
      "action_relation_loss": 0.0035268751438707114,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.80823199251637,
      "flow_matching_loss": 0.02864908557385206,
      "grad_norm": 0.11296462267637253,
      "h_stock_valid_norm": 98.16477279663086,
      "invalid_tail_samples": 24.7,
      "learning_rate": 4.956459178452079e-05,
      "loss": 0.0322,
      "positive_teacher_pairs": 26648.6,
      "query_h_norm_ratio": 0.43111182749271393,
      "relation_loss": 0.0035268751438707114,
      "selected_pairs": 26648.6,
      "step": 15830,
      "student_distance_mean": 24.618163299560546,
      "teacher_distance_mean": 0.8567769825458527,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0035268751438707114,
      "weighted_relation_loss": 0.0035268751438707114
    },
    {
      "action_query_batch_std": 0.3723233163356781,
      "action_query_norm": 42.30090866088867,
      "action_relation_loss": 0.0035349716898053886,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.81758652946679,
      "flow_matching_loss": 0.029851914569735526,
      "grad_norm": 0.1424683928489685,
      "h_stock_valid_norm": 97.60926971435546,
      "invalid_tail_samples": 22.3,
      "learning_rate": 4.950947848943235e-05,
      "loss": 0.0334,
      "positive_teacher_pairs": 27200.2,
      "query_h_norm_ratio": 0.4333717465400696,
      "relation_loss": 0.0035349716898053886,
      "selected_pairs": 27200.2,
      "step": 15840,
      "student_distance_mean": 24.657330894470213,
      "teacher_distance_mean": 0.8561438739299774,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0035349716898053886,
      "weighted_relation_loss": 0.0035349716898053886
    },
    {
      "action_query_batch_std": 0.37062239050865176,
      "action_query_norm": 42.302999877929686,
      "action_relation_loss": 0.003607281041331589,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.826941066417213,
      "flow_matching_loss": 0.028345219790935516,
      "grad_norm": 0.10361967980861664,
      "h_stock_valid_norm": 97.2688980102539,
      "invalid_tail_samples": 23.3,
      "learning_rate": 4.9454365790373805e-05,
      "loss": 0.032,
      "positive_teacher_pairs": 26966.3,
      "query_h_norm_ratio": 0.4349106580018997,
      "relation_loss": 0.003607281041331589,
      "selected_pairs": 26966.3,
      "step": 15850,
      "student_distance_mean": 24.57534828186035,
      "teacher_distance_mean": 0.8506567656993866,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.003607281041331589,
      "weighted_relation_loss": 0.003607281041331589
    },
    {
      "action_query_batch_std": 0.36989575922489165,
      "action_query_norm": 42.291111373901366,
      "action_relation_loss": 0.0037640429567545653,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.836295603367633,
      "flow_matching_loss": 0.030477827042341234,
      "grad_norm": 0.11650381237268448,
      "h_stock_valid_norm": 97.21845779418945,
      "invalid_tail_samples": 24.2,
      "learning_rate": 4.939925375431226e-05,
      "loss": 0.0342,
      "positive_teacher_pairs": 26754.3,
      "query_h_norm_ratio": 0.4350143551826477,
      "relation_loss": 0.0037640429567545653,
      "selected_pairs": 26754.3,
      "step": 15860,
      "student_distance_mean": 24.595304107666017,
      "teacher_distance_mean": 0.8549624502658844,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0037640429567545653,
      "weighted_relation_loss": 0.0037640429567545653
    },
    {
      "action_query_batch_std": 0.3740400969982147,
      "action_query_norm": 42.30395431518555,
      "action_relation_loss": 0.0034598056226968765,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.845650140318055,
      "flow_matching_loss": 0.029138810001313686,
      "grad_norm": 0.09886764734983444,
      "h_stock_valid_norm": 98.06162719726562,
      "invalid_tail_samples": 21.5,
      "learning_rate": 4.9344142448214046e-05,
      "loss": 0.0326,
      "positive_teacher_pairs": 27386.8,
      "query_h_norm_ratio": 0.43141249716281893,
      "relation_loss": 0.0034598056226968765,
      "selected_pairs": 27386.8,
      "step": 15870,
      "student_distance_mean": 24.809230613708497,
      "teacher_distance_mean": 0.8531561255455017,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.0034598056226968765,
      "weighted_relation_loss": 0.0034598056226968765
    },
    {
      "action_query_batch_std": 0.369573125243187,
      "action_query_norm": 42.28311386108398,
      "action_relation_loss": 0.0034805056639015675,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.855004677268475,
      "flow_matching_loss": 0.0286523612216115,
      "grad_norm": 0.11314454674720764,
      "h_stock_valid_norm": 98.70523376464844,
      "invalid_tail_samples": 21.8,
      "learning_rate": 4.928903193904461e-05,
      "loss": 0.0321,
      "positive_teacher_pairs": 27322.8,
      "query_h_norm_ratio": 0.4283799290657043,
      "relation_loss": 0.0034805056639015675,
      "selected_pairs": 27322.8,
      "step": 15880,
      "student_distance_mean": 24.489509201049806,
      "teacher_distance_mean": 0.8479480385780335,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.0034805056639015675,
      "weighted_relation_loss": 0.0034805056639015675
    },
    {
      "action_query_batch_std": 0.37229294180870054,
      "action_query_norm": 42.291303634643555,
      "action_relation_loss": 0.0035322470357641578,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.864359214218895,
      "flow_matching_loss": 0.029006754420697688,
      "grad_norm": 0.12243045121431351,
      "h_stock_valid_norm": 98.5049560546875,
      "invalid_tail_samples": 22.8,
      "learning_rate": 4.923392229376842e-05,
      "loss": 0.0325,
      "positive_teacher_pairs": 27087.7,
      "query_h_norm_ratio": 0.42933603525161745,
      "relation_loss": 0.0035322470357641578,
      "selected_pairs": 27087.7,
      "step": 15890,
      "student_distance_mean": 24.72064838409424,
      "teacher_distance_mean": 0.8566983222961426,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0035322470357641578,
      "weighted_relation_loss": 0.0035322470357641578
    },
    {
      "action_query_batch_std": 0.3743597358465195,
      "action_query_norm": 42.30734405517578,
      "action_relation_loss": 0.0034998069517314435,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.873713751169317,
      "flow_matching_loss": 0.028006328828632833,
      "grad_norm": 0.0988335981965065,
      "h_stock_valid_norm": 98.2081916809082,
      "invalid_tail_samples": 24.5,
      "learning_rate": 4.9178813579348865e-05,
      "loss": 0.0315,
      "positive_teacher_pairs": 26689.0,
      "query_h_norm_ratio": 0.4307984381914139,
      "relation_loss": 0.0034998069517314435,
      "selected_pairs": 26689.0,
      "step": 15900,
      "student_distance_mean": 24.927285575866698,
      "teacher_distance_mean": 0.8622668981552124,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0034998069517314435,
      "weighted_relation_loss": 0.0034998069517314435
    },
    {
      "action_query_batch_std": 0.37264117896556853,
      "action_query_norm": 42.301177215576175,
      "action_relation_loss": 0.0034102284582331775,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.883068288119738,
      "flow_matching_loss": 0.027762259170413018,
      "grad_norm": 0.1101214736700058,
      "h_stock_valid_norm": 97.69523544311524,
      "invalid_tail_samples": 22.3,
      "learning_rate": 4.9123705862748253e-05,
      "loss": 0.0312,
      "positive_teacher_pairs": 27201.5,
      "query_h_norm_ratio": 0.4329959720373154,
      "relation_loss": 0.0034102284582331775,
      "selected_pairs": 27201.5,
      "step": 15910,
      "student_distance_mean": 24.716227340698243,
      "teacher_distance_mean": 0.8558797717094422,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0034102284582331775,
      "weighted_relation_loss": 0.0034102284582331775
    },
    {
      "action_query_batch_std": 0.36887621581554414,
      "action_query_norm": 42.274040603637694,
      "action_relation_loss": 0.0035524438600987194,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.89242282507016,
      "flow_matching_loss": 0.028118703700602054,
      "grad_norm": 0.11074820905923843,
      "h_stock_valid_norm": 97.36916656494141,
      "invalid_tail_samples": 24.1,
      "learning_rate": 4.9068599210927627e-05,
      "loss": 0.0317,
      "positive_teacher_pairs": 26781.0,
      "query_h_norm_ratio": 0.43416473269462585,
      "relation_loss": 0.0035524438600987194,
      "selected_pairs": 26781.0,
      "step": 15920,
      "student_distance_mean": 24.53345355987549,
      "teacher_distance_mean": 0.8520523607730865,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0035524438600987194,
      "weighted_relation_loss": 0.0035524438600987194
    },
    {
      "action_query_batch_std": 0.3750851511955261,
      "action_query_norm": 42.28946037292481,
      "action_relation_loss": 0.0036337312776595354,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.90177736202058,
      "flow_matching_loss": 0.02942880056798458,
      "grad_norm": 0.1251942217350006,
      "h_stock_valid_norm": 97.67975540161133,
      "invalid_tail_samples": 23.8,
      "learning_rate": 4.901349369084681e-05,
      "loss": 0.0331,
      "positive_teacher_pairs": 26867.7,
      "query_h_norm_ratio": 0.43294318318367003,
      "relation_loss": 0.0036337312776595354,
      "selected_pairs": 26867.7,
      "step": 15930,
      "student_distance_mean": 24.865607452392577,
      "teacher_distance_mean": 0.8657507419586181,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0036337312776595354,
      "weighted_relation_loss": 0.0036337312776595354
    },
    {
      "action_query_batch_std": 0.37589321434497835,
      "action_query_norm": 42.309383392333984,
      "action_relation_loss": 0.0034694016678258775,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.911131898971,
      "flow_matching_loss": 0.028598296642303466,
      "grad_norm": 0.09830988198518753,
      "h_stock_valid_norm": 98.30045089721679,
      "invalid_tail_samples": 22.2,
      "learning_rate": 4.8958389369464156e-05,
      "loss": 0.0321,
      "positive_teacher_pairs": 27218.8,
      "query_h_norm_ratio": 0.43040957748889924,
      "relation_loss": 0.0034694016678258775,
      "selected_pairs": 27218.8,
      "step": 15940,
      "student_distance_mean": 24.904125595092772,
      "teacher_distance_mean": 0.8628336191177368,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0034694016678258775,
      "weighted_relation_loss": 0.0034694016678258775
    },
    {
      "action_query_batch_std": 0.3712060064077377,
      "action_query_norm": 42.28838920593262,
      "action_relation_loss": 0.0035593306412920354,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.920486435921422,
      "flow_matching_loss": 0.029825954884290694,
      "grad_norm": 0.12348476052284241,
      "h_stock_valid_norm": 98.5825584411621,
      "invalid_tail_samples": 23.9,
      "learning_rate": 4.890328631373666e-05,
      "loss": 0.0334,
      "positive_teacher_pairs": 26825.5,
      "query_h_norm_ratio": 0.4289709120988846,
      "relation_loss": 0.0035593306412920354,
      "selected_pairs": 26825.5,
      "step": 15950,
      "student_distance_mean": 24.587683868408202,
      "teacher_distance_mean": 0.8486961424350739,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0035593306412920354,
      "weighted_relation_loss": 0.0035593306412920354
    },
    {
      "action_query_batch_std": 0.3699746787548065,
      "action_query_norm": 42.27243156433106,
      "action_relation_loss": 0.003532271459698677,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.929840972871842,
      "flow_matching_loss": 0.028568545542657375,
      "grad_norm": 0.09073515981435776,
      "h_stock_valid_norm": 98.00479507446289,
      "invalid_tail_samples": 25.1,
      "learning_rate": 4.88481845906197e-05,
      "loss": 0.0321,
      "positive_teacher_pairs": 26560.2,
      "query_h_norm_ratio": 0.4313351184129715,
      "relation_loss": 0.003532271459698677,
      "selected_pairs": 26560.2,
      "step": 15960,
      "student_distance_mean": 24.562023162841797,
      "teacher_distance_mean": 0.8492038726806641,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.003532271459698677,
      "weighted_relation_loss": 0.003532271459698677
    },
    {
      "action_query_batch_std": 0.37403195798397065,
      "action_query_norm": 42.28888130187988,
      "action_relation_loss": 0.0036111035849899055,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.939195509822264,
      "flow_matching_loss": 0.02891895081847906,
      "grad_norm": 0.08766420930624008,
      "h_stock_valid_norm": 97.5211395263672,
      "invalid_tail_samples": 23.7,
      "learning_rate": 4.879308426706707e-05,
      "loss": 0.0325,
      "positive_teacher_pairs": 26880.1,
      "query_h_norm_ratio": 0.4336414188146591,
      "relation_loss": 0.0036111035849899055,
      "selected_pairs": 26880.1,
      "step": 15970,
      "student_distance_mean": 24.729685401916505,
      "teacher_distance_mean": 0.8588651120662689,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0036111035849899055,
      "weighted_relation_loss": 0.0036111035849899055
    },
    {
      "action_query_batch_std": 0.36673955619335175,
      "action_query_norm": 42.25426979064942,
      "action_relation_loss": 0.0036723410245031118,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.948550046772684,
      "flow_matching_loss": 0.02956363055855036,
      "grad_norm": 0.08679758012294769,
      "h_stock_valid_norm": 97.55075302124024,
      "invalid_tail_samples": 20.7,
      "learning_rate": 4.873798541003084e-05,
      "loss": 0.0332,
      "positive_teacher_pairs": 27573.8,
      "query_h_norm_ratio": 0.43315501511096954,
      "relation_loss": 0.0036723410245031118,
      "selected_pairs": 27573.8,
      "step": 15980,
      "student_distance_mean": 24.3538610458374,
      "teacher_distance_mean": 0.851254403591156,
      "valid_action_samples": 235.3,
      "weighted_auxiliary_loss": 0.0036723410245031118,
      "weighted_relation_loss": 0.0036723410245031118
    },
    {
      "action_query_batch_std": 0.3652563840150833,
      "action_query_norm": 42.23638076782227,
      "action_relation_loss": 0.0034902219427749514,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.957904583723106,
      "flow_matching_loss": 0.029278682731091975,
      "grad_norm": 0.08466428518295288,
      "h_stock_valid_norm": 98.01842956542968,
      "invalid_tail_samples": 24.4,
      "learning_rate": 4.868288808646136e-05,
      "loss": 0.0328,
      "positive_teacher_pairs": 26729.0,
      "query_h_norm_ratio": 0.4309041053056717,
      "relation_loss": 0.0034902219427749514,
      "selected_pairs": 26729.0,
      "step": 15990,
      "student_distance_mean": 24.1931734085083,
      "teacher_distance_mean": 0.8461026966571807,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0034902219427749514,
      "weighted_relation_loss": 0.0034902219427749514
    },
    {
      "action_query_batch_std": 0.36916052401065824,
      "action_query_norm": 42.26083564758301,
      "action_relation_loss": 0.0034426807425916193,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.967259120673527,
      "flow_matching_loss": 0.02813788577914238,
      "grad_norm": 0.0986369177699089,
      "h_stock_valid_norm": 97.93031845092773,
      "invalid_tail_samples": 25.5,
      "learning_rate": 4.862779236330705e-05,
      "loss": 0.0316,
      "positive_teacher_pairs": 26467.5,
      "query_h_norm_ratio": 0.43154231607913973,
      "relation_loss": 0.0034426807425916193,
      "selected_pairs": 26467.5,
      "step": 16000,
      "student_distance_mean": 24.509980583190917,
      "teacher_distance_mean": 0.8566137671470642,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0034426807425916193,
      "weighted_relation_loss": 0.0034426807425916193
    },
    {
      "action_query_batch_std": 0.36984296441078185,
      "action_query_norm": 42.25140724182129,
      "action_relation_loss": 0.003443822148256004,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.976613657623947,
      "flow_matching_loss": 0.029544254206120966,
      "grad_norm": 0.13034090399742126,
      "h_stock_valid_norm": 97.67182388305665,
      "invalid_tail_samples": 26.7,
      "learning_rate": 4.8572698307514395e-05,
      "loss": 0.033,
      "positive_teacher_pairs": 26179.4,
      "query_h_norm_ratio": 0.432586869597435,
      "relation_loss": 0.003443822148256004,
      "selected_pairs": 26179.4,
      "step": 16010,
      "student_distance_mean": 24.508717918395995,
      "teacher_distance_mean": 0.8521770775318146,
      "valid_action_samples": 229.3,
      "weighted_auxiliary_loss": 0.003443822148256004,
      "weighted_relation_loss": 0.003443822148256004
    },
    {
      "action_query_batch_std": 0.3693435102701187,
      "action_query_norm": 42.26767501831055,
      "action_relation_loss": 0.0035952214151620866,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.985968194574369,
      "flow_matching_loss": 0.02833227925002575,
      "grad_norm": 0.09601395577192307,
      "h_stock_valid_norm": 98.23910598754883,
      "invalid_tail_samples": 23.4,
      "learning_rate": 4.85176059860279e-05,
      "loss": 0.0319,
      "positive_teacher_pairs": 26940.3,
      "query_h_norm_ratio": 0.43025805056095123,
      "relation_loss": 0.0035952214151620866,
      "selected_pairs": 26940.3,
      "step": 16020,
      "student_distance_mean": 24.5790735244751,
      "teacher_distance_mean": 0.8496523916721344,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0035952214151620866,
      "weighted_relation_loss": 0.0035952214151620866
    },
    {
      "action_query_batch_std": 0.37208672165870665,
      "action_query_norm": 42.266876220703125,
      "action_relation_loss": 0.0037974935956299307,
      "auxiliary_loss_weight": 1.0,
      "epoch": 14.995322731524789,
      "flow_matching_loss": 0.03004743829369545,
      "grad_norm": 0.08843757212162018,
      "h_stock_valid_norm": 98.47245483398437,
      "invalid_tail_samples": 24.0,
      "learning_rate": 4.846251546578989e-05,
      "loss": 0.0338,
      "positive_teacher_pairs": 26806.9,
      "query_h_norm_ratio": 0.4292292147874832,
      "relation_loss": 0.0037974935956299307,
      "selected_pairs": 26806.9,
      "step": 16030,
      "student_distance_mean": 24.579993438720702,
      "teacher_distance_mean": 0.8524382591247559,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0037974935956299307,
      "weighted_relation_loss": 0.0037974935956299307
    },
    {
      "action_query_batch_std": 0.3712505429983139,
      "action_query_norm": 42.23930778503418,
      "action_relation_loss": 0.0035018582362681626,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.004677268475211,
      "flow_matching_loss": 0.02888827435672283,
      "grad_norm": 0.11127371340990067,
      "h_stock_valid_norm": 97.87435302734374,
      "invalid_tail_samples": 23.8,
      "learning_rate": 4.840742681374058e-05,
      "loss": 0.0324,
      "positive_teacher_pairs": 23953.5,
      "query_h_norm_ratio": 0.4315704643726349,
      "relation_loss": 0.0035018582362681626,
      "selected_pairs": 23953.5,
      "step": 16040,
      "student_distance_mean": 24.602047538757326,
      "teacher_distance_mean": 0.8579780220985412,
      "valid_action_samples": 212.3,
      "weighted_auxiliary_loss": 0.0035018582362681626,
      "weighted_relation_loss": 0.0035018582362681626
    },
    {
      "action_query_batch_std": 0.3695171743631363,
      "action_query_norm": 42.232743453979495,
      "action_relation_loss": 0.0035563044948503375,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.014031805425631,
      "flow_matching_loss": 0.027732823975384235,
      "grad_norm": 0.11427433043718338,
      "h_stock_valid_norm": 97.45535202026367,
      "invalid_tail_samples": 25.6,
      "learning_rate": 4.835234009681787e-05,
      "loss": 0.0313,
      "positive_teacher_pairs": 26436.1,
      "query_h_norm_ratio": 0.433357435464859,
      "relation_loss": 0.0035563044948503375,
      "selected_pairs": 26436.1,
      "step": 16050,
      "student_distance_mean": 24.487583923339844,
      "teacher_distance_mean": 0.8552719354629517,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0035563044948503375,
      "weighted_relation_loss": 0.0035563044948503375
    },
    {
      "action_query_batch_std": 0.37202642261981966,
      "action_query_norm": 42.265652084350585,
      "action_relation_loss": 0.0033131354255601765,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.023386342376053,
      "flow_matching_loss": 0.02843759097158909,
      "grad_norm": 0.11257725208997726,
      "h_stock_valid_norm": 98.19140625,
      "invalid_tail_samples": 22.8,
      "learning_rate": 4.8297255381957296e-05,
      "loss": 0.0318,
      "positive_teacher_pairs": 27079.7,
      "query_h_norm_ratio": 0.4304474055767059,
      "relation_loss": 0.0033131354255601765,
      "selected_pairs": 27079.7,
      "step": 16060,
      "student_distance_mean": 24.730133628845216,
      "teacher_distance_mean": 0.8575641274452209,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0033131354255601765,
      "weighted_relation_loss": 0.0033131354255601765
    },
    {
      "action_query_batch_std": 0.375320628285408,
      "action_query_norm": 42.26989974975586,
      "action_relation_loss": 0.0033476166194304825,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.032740879326473,
      "flow_matching_loss": 0.027383730560541154,
      "grad_norm": 0.12303037196397781,
      "h_stock_valid_norm": 98.48776168823242,
      "invalid_tail_samples": 24.8,
      "learning_rate": 4.824217273609198e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 26614.3,
      "query_h_norm_ratio": 0.42919124364852906,
      "relation_loss": 0.0033476166194304825,
      "selected_pairs": 26614.3,
      "step": 16070,
      "student_distance_mean": 24.843472480773926,
      "teacher_distance_mean": 0.8633100748062134,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0033476166194304825,
      "weighted_relation_loss": 0.0033476166194304825
    },
    {
      "action_query_batch_std": 0.3608678638935089,
      "action_query_norm": 42.186058044433594,
      "action_relation_loss": 0.0036606609588488936,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.042095416276894,
      "flow_matching_loss": 0.026354268565773963,
      "grad_norm": 0.09659027308225632,
      "h_stock_valid_norm": 98.55000305175781,
      "invalid_tail_samples": 23.9,
      "learning_rate": 4.8187092226152555e-05,
      "loss": 0.03,
      "positive_teacher_pairs": 26833.3,
      "query_h_norm_ratio": 0.42806979417800906,
      "relation_loss": 0.0036606609588488936,
      "selected_pairs": 26833.3,
      "step": 16080,
      "student_distance_mean": 23.939733505249023,
      "teacher_distance_mean": 0.8334941744804383,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0036606609588488936,
      "weighted_relation_loss": 0.0036606609588488936
    },
    {
      "action_query_batch_std": 0.37358459532260896,
      "action_query_norm": 42.2586124420166,
      "action_relation_loss": 0.003554784995503724,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.051449953227316,
      "flow_matching_loss": 0.027699359320104124,
      "grad_norm": 0.13198906183242798,
      "h_stock_valid_norm": 98.28139038085938,
      "invalid_tail_samples": 24.1,
      "learning_rate": 4.8132013919067016e-05,
      "loss": 0.0313,
      "positive_teacher_pairs": 26782.4,
      "query_h_norm_ratio": 0.4299836397171021,
      "relation_loss": 0.003554784995503724,
      "selected_pairs": 26782.4,
      "step": 16090,
      "student_distance_mean": 24.638010787963868,
      "teacher_distance_mean": 0.8522696316242218,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.003554784995503724,
      "weighted_relation_loss": 0.003554784995503724
    },
    {
      "action_query_batch_std": 0.3698675841093063,
      "action_query_norm": 42.2379566192627,
      "action_relation_loss": 0.0034247297327965496,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.060804490177736,
      "flow_matching_loss": 0.0279952485114336,
      "grad_norm": 0.12347127497196198,
      "h_stock_valid_norm": 98.24254684448242,
      "invalid_tail_samples": 23.0,
      "learning_rate": 4.807693788176071e-05,
      "loss": 0.0314,
      "positive_teacher_pairs": 27039.5,
      "query_h_norm_ratio": 0.4299371808767319,
      "relation_loss": 0.0034247297327965496,
      "selected_pairs": 27039.5,
      "step": 16100,
      "student_distance_mean": 24.462355041503905,
      "teacher_distance_mean": 0.8432645618915557,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0034247297327965496,
      "weighted_relation_loss": 0.0034247297327965496
    },
    {
      "action_query_batch_std": 0.37111090421676635,
      "action_query_norm": 42.24214477539063,
      "action_relation_loss": 0.0033643554197624326,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.070159027128158,
      "flow_matching_loss": 0.0290520953014493,
      "grad_norm": 0.11947444826364517,
      "h_stock_valid_norm": 98.38134307861328,
      "invalid_tail_samples": 25.8,
      "learning_rate": 4.8021864181156214e-05,
      "loss": 0.0324,
      "positive_teacher_pairs": 26386.3,
      "query_h_norm_ratio": 0.4293725430965424,
      "relation_loss": 0.0033643554197624326,
      "selected_pairs": 26386.3,
      "step": 16110,
      "student_distance_mean": 24.652763748168944,
      "teacher_distance_mean": 0.859071010351181,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0033643554197624326,
      "weighted_relation_loss": 0.0033643554197624326
    },
    {
      "action_query_batch_std": 0.3645347416400909,
      "action_query_norm": 42.20388488769531,
      "action_relation_loss": 0.0033744366141036154,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.079513564078578,
      "flow_matching_loss": 0.028967385925352574,
      "grad_norm": 0.13276460766792297,
      "h_stock_valid_norm": 98.3123565673828,
      "invalid_tail_samples": 24.8,
      "learning_rate": 4.796679288417326e-05,
      "loss": 0.0323,
      "positive_teacher_pairs": 26623.9,
      "query_h_norm_ratio": 0.4292886644601822,
      "relation_loss": 0.0033744366141036154,
      "selected_pairs": 26623.9,
      "step": 16120,
      "student_distance_mean": 24.197028160095215,
      "teacher_distance_mean": 0.8432132422924041,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0033744366141036154,
      "weighted_relation_loss": 0.0033744366141036154
    },
    {
      "action_query_batch_std": 0.3689657926559448,
      "action_query_norm": 42.22201766967773,
      "action_relation_loss": 0.0034179548965767026,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.088868101028998,
      "flow_matching_loss": 0.026610898785293102,
      "grad_norm": 0.1396607756614685,
      "h_stock_valid_norm": 97.57110443115235,
      "invalid_tail_samples": 23.5,
      "learning_rate": 4.791172405772866e-05,
      "loss": 0.03,
      "positive_teacher_pairs": 26922.5,
      "query_h_norm_ratio": 0.4327367305755615,
      "relation_loss": 0.0034179548965767026,
      "selected_pairs": 26922.5,
      "step": 16130,
      "student_distance_mean": 24.396491622924806,
      "teacher_distance_mean": 0.8497547209262848,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0034179548965767026,
      "weighted_relation_loss": 0.0034179548965767026
    },
    {
      "action_query_batch_std": 0.3728908449411392,
      "action_query_norm": 42.22887535095215,
      "action_relation_loss": 0.0033799851313233375,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.09822263797942,
      "flow_matching_loss": 0.027542112953960897,
      "grad_norm": 0.10850153863430023,
      "h_stock_valid_norm": 97.32072525024414,
      "invalid_tail_samples": 25.4,
      "learning_rate": 4.785665776873625e-05,
      "loss": 0.0309,
      "positive_teacher_pairs": 26484.2,
      "query_h_norm_ratio": 0.4339174598455429,
      "relation_loss": 0.0033799851313233375,
      "selected_pairs": 26484.2,
      "step": 16140,
      "student_distance_mean": 24.680173301696776,
      "teacher_distance_mean": 0.859623384475708,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0033799851313233375,
      "weighted_relation_loss": 0.0033799851313233375
    },
    {
      "action_query_batch_std": 0.3708795517683029,
      "action_query_norm": 42.231099700927736,
      "action_relation_loss": 0.003350255452096462,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.10757717492984,
      "flow_matching_loss": 0.027895545959472655,
      "grad_norm": 0.11250709742307663,
      "h_stock_valid_norm": 97.98817749023438,
      "invalid_tail_samples": 24.9,
      "learning_rate": 4.7801594084106763e-05,
      "loss": 0.0312,
      "positive_teacher_pairs": 26597.9,
      "query_h_norm_ratio": 0.4309853434562683,
      "relation_loss": 0.003350255452096462,
      "selected_pairs": 26597.9,
      "step": 16150,
      "student_distance_mean": 24.584135818481446,
      "teacher_distance_mean": 0.8540557265281677,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.003350255452096462,
      "weighted_relation_loss": 0.003350255452096462
    },
    {
      "action_query_batch_std": 0.3738363325595856,
      "action_query_norm": 42.247079849243164,
      "action_relation_loss": 0.00342083009891212,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.116931711880262,
      "flow_matching_loss": 0.02769629843533039,
      "grad_norm": 0.1150728389620781,
      "h_stock_valid_norm": 98.64256286621094,
      "invalid_tail_samples": 22.3,
      "learning_rate": 4.774653307074775e-05,
      "loss": 0.0311,
      "positive_teacher_pairs": 27200.5,
      "query_h_norm_ratio": 0.4282872945070267,
      "relation_loss": 0.00342083009891212,
      "selected_pairs": 27200.5,
      "step": 16160,
      "student_distance_mean": 24.854616165161133,
      "teacher_distance_mean": 0.8685772001743317,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.00342083009891212,
      "weighted_relation_loss": 0.00342083009891212
    },
    {
      "action_query_batch_std": 0.37254887223243716,
      "action_query_norm": 42.23448143005371,
      "action_relation_loss": 0.0033102638088166715,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.126286248830683,
      "flow_matching_loss": 0.026266538351774216,
      "grad_norm": 0.09975509345531464,
      "h_stock_valid_norm": 98.47253875732422,
      "invalid_tail_samples": 24.9,
      "learning_rate": 4.769147479556355e-05,
      "loss": 0.0296,
      "positive_teacher_pairs": 26597.5,
      "query_h_norm_ratio": 0.42890151739120486,
      "relation_loss": 0.0033102638088166715,
      "selected_pairs": 26597.5,
      "step": 16170,
      "student_distance_mean": 24.77332000732422,
      "teacher_distance_mean": 0.8617823421955109,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0033102638088166715,
      "weighted_relation_loss": 0.0033102638088166715
    },
    {
      "action_query_batch_std": 0.3669320583343506,
      "action_query_norm": 42.20323295593262,
      "action_relation_loss": 0.0034927936038002374,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.135640785781105,
      "flow_matching_loss": 0.027610334008932112,
      "grad_norm": 0.13684199750423431,
      "h_stock_valid_norm": 98.88290100097656,
      "invalid_tail_samples": 24.8,
      "learning_rate": 4.763641932545515e-05,
      "loss": 0.0311,
      "positive_teacher_pairs": 26625.3,
      "query_h_norm_ratio": 0.42680488526821136,
      "relation_loss": 0.0034927936038002374,
      "selected_pairs": 26625.3,
      "step": 16180,
      "student_distance_mean": 24.361628913879393,
      "teacher_distance_mean": 0.8498121321201324,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0034927936038002374,
      "weighted_relation_loss": 0.0034927936038002374
    },
    {
      "action_query_batch_std": 0.3693687915802002,
      "action_query_norm": 42.20548629760742,
      "action_relation_loss": 0.0032994292676448824,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.144995322731525,
      "flow_matching_loss": 0.02906333766877651,
      "grad_norm": 0.11983976513147354,
      "h_stock_valid_norm": 98.59452896118164,
      "invalid_tail_samples": 24.5,
      "learning_rate": 4.758136672732013e-05,
      "loss": 0.0324,
      "positive_teacher_pairs": 26684.1,
      "query_h_norm_ratio": 0.42807572782039643,
      "relation_loss": 0.0032994292676448824,
      "selected_pairs": 26684.1,
      "step": 16190,
      "student_distance_mean": 24.49902973175049,
      "teacher_distance_mean": 0.858208155632019,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0032994292676448824,
      "weighted_relation_loss": 0.0032994292676448824
    },
    {
      "action_query_batch_std": 0.36732186675071715,
      "action_query_norm": 42.19128150939942,
      "action_relation_loss": 0.00334741564001888,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.154349859681945,
      "flow_matching_loss": 0.027854178659617902,
      "grad_norm": 0.09471483528614044,
      "h_stock_valid_norm": 97.67786026000977,
      "invalid_tail_samples": 25.8,
      "learning_rate": 4.752631706805261e-05,
      "loss": 0.0312,
      "positive_teacher_pairs": 26397.4,
      "query_h_norm_ratio": 0.43194715976715087,
      "relation_loss": 0.00334741564001888,
      "selected_pairs": 26397.4,
      "step": 16200,
      "student_distance_mean": 24.404528617858887,
      "teacher_distance_mean": 0.8498863101005554,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.00334741564001888,
      "weighted_relation_loss": 0.00334741564001888
    },
    {
      "action_query_batch_std": 0.37497070133686067,
      "action_query_norm": 42.23119010925293,
      "action_relation_loss": 0.0034671111265197397,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.163704396632367,
      "flow_matching_loss": 0.028060444444417954,
      "grad_norm": 0.09082827717065811,
      "h_stock_valid_norm": 97.90629806518555,
      "invalid_tail_samples": 23.9,
      "learning_rate": 4.7471270414543114e-05,
      "loss": 0.0315,
      "positive_teacher_pairs": 26830.7,
      "query_h_norm_ratio": 0.4313506007194519,
      "relation_loss": 0.0034671111265197397,
      "selected_pairs": 26830.7,
      "step": 16210,
      "student_distance_mean": 24.841860008239745,
      "teacher_distance_mean": 0.864267772436142,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0034671111265197397,
      "weighted_relation_loss": 0.0034671111265197397
    },
    {
      "action_query_batch_std": 0.37240190505981446,
      "action_query_norm": 42.21708183288574,
      "action_relation_loss": 0.003364361543208361,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.173058933582787,
      "flow_matching_loss": 0.027771610580384732,
      "grad_norm": 0.1180015280842781,
      "h_stock_valid_norm": 98.35070571899413,
      "invalid_tail_samples": 24.0,
      "learning_rate": 4.741622683367849e-05,
      "loss": 0.0311,
      "positive_teacher_pairs": 26804.1,
      "query_h_norm_ratio": 0.42925148606300356,
      "relation_loss": 0.003364361543208361,
      "selected_pairs": 26804.1,
      "step": 16220,
      "student_distance_mean": 24.760676765441893,
      "teacher_distance_mean": 0.8597762525081635,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.003364361543208361,
      "weighted_relation_loss": 0.003364361543208361
    },
    {
      "action_query_batch_std": 0.3687775909900665,
      "action_query_norm": 42.20530815124512,
      "action_relation_loss": 0.0034156391164287927,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.18241347053321,
      "flow_matching_loss": 0.027772743441164493,
      "grad_norm": 0.1088019385933876,
      "h_stock_valid_norm": 98.43384017944337,
      "invalid_tail_samples": 25.4,
      "learning_rate": 4.736118639234191e-05,
      "loss": 0.0312,
      "positive_teacher_pairs": 26479.2,
      "query_h_norm_ratio": 0.4287708669900894,
      "relation_loss": 0.0034156391164287927,
      "selected_pairs": 26479.2,
      "step": 16230,
      "student_distance_mean": 24.56536731719971,
      "teacher_distance_mean": 0.8542441725730896,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0034156391164287927,
      "weighted_relation_loss": 0.0034156391164287927
    },
    {
      "action_query_batch_std": 0.36571009159088136,
      "action_query_norm": 42.16634864807129,
      "action_relation_loss": 0.0035293590510264037,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.19176800748363,
      "flow_matching_loss": 0.024509733356535435,
      "grad_norm": 0.1036384105682373,
      "h_stock_valid_norm": 98.34468078613281,
      "invalid_tail_samples": 24.6,
      "learning_rate": 4.730614915741267e-05,
      "loss": 0.028,
      "positive_teacher_pairs": 26667.5,
      "query_h_norm_ratio": 0.4287624925374985,
      "relation_loss": 0.0035293590510264037,
      "selected_pairs": 26667.5,
      "step": 16240,
      "student_distance_mean": 24.273466300964355,
      "teacher_distance_mean": 0.8489972174167633,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0035293590510264037,
      "weighted_relation_loss": 0.0035293590510264037
    },
    {
      "action_query_batch_std": 0.3620311379432678,
      "action_query_norm": 42.169677734375,
      "action_relation_loss": 0.0034593203803524376,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.20112254443405,
      "flow_matching_loss": 0.027822163701057435,
      "grad_norm": 0.10390966385602951,
      "h_stock_valid_norm": 98.32192001342773,
      "invalid_tail_samples": 23.0,
      "learning_rate": 4.7251115195766234e-05,
      "loss": 0.0313,
      "positive_teacher_pairs": 27042.2,
      "query_h_norm_ratio": 0.42890200316905974,
      "relation_loss": 0.0034593203803524376,
      "selected_pairs": 27042.2,
      "step": 16250,
      "student_distance_mean": 23.9991397857666,
      "teacher_distance_mean": 0.8442302048206329,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0034593203803524376,
      "weighted_relation_loss": 0.0034593203803524376
    },
    {
      "action_query_batch_std": 0.3710984498262405,
      "action_query_norm": 42.21621894836426,
      "action_relation_loss": 0.0034319489495828748,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.210477081384472,
      "flow_matching_loss": 0.027178778685629366,
      "grad_norm": 0.10887134820222855,
      "h_stock_valid_norm": 98.17983932495117,
      "invalid_tail_samples": 24.9,
      "learning_rate": 4.719608457427404e-05,
      "loss": 0.0306,
      "positive_teacher_pairs": 26597.8,
      "query_h_norm_ratio": 0.4299913376569748,
      "relation_loss": 0.0034319489495828748,
      "selected_pairs": 26597.8,
      "step": 16260,
      "student_distance_mean": 24.60424633026123,
      "teacher_distance_mean": 0.8598739504814148,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0034319489495828748,
      "weighted_relation_loss": 0.0034319489495828748
    },
    {
      "action_query_batch_std": 0.36776831448078157,
      "action_query_norm": 42.17193717956543,
      "action_relation_loss": 0.0034498796798288823,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.219831618334892,
      "flow_matching_loss": 0.028392320871353148,
      "grad_norm": 0.137313112616539,
      "h_stock_valid_norm": 97.69404220581055,
      "invalid_tail_samples": 25.4,
      "learning_rate": 4.714105735980347e-05,
      "loss": 0.0318,
      "positive_teacher_pairs": 26482.7,
      "query_h_norm_ratio": 0.4316802442073822,
      "relation_loss": 0.0034498796798288823,
      "selected_pairs": 26482.7,
      "step": 16270,
      "student_distance_mean": 24.25522232055664,
      "teacher_distance_mean": 0.8472562611103058,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0034498796798288823,
      "weighted_relation_loss": 0.0034498796798288823
    },
    {
      "action_query_batch_std": 0.36778818964958193,
      "action_query_norm": 42.16898384094238,
      "action_relation_loss": 0.0034638823010027407,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.229186155285314,
      "flow_matching_loss": 0.02631949316710234,
      "grad_norm": 0.12842364609241486,
      "h_stock_valid_norm": 97.41207656860351,
      "invalid_tail_samples": 25.8,
      "learning_rate": 4.70860336192178e-05,
      "loss": 0.0298,
      "positive_teacher_pairs": 26388.8,
      "query_h_norm_ratio": 0.43289532959461213,
      "relation_loss": 0.0034638823010027407,
      "selected_pairs": 26388.8,
      "step": 16280,
      "student_distance_mean": 24.2005916595459,
      "teacher_distance_mean": 0.8471508502960206,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0034638823010027407,
      "weighted_relation_loss": 0.0034638823010027407
    },
    {
      "action_query_batch_std": 0.3684330075979233,
      "action_query_norm": 42.18818130493164,
      "action_relation_loss": 0.003203184879384935,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.238540692235734,
      "flow_matching_loss": 0.02580223623663187,
      "grad_norm": 0.1130744069814682,
      "h_stock_valid_norm": 98.11561126708985,
      "invalid_tail_samples": 23.6,
      "learning_rate": 4.7031013419376034e-05,
      "loss": 0.029,
      "positive_teacher_pairs": 26894.5,
      "query_h_norm_ratio": 0.4299889445304871,
      "relation_loss": 0.003203184879384935,
      "selected_pairs": 26894.5,
      "step": 16290,
      "student_distance_mean": 24.37981014251709,
      "teacher_distance_mean": 0.8512989640235901,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.003203184879384935,
      "weighted_relation_loss": 0.003203184879384935
    },
    {
      "action_query_batch_std": 0.3722010225057602,
      "action_query_norm": 42.20621185302734,
      "action_relation_loss": 0.0032957112416625023,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.247895229186156,
      "flow_matching_loss": 0.026247877255082132,
      "grad_norm": 0.08751101046800613,
      "h_stock_valid_norm": 98.19062728881836,
      "invalid_tail_samples": 22.9,
      "learning_rate": 4.6975996827132926e-05,
      "loss": 0.0295,
      "positive_teacher_pairs": 27060.2,
      "query_h_norm_ratio": 0.42984348237514497,
      "relation_loss": 0.0032957112416625023,
      "selected_pairs": 27060.2,
      "step": 16300,
      "student_distance_mean": 24.77288875579834,
      "teacher_distance_mean": 0.8603724837303162,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0032957112416625023,
      "weighted_relation_loss": 0.0032957112416625023
    },
    {
      "action_query_batch_std": 0.36706002354621886,
      "action_query_norm": 42.18067588806152,
      "action_relation_loss": 0.0034597330959513783,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.257249766136576,
      "flow_matching_loss": 0.027712604217231273,
      "grad_norm": 0.11149165779352188,
      "h_stock_valid_norm": 98.21174850463868,
      "invalid_tail_samples": 24.2,
      "learning_rate": 4.692098390933883e-05,
      "loss": 0.0312,
      "positive_teacher_pairs": 26761.0,
      "query_h_norm_ratio": 0.4294897705316544,
      "relation_loss": 0.0034597330959513783,
      "selected_pairs": 26761.0,
      "step": 16310,
      "student_distance_mean": 24.455809593200684,
      "teacher_distance_mean": 0.8520193517208099,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0034597330959513783,
      "weighted_relation_loss": 0.0034597330959513783
    },
    {
      "action_query_batch_std": 0.3678145080804825,
      "action_query_norm": 42.178397750854494,
      "action_relation_loss": 0.0034001901978626846,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.266604303086996,
      "flow_matching_loss": 0.02868141941726208,
      "grad_norm": 0.11816791445016861,
      "h_stock_valid_norm": 97.8349838256836,
      "invalid_tail_samples": 22.3,
      "learning_rate": 4.6865974732839615e-05,
      "loss": 0.0321,
      "positive_teacher_pairs": 27195.6,
      "query_h_norm_ratio": 0.43111926019191743,
      "relation_loss": 0.0034001901978626846,
      "selected_pairs": 27195.6,
      "step": 16320,
      "student_distance_mean": 24.483168029785155,
      "teacher_distance_mean": 0.8532697856426239,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0034001901978626846,
      "weighted_relation_loss": 0.0034001901978626846
    },
    {
      "action_query_batch_std": 0.36846196055412295,
      "action_query_norm": 42.180606842041016,
      "action_relation_loss": 0.003382595698349178,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.275958840037418,
      "flow_matching_loss": 0.029811470955610275,
      "grad_norm": 0.11002874374389648,
      "h_stock_valid_norm": 98.17728042602539,
      "invalid_tail_samples": 25.8,
      "learning_rate": 4.6810969364476615e-05,
      "loss": 0.0332,
      "positive_teacher_pairs": 26384.0,
      "query_h_norm_ratio": 0.4296436786651611,
      "relation_loss": 0.003382595698349178,
      "selected_pairs": 26384.0,
      "step": 16330,
      "student_distance_mean": 24.499203491210938,
      "teacher_distance_mean": 0.854115092754364,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.003382595698349178,
      "weighted_relation_loss": 0.003382595698349178
    },
    {
      "action_query_batch_std": 0.37104407548904417,
      "action_query_norm": 42.19112281799316,
      "action_relation_loss": 0.0034666325896978377,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.285313376987839,
      "flow_matching_loss": 0.029096343927085398,
      "grad_norm": 0.13320451974868774,
      "h_stock_valid_norm": 98.4868064880371,
      "invalid_tail_samples": 21.2,
      "learning_rate": 4.675596787108653e-05,
      "loss": 0.0326,
      "positive_teacher_pairs": 27460.7,
      "query_h_norm_ratio": 0.42839740216732025,
      "relation_loss": 0.0034666325896978377,
      "selected_pairs": 27460.7,
      "step": 16340,
      "student_distance_mean": 24.661850357055663,
      "teacher_distance_mean": 0.8581232786178589,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.0034666325896978377,
      "weighted_relation_loss": 0.0034666325896978377
    },
    {
      "action_query_batch_std": 0.36574632227420806,
      "action_query_norm": 42.15890274047852,
      "action_relation_loss": 0.0034615165553987027,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.29466791393826,
      "flow_matching_loss": 0.02783771026879549,
      "grad_norm": 0.1407420039176941,
      "h_stock_valid_norm": 98.21923217773437,
      "invalid_tail_samples": 21.7,
      "learning_rate": 4.670097031950139e-05,
      "loss": 0.0313,
      "positive_teacher_pairs": 27337.5,
      "query_h_norm_ratio": 0.4292343854904175,
      "relation_loss": 0.0034615165553987027,
      "selected_pairs": 27337.5,
      "step": 16350,
      "student_distance_mean": 24.19458465576172,
      "teacher_distance_mean": 0.8452439427375793,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0034615165553987027,
      "weighted_relation_loss": 0.0034615165553987027
    },
    {
      "action_query_batch_std": 0.36846359372138976,
      "action_query_norm": 42.17997169494629,
      "action_relation_loss": 0.003272340865805745,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.30402245088868,
      "flow_matching_loss": 0.027402660995721816,
      "grad_norm": 0.1108214482665062,
      "h_stock_valid_norm": 98.5215934753418,
      "invalid_tail_samples": 20.7,
      "learning_rate": 4.664597677654838e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 27567.8,
      "query_h_norm_ratio": 0.4281372129917145,
      "relation_loss": 0.003272340865805745,
      "selected_pairs": 27567.8,
      "step": 16360,
      "student_distance_mean": 24.48375186920166,
      "teacher_distance_mean": 0.8608154714107513,
      "valid_action_samples": 235.3,
      "weighted_auxiliary_loss": 0.003272340865805745,
      "weighted_relation_loss": 0.003272340865805745
    },
    {
      "action_query_batch_std": 0.3687000095844269,
      "action_query_norm": 42.159394073486325,
      "action_relation_loss": 0.003386930888518691,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.313376987839103,
      "flow_matching_loss": 0.028575108759105204,
      "grad_norm": 0.10975037515163422,
      "h_stock_valid_norm": 99.08548278808594,
      "invalid_tail_samples": 20.6,
      "learning_rate": 4.659098730904986e-05,
      "loss": 0.032,
      "positive_teacher_pairs": 27596.4,
      "query_h_norm_ratio": 0.4254879206418991,
      "relation_loss": 0.003386930888518691,
      "selected_pairs": 27596.4,
      "step": 16370,
      "student_distance_mean": 24.50202865600586,
      "teacher_distance_mean": 0.8576019644737244,
      "valid_action_samples": 235.4,
      "weighted_auxiliary_loss": 0.003386930888518691,
      "weighted_relation_loss": 0.003386930888518691
    },
    {
      "action_query_batch_std": 0.3712283045053482,
      "action_query_norm": 42.17601051330566,
      "action_relation_loss": 0.0034254005644470454,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.322731524789523,
      "flow_matching_loss": 0.027267960272729395,
      "grad_norm": 0.11213012039661407,
      "h_stock_valid_norm": 98.94256591796875,
      "invalid_tail_samples": 25.0,
      "learning_rate": 4.6536001983823204e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 26575.6,
      "query_h_norm_ratio": 0.4262742042541504,
      "relation_loss": 0.0034254005644470454,
      "selected_pairs": 26575.6,
      "step": 16380,
      "student_distance_mean": 24.69382038116455,
      "teacher_distance_mean": 0.8600112438201905,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0034254005644470454,
      "weighted_relation_loss": 0.0034254005644470454
    },
    {
      "action_query_batch_std": 0.36744544804096224,
      "action_query_norm": 42.13127746582031,
      "action_relation_loss": 0.0033650960307568313,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.332086061739943,
      "flow_matching_loss": 0.02558322940021753,
      "grad_norm": 0.14194487035274506,
      "h_stock_valid_norm": 98.24185791015626,
      "invalid_tail_samples": 24.0,
      "learning_rate": 4.648102086768078e-05,
      "loss": 0.0289,
      "positive_teacher_pairs": 26801.4,
      "query_h_norm_ratio": 0.4288532942533493,
      "relation_loss": 0.0033650960307568313,
      "selected_pairs": 26801.4,
      "step": 16390,
      "student_distance_mean": 24.381030464172362,
      "teacher_distance_mean": 0.8568323016166687,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0033650960307568313,
      "weighted_relation_loss": 0.0033650960307568313
    },
    {
      "action_query_batch_std": 0.371404168009758,
      "action_query_norm": 42.1614803314209,
      "action_relation_loss": 0.0033344871597364547,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.341440598690365,
      "flow_matching_loss": 0.02677299063652754,
      "grad_norm": 0.1215781420469284,
      "h_stock_valid_norm": 98.1615089416504,
      "invalid_tail_samples": 23.3,
      "learning_rate": 4.6426044027429786e-05,
      "loss": 0.0301,
      "positive_teacher_pairs": 26965.0,
      "query_h_norm_ratio": 0.4295151591300964,
      "relation_loss": 0.0033344871597364547,
      "selected_pairs": 26965.0,
      "step": 16400,
      "student_distance_mean": 24.601388359069823,
      "teacher_distance_mean": 0.8625072956085205,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0033344871597364547,
      "weighted_relation_loss": 0.0033344871597364547
    },
    {
      "action_query_batch_std": 0.3679298430681229,
      "action_query_norm": 42.14541015625,
      "action_relation_loss": 0.0033374505117535593,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.350795135640785,
      "flow_matching_loss": 0.02579459734261036,
      "grad_norm": 0.08380992710590363,
      "h_stock_valid_norm": 98.44286880493163,
      "invalid_tail_samples": 27.0,
      "learning_rate": 4.637107152987234e-05,
      "loss": 0.0291,
      "positive_teacher_pairs": 26121.3,
      "query_h_norm_ratio": 0.4281220018863678,
      "relation_loss": 0.0033374505117535593,
      "selected_pairs": 26121.3,
      "step": 16410,
      "student_distance_mean": 24.59557819366455,
      "teacher_distance_mean": 0.8581142663955689,
      "valid_action_samples": 229.0,
      "weighted_auxiliary_loss": 0.0033374505117535593,
      "weighted_relation_loss": 0.0033374505117535593
    },
    {
      "action_query_batch_std": 0.36046276390552523,
      "action_query_norm": 42.11761245727539,
      "action_relation_loss": 0.003582382528111339,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.360149672591207,
      "flow_matching_loss": 0.028901709243655205,
      "grad_norm": 0.13591498136520386,
      "h_stock_valid_norm": 98.24734420776367,
      "invalid_tail_samples": 21.0,
      "learning_rate": 4.631610344180515e-05,
      "loss": 0.0325,
      "positive_teacher_pairs": 27503.0,
      "query_h_norm_ratio": 0.42869213223457336,
      "relation_loss": 0.003582382528111339,
      "selected_pairs": 27503.0,
      "step": 16420,
      "student_distance_mean": 24.015577697753905,
      "teacher_distance_mean": 0.8370947718620301,
      "valid_action_samples": 235.0,
      "weighted_auxiliary_loss": 0.003582382528111339,
      "weighted_relation_loss": 0.003582382528111339
    },
    {
      "action_query_batch_std": 0.3665233552455902,
      "action_query_norm": 42.13292922973633,
      "action_relation_loss": 0.0033958867425099015,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.369504209541628,
      "flow_matching_loss": 0.026682513020932674,
      "grad_norm": 0.12519718706607819,
      "h_stock_valid_norm": 97.84214401245117,
      "invalid_tail_samples": 23.9,
      "learning_rate": 4.626113983001965e-05,
      "loss": 0.0301,
      "positive_teacher_pairs": 26832.9,
      "query_h_norm_ratio": 0.4306242108345032,
      "relation_loss": 0.0033958867425099015,
      "selected_pairs": 26832.9,
      "step": 16430,
      "student_distance_mean": 24.32419834136963,
      "teacher_distance_mean": 0.8534272134304046,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0033958867425099015,
      "weighted_relation_loss": 0.0033958867425099015
    },
    {
      "action_query_batch_std": 0.3674194633960724,
      "action_query_norm": 42.13459587097168,
      "action_relation_loss": 0.0033322978066280483,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.378858746492048,
      "flow_matching_loss": 0.02739719711244106,
      "grad_norm": 0.10108437389135361,
      "h_stock_valid_norm": 97.59834060668945,
      "invalid_tail_samples": 23.2,
      "learning_rate": 4.620618076130182e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 26994.1,
      "query_h_norm_ratio": 0.4317159354686737,
      "relation_loss": 0.0033322978066280483,
      "selected_pairs": 26994.1,
      "step": 16440,
      "student_distance_mean": 24.37782211303711,
      "teacher_distance_mean": 0.8538198649883271,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0033322978066280483,
      "weighted_relation_loss": 0.0033322978066280483
    },
    {
      "action_query_batch_std": 0.37032913863658906,
      "action_query_norm": 42.14912986755371,
      "action_relation_loss": 0.0033152289222925903,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.38821328344247,
      "flow_matching_loss": 0.02877077031880617,
      "grad_norm": 0.1072344034910202,
      "h_stock_valid_norm": 97.67538986206054,
      "invalid_tail_samples": 24.8,
      "learning_rate": 4.6151226302432074e-05,
      "loss": 0.0321,
      "positive_teacher_pairs": 26627.0,
      "query_h_norm_ratio": 0.4315235078334808,
      "relation_loss": 0.0033152289222925903,
      "selected_pairs": 26627.0,
      "step": 16450,
      "student_distance_mean": 24.46745185852051,
      "teacher_distance_mean": 0.8544259488582611,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0033152289222925903,
      "weighted_relation_loss": 0.0033152289222925903
    },
    {
      "action_query_batch_std": 0.37743220925331117,
      "action_query_norm": 42.18508758544922,
      "action_relation_loss": 0.0034314409596845506,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.39756782039289,
      "flow_matching_loss": 0.02920955065637827,
      "grad_norm": 0.11702118068933487,
      "h_stock_valid_norm": 98.3836883544922,
      "invalid_tail_samples": 24.2,
      "learning_rate": 4.6096276520185304e-05,
      "loss": 0.0326,
      "positive_teacher_pairs": 26763.7,
      "query_h_norm_ratio": 0.4287832140922546,
      "relation_loss": 0.0034314409596845506,
      "selected_pairs": 26763.7,
      "step": 16460,
      "student_distance_mean": 24.922007751464843,
      "teacher_distance_mean": 0.8701772570610047,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0034314409596845506,
      "weighted_relation_loss": 0.0034314409596845506
    },
    {
      "action_query_batch_std": 0.3630890816450119,
      "action_query_norm": 42.12614860534668,
      "action_relation_loss": 0.003428851556964219,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.406922357343312,
      "flow_matching_loss": 0.027260312624275685,
      "grad_norm": 0.12173790484666824,
      "h_stock_valid_norm": 98.52805938720704,
      "invalid_tail_samples": 20.7,
      "learning_rate": 4.604133148133066e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 27572.1,
      "query_h_norm_ratio": 0.42755970656871795,
      "relation_loss": 0.003428851556964219,
      "selected_pairs": 27572.1,
      "step": 16470,
      "student_distance_mean": 24.097291946411133,
      "teacher_distance_mean": 0.8460890829563141,
      "valid_action_samples": 235.3,
      "weighted_auxiliary_loss": 0.003428851556964219,
      "weighted_relation_loss": 0.003428851556964219
    },
    {
      "action_query_batch_std": 0.3698056548833847,
      "action_query_norm": 42.14903106689453,
      "action_relation_loss": 0.00328300166875124,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.416276894293732,
      "flow_matching_loss": 0.026280737668275832,
      "grad_norm": 0.12676285207271576,
      "h_stock_valid_norm": 98.34718017578125,
      "invalid_tail_samples": 25.2,
      "learning_rate": 4.598639125263155e-05,
      "loss": 0.0296,
      "positive_teacher_pairs": 26531.1,
      "query_h_norm_ratio": 0.42857798635959626,
      "relation_loss": 0.00328300166875124,
      "selected_pairs": 26531.1,
      "step": 16480,
      "student_distance_mean": 24.66131763458252,
      "teacher_distance_mean": 0.862420243024826,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.00328300166875124,
      "weighted_relation_loss": 0.00328300166875124
    },
    {
      "action_query_batch_std": 0.36469348073005675,
      "action_query_norm": 42.10180778503418,
      "action_relation_loss": 0.0037043163087219,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.425631431244154,
      "flow_matching_loss": 0.028542117774486543,
      "grad_norm": 0.11762524396181107,
      "h_stock_valid_norm": 98.16589813232422,
      "invalid_tail_samples": 23.4,
      "learning_rate": 4.593145590084553e-05,
      "loss": 0.0322,
      "positive_teacher_pairs": 26943.6,
      "query_h_norm_ratio": 0.42888695895671847,
      "relation_loss": 0.0037043163087219,
      "selected_pairs": 26943.6,
      "step": 16490,
      "student_distance_mean": 24.277724075317384,
      "teacher_distance_mean": 0.8492116451263427,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0037043163087219,
      "weighted_relation_loss": 0.0037043163087219
    },
    {
      "action_query_batch_std": 0.369326239824295,
      "action_query_norm": 42.12326545715332,
      "action_relation_loss": 0.003433503583073616,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.434985968194574,
      "flow_matching_loss": 0.02678361292928457,
      "grad_norm": 0.11729495227336884,
      "h_stock_valid_norm": 98.58715286254883,
      "invalid_tail_samples": 23.2,
      "learning_rate": 4.58765254927242e-05,
      "loss": 0.0302,
      "positive_teacher_pairs": 26988.8,
      "query_h_norm_ratio": 0.42727283835411073,
      "relation_loss": 0.003433503583073616,
      "selected_pairs": 26988.8,
      "step": 16500,
      "student_distance_mean": 24.506174278259277,
      "teacher_distance_mean": 0.8539893984794616,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.003433503583073616,
      "weighted_relation_loss": 0.003433503583073616
    },
    {
      "action_query_batch_std": 0.365445938706398,
      "action_query_norm": 42.11574859619141,
      "action_relation_loss": 0.003269850159995258,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.444340505144995,
      "flow_matching_loss": 0.027274400182068347,
      "grad_norm": 0.09399476647377014,
      "h_stock_valid_norm": 99.16539993286133,
      "invalid_tail_samples": 25.2,
      "learning_rate": 4.582160009501323e-05,
      "loss": 0.0305,
      "positive_teacher_pairs": 26527.6,
      "query_h_norm_ratio": 0.424703973531723,
      "relation_loss": 0.003269850159995258,
      "selected_pairs": 26527.6,
      "step": 16510,
      "student_distance_mean": 24.38475093841553,
      "teacher_distance_mean": 0.8533028602600098,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.003269850159995258,
      "weighted_relation_loss": 0.003269850159995258
    },
    {
      "action_query_batch_std": 0.3660746246576309,
      "action_query_norm": 42.10509757995605,
      "action_relation_loss": 0.003457151562906802,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.453695042095417,
      "flow_matching_loss": 0.02804997731000185,
      "grad_norm": 0.09071727097034454,
      "h_stock_valid_norm": 98.84693603515625,
      "invalid_tail_samples": 21.3,
      "learning_rate": 4.5766679774452143e-05,
      "loss": 0.0315,
      "positive_teacher_pairs": 27430.7,
      "query_h_norm_ratio": 0.42597348988056183,
      "relation_loss": 0.003457151562906802,
      "selected_pairs": 27430.7,
      "step": 16520,
      "student_distance_mean": 24.220600509643553,
      "teacher_distance_mean": 0.850846266746521,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.003457151562906802,
      "weighted_relation_loss": 0.003457151562906802
    },
    {
      "action_query_batch_std": 0.369197753071785,
      "action_query_norm": 42.121005630493165,
      "action_relation_loss": 0.00344561783131212,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.463049579045837,
      "flow_matching_loss": 0.027700531855225564,
      "grad_norm": 0.10176273435354233,
      "h_stock_valid_norm": 98.81494903564453,
      "invalid_tail_samples": 26.5,
      "learning_rate": 4.571176459777431e-05,
      "loss": 0.0311,
      "positive_teacher_pairs": 26241.3,
      "query_h_norm_ratio": 0.4262639135122299,
      "relation_loss": 0.00344561783131212,
      "selected_pairs": 26241.3,
      "step": 16530,
      "student_distance_mean": 24.55042381286621,
      "teacher_distance_mean": 0.8592863261699677,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.00344561783131212,
      "weighted_relation_loss": 0.00344561783131212
    },
    {
      "action_query_batch_std": 0.3686649650335312,
      "action_query_norm": 42.131792068481445,
      "action_relation_loss": 0.0034389936598017814,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.472404115996259,
      "flow_matching_loss": 0.027258863672614098,
      "grad_norm": 0.10393153876066208,
      "h_stock_valid_norm": 98.86575927734376,
      "invalid_tail_samples": 23.3,
      "learning_rate": 4.5656854631706846e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 26967.1,
      "query_h_norm_ratio": 0.42615805864334105,
      "relation_loss": 0.0034389936598017814,
      "selected_pairs": 26967.1,
      "step": 16540,
      "student_distance_mean": 24.4143367767334,
      "teacher_distance_mean": 0.8488621294498444,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0034389936598017814,
      "weighted_relation_loss": 0.0034389936598017814
    },
    {
      "action_query_batch_std": 0.366060808300972,
      "action_query_norm": 42.11184844970703,
      "action_relation_loss": 0.003482858114875853,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.481758652946679,
      "flow_matching_loss": 0.029088368453085423,
      "grad_norm": 0.1013958752155304,
      "h_stock_valid_norm": 98.94407424926757,
      "invalid_tail_samples": 21.6,
      "learning_rate": 4.560194994297054e-05,
      "loss": 0.0326,
      "positive_teacher_pairs": 27369.5,
      "query_h_norm_ratio": 0.42561748027801516,
      "relation_loss": 0.003482858114875853,
      "selected_pairs": 27369.5,
      "step": 16550,
      "student_distance_mean": 24.26728057861328,
      "teacher_distance_mean": 0.8483339965343475,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.003482858114875853,
      "weighted_relation_loss": 0.003482858114875853
    },
    {
      "action_query_batch_std": 0.3702181339263916,
      "action_query_norm": 42.143301010131836,
      "action_relation_loss": 0.003325372003018856,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.4911131898971,
      "flow_matching_loss": 0.02511575259268284,
      "grad_norm": 0.11005431413650513,
      "h_stock_valid_norm": 98.33261489868164,
      "invalid_tail_samples": 25.5,
      "learning_rate": 4.554705059827974e-05,
      "loss": 0.0284,
      "positive_teacher_pairs": 26457.3,
      "query_h_norm_ratio": 0.428581964969635,
      "relation_loss": 0.003325372003018856,
      "selected_pairs": 26457.3,
      "step": 16560,
      "student_distance_mean": 24.657391357421876,
      "teacher_distance_mean": 0.8604857444763183,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.003325372003018856,
      "weighted_relation_loss": 0.003325372003018856
    },
    {
      "action_query_batch_std": 0.36464929282665254,
      "action_query_norm": 42.106794357299805,
      "action_relation_loss": 0.0033159459475427867,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.500467726847521,
      "flow_matching_loss": 0.02743746414780617,
      "grad_norm": 0.10193876177072525,
      "h_stock_valid_norm": 98.39332504272461,
      "invalid_tail_samples": 23.9,
      "learning_rate": 4.5492156664342365e-05,
      "loss": 0.0308,
      "positive_teacher_pairs": 26831.7,
      "query_h_norm_ratio": 0.427945539355278,
      "relation_loss": 0.0033159459475427867,
      "selected_pairs": 26831.7,
      "step": 16570,
      "student_distance_mean": 24.217823791503907,
      "teacher_distance_mean": 0.844078540802002,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0033159459475427867,
      "weighted_relation_loss": 0.0033159459475427867
    },
    {
      "action_query_batch_std": 0.3657674491405487,
      "action_query_norm": 42.09782028198242,
      "action_relation_loss": 0.0035924725234508514,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.509822263797941,
      "flow_matching_loss": 0.028293533809483053,
      "grad_norm": 0.1300988793373108,
      "h_stock_valid_norm": 98.69308547973633,
      "invalid_tail_samples": 25.0,
      "learning_rate": 4.5437268207859695e-05,
      "loss": 0.0319,
      "positive_teacher_pairs": 26573.1,
      "query_h_norm_ratio": 0.4265555262565613,
      "relation_loss": 0.0035924725234508514,
      "selected_pairs": 26573.1,
      "step": 16580,
      "student_distance_mean": 24.31162281036377,
      "teacher_distance_mean": 0.8512981414794922,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0035924725234508514,
      "weighted_relation_loss": 0.0035924725234508514
    },
    {
      "action_query_batch_std": 0.3685363411903381,
      "action_query_norm": 42.11742324829102,
      "action_relation_loss": 0.0034863857552409174,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.519176800748363,
      "flow_matching_loss": 0.02768109105527401,
      "grad_norm": 0.11420222371816635,
      "h_stock_valid_norm": 98.93998947143555,
      "invalid_tail_samples": 22.0,
      "learning_rate": 4.538238529552641e-05,
      "loss": 0.0312,
      "positive_teacher_pairs": 27269.3,
      "query_h_norm_ratio": 0.4256895869970322,
      "relation_loss": 0.0034863857552409174,
      "selected_pairs": 27269.3,
      "step": 16590,
      "student_distance_mean": 24.416251945495606,
      "teacher_distance_mean": 0.8515370070934296,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.0034863857552409174,
      "weighted_relation_loss": 0.0034863857552409174
    },
    {
      "action_query_batch_std": 0.37141880989074705,
      "action_query_norm": 42.12929496765137,
      "action_relation_loss": 0.003332370985299349,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.528531337698784,
      "flow_matching_loss": 0.02690014336258173,
      "grad_norm": 0.09440203011035919,
      "h_stock_valid_norm": 98.71115646362304,
      "invalid_tail_samples": 23.1,
      "learning_rate": 4.5327507994030394e-05,
      "loss": 0.0302,
      "positive_teacher_pairs": 27010.5,
      "query_h_norm_ratio": 0.4267997443675995,
      "relation_loss": 0.003332370985299349,
      "selected_pairs": 27010.5,
      "step": 16600,
      "student_distance_mean": 24.573775863647462,
      "teacher_distance_mean": 0.8563944280147553,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.003332370985299349,
      "weighted_relation_loss": 0.003332370985299349
    },
    {
      "action_query_batch_std": 0.3735478162765503,
      "action_query_norm": 42.13366203308105,
      "action_relation_loss": 0.0035033069318160413,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.537885874649206,
      "flow_matching_loss": 0.026454827934503555,
      "grad_norm": 0.10748504102230072,
      "h_stock_valid_norm": 97.98267135620117,
      "invalid_tail_samples": 24.0,
      "learning_rate": 4.527263637005273e-05,
      "loss": 0.03,
      "positive_teacher_pairs": 26805.8,
      "query_h_norm_ratio": 0.4300157934427261,
      "relation_loss": 0.0035033069318160413,
      "selected_pairs": 26805.8,
      "step": 16610,
      "student_distance_mean": 24.8007474899292,
      "teacher_distance_mean": 0.8672613322734832,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0035033069318160413,
      "weighted_relation_loss": 0.0035033069318160413
    },
    {
      "action_query_batch_std": 0.3677645683288574,
      "action_query_norm": 42.08515892028809,
      "action_relation_loss": 0.0034411583794280885,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.547240411599626,
      "flow_matching_loss": 0.026750822737812995,
      "grad_norm": 0.11295399814844131,
      "h_stock_valid_norm": 97.80359497070313,
      "invalid_tail_samples": 24.4,
      "learning_rate": 4.521777049026766e-05,
      "loss": 0.0302,
      "positive_teacher_pairs": 26711.9,
      "query_h_norm_ratio": 0.4303048670291901,
      "relation_loss": 0.0034411583794280885,
      "selected_pairs": 26711.9,
      "step": 16620,
      "student_distance_mean": 24.41774215698242,
      "teacher_distance_mean": 0.8539333522319794,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0034411583794280885,
      "weighted_relation_loss": 0.0034411583794280885
    },
    {
      "action_query_batch_std": 0.3704517543315887,
      "action_query_norm": 42.107888793945314,
      "action_relation_loss": 0.003393308003433049,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.556594948550046,
      "flow_matching_loss": 0.026594674587249754,
      "grad_norm": 0.08457762748003006,
      "h_stock_valid_norm": 97.96983413696289,
      "invalid_tail_samples": 27.3,
      "learning_rate": 4.5162910421342386e-05,
      "loss": 0.03,
      "positive_teacher_pairs": 26051.0,
      "query_h_norm_ratio": 0.42981420159339906,
      "relation_loss": 0.003393308003433049,
      "selected_pairs": 26051.0,
      "step": 16630,
      "student_distance_mean": 24.565716361999513,
      "teacher_distance_mean": 0.8558913171291351,
      "valid_action_samples": 228.7,
      "weighted_auxiliary_loss": 0.003393308003433049,
      "weighted_relation_loss": 0.003393308003433049
    },
    {
      "action_query_batch_std": 0.37137049436569214,
      "action_query_norm": 42.10997695922852,
      "action_relation_loss": 0.0033201931044459345,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.565949485500468,
      "flow_matching_loss": 0.02666052244603634,
      "grad_norm": 0.1252656728029251,
      "h_stock_valid_norm": 98.47429580688477,
      "invalid_tail_samples": 23.4,
      "learning_rate": 4.510805622993706e-05,
      "loss": 0.03,
      "positive_teacher_pairs": 26943.8,
      "query_h_norm_ratio": 0.42762952446937563,
      "relation_loss": 0.0033201931044459345,
      "selected_pairs": 26943.8,
      "step": 16640,
      "student_distance_mean": 24.69671630859375,
      "teacher_distance_mean": 0.8603367328643798,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0033201931044459345,
      "weighted_relation_loss": 0.0033201931044459345
    },
    {
      "action_query_batch_std": 0.3651233226060867,
      "action_query_norm": 42.070214080810544,
      "action_relation_loss": 0.0033289317274466158,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.575304022450888,
      "flow_matching_loss": 0.027967104502022267,
      "grad_norm": 0.1237742006778717,
      "h_stock_valid_norm": 98.8758171081543,
      "invalid_tail_samples": 25.9,
      "learning_rate": 4.505320798270467e-05,
      "loss": 0.0313,
      "positive_teacher_pairs": 26361.5,
      "query_h_norm_ratio": 0.4254884958267212,
      "relation_loss": 0.0033289317274466158,
      "selected_pairs": 26361.5,
      "step": 16650,
      "student_distance_mean": 24.393505668640138,
      "teacher_distance_mean": 0.8583408415317535,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.0033289317274466158,
      "weighted_relation_loss": 0.0033289317274466158
    },
    {
      "action_query_batch_std": 0.3698570400476456,
      "action_query_norm": 42.09440231323242,
      "action_relation_loss": 0.0032520468113943933,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.58465855940131,
      "flow_matching_loss": 0.026225783675909043,
      "grad_norm": 0.08390016853809357,
      "h_stock_valid_norm": 98.56593704223633,
      "invalid_tail_samples": 22.6,
      "learning_rate": 4.499836574629105e-05,
      "loss": 0.0295,
      "positive_teacher_pairs": 27128.7,
      "query_h_norm_ratio": 0.42707250714302064,
      "relation_loss": 0.0032520468113943933,
      "selected_pairs": 27128.7,
      "step": 16660,
      "student_distance_mean": 24.495741844177246,
      "teacher_distance_mean": 0.8554823100566864,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.0032520468113943933,
      "weighted_relation_loss": 0.0032520468113943933
    },
    {
      "action_query_batch_std": 0.3710401266813278,
      "action_query_norm": 42.114989852905275,
      "action_relation_loss": 0.003320154012180865,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.59401309635173,
      "flow_matching_loss": 0.02685294169932604,
      "grad_norm": 0.1490928679704666,
      "h_stock_valid_norm": 98.03767318725586,
      "invalid_tail_samples": 22.6,
      "learning_rate": 4.494352958733465e-05,
      "loss": 0.0302,
      "positive_teacher_pairs": 27134.4,
      "query_h_norm_ratio": 0.4295853853225708,
      "relation_loss": 0.003320154012180865,
      "selected_pairs": 27134.4,
      "step": 16670,
      "student_distance_mean": 24.560794067382812,
      "teacher_distance_mean": 0.8543682098388672,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.003320154012180865,
      "weighted_relation_loss": 0.003320154012180865
    },
    {
      "action_query_batch_std": 0.369013375043869,
      "action_query_norm": 42.08127059936523,
      "action_relation_loss": 0.003277900186367333,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.603367633302152,
      "flow_matching_loss": 0.02745295371860266,
      "grad_norm": 0.09189144521951675,
      "h_stock_valid_norm": 97.67435913085937,
      "invalid_tail_samples": 24.2,
      "learning_rate": 4.488869957246663e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 26765.7,
      "query_h_norm_ratio": 0.4308340311050415,
      "relation_loss": 0.003277900186367333,
      "selected_pairs": 26765.7,
      "step": 16680,
      "student_distance_mean": 24.486373519897462,
      "teacher_distance_mean": 0.856851464509964,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.003277900186367333,
      "weighted_relation_loss": 0.003277900186367333
    },
    {
      "action_query_batch_std": 0.3658041566610336,
      "action_query_norm": 42.05382881164551,
      "action_relation_loss": 0.0033596235793083907,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.612722170252573,
      "flow_matching_loss": 0.025806276127696037,
      "grad_norm": 0.11322961747646332,
      "h_stock_valid_norm": 97.68572769165038,
      "invalid_tail_samples": 27.2,
      "learning_rate": 4.4833875768310575e-05,
      "loss": 0.0292,
      "positive_teacher_pairs": 26068.2,
      "query_h_norm_ratio": 0.4305055797100067,
      "relation_loss": 0.0033596235793083907,
      "selected_pairs": 26068.2,
      "step": 16690,
      "student_distance_mean": 24.392297172546385,
      "teacher_distance_mean": 0.8568349421024323,
      "valid_action_samples": 228.8,
      "weighted_auxiliary_loss": 0.0033596235793083907,
      "weighted_relation_loss": 0.0033596235793083907
    },
    {
      "action_query_batch_std": 0.3660913646221161,
      "action_query_norm": 42.04290580749512,
      "action_relation_loss": 0.0033205810468643904,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.622076707202993,
      "flow_matching_loss": 0.027391642704606055,
      "grad_norm": 0.11813046783208847,
      "h_stock_valid_norm": 97.44355621337891,
      "invalid_tail_samples": 23.5,
      "learning_rate": 4.47790582414826e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 26926.0,
      "query_h_norm_ratio": 0.43145962357521056,
      "relation_loss": 0.0033205810468643904,
      "selected_pairs": 26926.0,
      "step": 16700,
      "student_distance_mean": 24.347737503051757,
      "teacher_distance_mean": 0.8553817808628082,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0033205810468643904,
      "weighted_relation_loss": 0.0033205810468643904
    },
    {
      "action_query_batch_std": 0.36670414805412294,
      "action_query_norm": 42.063826370239255,
      "action_relation_loss": 0.0032925499603152276,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.631431244153415,
      "flow_matching_loss": 0.027953981794416903,
      "grad_norm": 0.10223790258169174,
      "h_stock_valid_norm": 97.5034194946289,
      "invalid_tail_samples": 21.0,
      "learning_rate": 4.472424705859114e-05,
      "loss": 0.0312,
      "positive_teacher_pairs": 27500.1,
      "query_h_norm_ratio": 0.4314126282930374,
      "relation_loss": 0.0032925499603152276,
      "selected_pairs": 27500.1,
      "step": 16710,
      "student_distance_mean": 24.306847763061523,
      "teacher_distance_mean": 0.8536655128002166,
      "valid_action_samples": 235.0,
      "weighted_auxiliary_loss": 0.0032925499603152276,
      "weighted_relation_loss": 0.0032925499603152276
    },
    {
      "action_query_batch_std": 0.36671191155910493,
      "action_query_norm": 42.0728099822998,
      "action_relation_loss": 0.003330624592490494,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.640785781103835,
      "flow_matching_loss": 0.02832864075899124,
      "grad_norm": 0.12834617495536804,
      "h_stock_valid_norm": 98.07755584716797,
      "invalid_tail_samples": 25.8,
      "learning_rate": 4.466944228623701e-05,
      "loss": 0.0317,
      "positive_teacher_pairs": 26386.0,
      "query_h_norm_ratio": 0.4289765626192093,
      "relation_loss": 0.003330624592490494,
      "selected_pairs": 26386.0,
      "step": 16720,
      "student_distance_mean": 24.209766578674316,
      "teacher_distance_mean": 0.8494864344596863,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.003330624592490494,
      "weighted_relation_loss": 0.003330624592490494
    },
    {
      "action_query_batch_std": 0.36976305544376376,
      "action_query_norm": 42.09824562072754,
      "action_relation_loss": 0.003237869474105537,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.650140318054257,
      "flow_matching_loss": 0.02706499267369509,
      "grad_norm": 0.11784136295318604,
      "h_stock_valid_norm": 98.70220718383788,
      "invalid_tail_samples": 26.2,
      "learning_rate": 4.461464399101312e-05,
      "loss": 0.0303,
      "positive_teacher_pairs": 26295.9,
      "query_h_norm_ratio": 0.42652090191841124,
      "relation_loss": 0.003237869474105537,
      "selected_pairs": 26295.9,
      "step": 16730,
      "student_distance_mean": 24.53924102783203,
      "teacher_distance_mean": 0.859980309009552,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.003237869474105537,
      "weighted_relation_loss": 0.003237869474105537
    },
    {
      "action_query_batch_std": 0.36990213990211485,
      "action_query_norm": 42.08327293395996,
      "action_relation_loss": 0.0033637480810284615,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.659494855004677,
      "flow_matching_loss": 0.02854931093752384,
      "grad_norm": 0.10580393671989441,
      "h_stock_valid_norm": 98.6792999267578,
      "invalid_tail_samples": 22.2,
      "learning_rate": 4.45598522395046e-05,
      "loss": 0.0319,
      "positive_teacher_pairs": 27226.0,
      "query_h_norm_ratio": 0.42646816074848176,
      "relation_loss": 0.0033637480810284615,
      "selected_pairs": 27226.0,
      "step": 16740,
      "student_distance_mean": 24.471387672424317,
      "teacher_distance_mean": 0.8586346983909607,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0033637480810284615,
      "weighted_relation_loss": 0.0033637480810284615
    },
    {
      "action_query_batch_std": 0.36723542511463164,
      "action_query_norm": 42.0639144897461,
      "action_relation_loss": 0.003186812344938517,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.668849391955098,
      "flow_matching_loss": 0.02815015334635973,
      "grad_norm": 0.10466455668210983,
      "h_stock_valid_norm": 98.52211151123046,
      "invalid_tail_samples": 23.3,
      "learning_rate": 4.4505067098288577e-05,
      "loss": 0.0313,
      "positive_teacher_pairs": 26974.2,
      "query_h_norm_ratio": 0.4269505113363266,
      "relation_loss": 0.003186812344938517,
      "selected_pairs": 26974.2,
      "step": 16750,
      "student_distance_mean": 24.368609046936037,
      "teacher_distance_mean": 0.8510184526443482,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.003186812344938517,
      "weighted_relation_loss": 0.003186812344938517
    },
    {
      "action_query_batch_std": 0.36714867651462557,
      "action_query_norm": 42.05079612731934,
      "action_relation_loss": 0.003413308225572109,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.67820392890552,
      "flow_matching_loss": 0.02786331344395876,
      "grad_norm": 0.1142314076423645,
      "h_stock_valid_norm": 98.39027481079101,
      "invalid_tail_samples": 26.1,
      "learning_rate": 4.445028863393417e-05,
      "loss": 0.0313,
      "positive_teacher_pairs": 26320.4,
      "query_h_norm_ratio": 0.4273891866207123,
      "relation_loss": 0.003413308225572109,
      "selected_pairs": 26320.4,
      "step": 16760,
      "student_distance_mean": 24.424539375305176,
      "teacher_distance_mean": 0.8577880144119263,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.003413308225572109,
      "weighted_relation_loss": 0.003413308225572109
    },
    {
      "action_query_batch_std": 0.3663324862718582,
      "action_query_norm": 42.059397506713864,
      "action_relation_loss": 0.0033886408200487494,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.68755846585594,
      "flow_matching_loss": 0.02536956835538149,
      "grad_norm": 0.09111382067203522,
      "h_stock_valid_norm": 98.06232986450195,
      "invalid_tail_samples": 24.6,
      "learning_rate": 4.439551691300236e-05,
      "loss": 0.0288,
      "positive_teacher_pairs": 26663.1,
      "query_h_norm_ratio": 0.4289101392030716,
      "relation_loss": 0.0033886408200487494,
      "selected_pairs": 26663.1,
      "step": 16770,
      "student_distance_mean": 24.33364067077637,
      "teacher_distance_mean": 0.8510065138339996,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0033886408200487494,
      "weighted_relation_loss": 0.0033886408200487494
    },
    {
      "action_query_batch_std": 0.36380737125873563,
      "action_query_norm": 42.043513488769534,
      "action_relation_loss": 0.0033518619602546096,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.696913002806362,
      "flow_matching_loss": 0.025372128188610076,
      "grad_norm": 0.09195270389318466,
      "h_stock_valid_norm": 97.41243896484374,
      "invalid_tail_samples": 22.2,
      "learning_rate": 4.434075200204599e-05,
      "loss": 0.0287,
      "positive_teacher_pairs": 27221.9,
      "query_h_norm_ratio": 0.43160520195961,
      "relation_loss": 0.0033518619602546096,
      "selected_pairs": 27221.9,
      "step": 16780,
      "student_distance_mean": 24.162517929077147,
      "teacher_distance_mean": 0.8443813323974609,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0033518619602546096,
      "weighted_relation_loss": 0.0033518619602546096
    },
    {
      "action_query_batch_std": 0.3750752002000809,
      "action_query_norm": 42.088599014282224,
      "action_relation_loss": 0.0031737906159833074,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.706267539756782,
      "flow_matching_loss": 0.025749505124986173,
      "grad_norm": 0.10416492074728012,
      "h_stock_valid_norm": 97.49630966186524,
      "invalid_tail_samples": 23.5,
      "learning_rate": 4.428599396760956e-05,
      "loss": 0.0289,
      "positive_teacher_pairs": 26919.6,
      "query_h_norm_ratio": 0.4316968023777008,
      "relation_loss": 0.0031737906159833074,
      "selected_pairs": 26919.6,
      "step": 16790,
      "student_distance_mean": 24.923043060302735,
      "teacher_distance_mean": 0.8757468283176422,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0031737906159833074,
      "weighted_relation_loss": 0.0031737906159833074
    },
    {
      "action_query_batch_std": 0.36990995407104493,
      "action_query_norm": 42.06708602905273,
      "action_relation_loss": 0.003352008434012532,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.715622076707202,
      "flow_matching_loss": 0.027532186172902585,
      "grad_norm": 0.10343784838914871,
      "h_stock_valid_norm": 97.94658126831055,
      "invalid_tail_samples": 23.9,
      "learning_rate": 4.423124287622926e-05,
      "loss": 0.0309,
      "positive_teacher_pairs": 26828.6,
      "query_h_norm_ratio": 0.4294928401708603,
      "relation_loss": 0.003352008434012532,
      "selected_pairs": 26828.6,
      "step": 16800,
      "student_distance_mean": 24.44271945953369,
      "teacher_distance_mean": 0.8567637860774994,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.003352008434012532,
      "weighted_relation_loss": 0.003352008434012532
    },
    {
      "action_query_batch_std": 0.3700698524713516,
      "action_query_norm": 42.06475982666016,
      "action_relation_loss": 0.003233236144296825,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.724976613657624,
      "flow_matching_loss": 0.027399713173508644,
      "grad_norm": 0.10937675088644028,
      "h_stock_valid_norm": 97.4498405456543,
      "invalid_tail_samples": 22.8,
      "learning_rate": 4.417649879443282e-05,
      "loss": 0.0306,
      "positive_teacher_pairs": 27083.9,
      "query_h_norm_ratio": 0.43165984749794006,
      "relation_loss": 0.003233236144296825,
      "selected_pairs": 27083.9,
      "step": 16810,
      "student_distance_mean": 24.511929321289063,
      "teacher_distance_mean": 0.8592478811740876,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.003233236144296825,
      "weighted_relation_loss": 0.003233236144296825
    },
    {
      "action_query_batch_std": 0.3671672612428665,
      "action_query_norm": 42.04961013793945,
      "action_relation_loss": 0.0031863137148320673,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.734331150608044,
      "flow_matching_loss": 0.02612305525690317,
      "grad_norm": 0.08509501069784164,
      "h_stock_valid_norm": 97.21794128417969,
      "invalid_tail_samples": 23.9,
      "learning_rate": 4.4121761788739444e-05,
      "loss": 0.0293,
      "positive_teacher_pairs": 26829.6,
      "query_h_norm_ratio": 0.4325343519449234,
      "relation_loss": 0.0031863137148320673,
      "selected_pairs": 26829.6,
      "step": 16820,
      "student_distance_mean": 24.317233276367187,
      "teacher_distance_mean": 0.853325468301773,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0031863137148320673,
      "weighted_relation_loss": 0.0031863137148320673
    },
    {
      "action_query_batch_std": 0.36576820611953736,
      "action_query_norm": 42.03378067016602,
      "action_relation_loss": 0.003128903009928763,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.743685687558466,
      "flow_matching_loss": 0.026820472814142703,
      "grad_norm": 0.10433905571699142,
      "h_stock_valid_norm": 97.88031387329102,
      "invalid_tail_samples": 22.1,
      "learning_rate": 4.4067031925659805e-05,
      "loss": 0.0299,
      "positive_teacher_pairs": 27243.5,
      "query_h_norm_ratio": 0.4294439643621445,
      "relation_loss": 0.003128903009928763,
      "selected_pairs": 27243.5,
      "step": 16830,
      "student_distance_mean": 24.15824775695801,
      "teacher_distance_mean": 0.8457901239395141,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.003128903009928763,
      "weighted_relation_loss": 0.003128903009928763
    },
    {
      "action_query_batch_std": 0.3684434562921524,
      "action_query_norm": 42.0489616394043,
      "action_relation_loss": 0.0034391464898362754,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.753040224508887,
      "flow_matching_loss": 0.02555875275284052,
      "grad_norm": 0.10189972817897797,
      "h_stock_valid_norm": 97.69954299926758,
      "invalid_tail_samples": 21.5,
      "learning_rate": 4.4012309271695816e-05,
      "loss": 0.029,
      "positive_teacher_pairs": 27393.9,
      "query_h_norm_ratio": 0.43039298951625826,
      "relation_loss": 0.0034391464898362754,
      "selected_pairs": 27393.9,
      "step": 16840,
      "student_distance_mean": 24.272682762145998,
      "teacher_distance_mean": 0.8488113403320312,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.0034391464898362754,
      "weighted_relation_loss": 0.0034391464898362754
    },
    {
      "action_query_batch_std": 0.365917032957077,
      "action_query_norm": 42.03849983215332,
      "action_relation_loss": 0.003208955796435475,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.762394761459309,
      "flow_matching_loss": 0.025892152450978756,
      "grad_norm": 0.10818881541490555,
      "h_stock_valid_norm": 98.03362884521485,
      "invalid_tail_samples": 21.1,
      "learning_rate": 4.395759389334067e-05,
      "loss": 0.0291,
      "positive_teacher_pairs": 27478.0,
      "query_h_norm_ratio": 0.42882082760334017,
      "relation_loss": 0.003208955796435475,
      "selected_pairs": 27478.0,
      "step": 16850,
      "student_distance_mean": 24.253173637390137,
      "teacher_distance_mean": 0.8523328840732575,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.003208955796435475,
      "weighted_relation_loss": 0.003208955796435475
    },
    {
      "action_query_batch_std": 0.3715310484170914,
      "action_query_norm": 42.073483276367185,
      "action_relation_loss": 0.003254495607689023,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.771749298409729,
      "flow_matching_loss": 0.025527724623680116,
      "grad_norm": 0.09365658462047577,
      "h_stock_valid_norm": 98.48377685546875,
      "invalid_tail_samples": 22.8,
      "learning_rate": 4.3902885857078685e-05,
      "loss": 0.0288,
      "positive_teacher_pairs": 27087.4,
      "query_h_norm_ratio": 0.4272142767906189,
      "relation_loss": 0.003254495607689023,
      "selected_pairs": 27087.4,
      "step": 16860,
      "student_distance_mean": 24.636596488952637,
      "teacher_distance_mean": 0.8694019675254822,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.003254495607689023,
      "weighted_relation_loss": 0.003254495607689023
    },
    {
      "action_query_batch_std": 0.36959314346313477,
      "action_query_norm": 42.05750160217285,
      "action_relation_loss": 0.003167901746928692,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.781103835360149,
      "flow_matching_loss": 0.026242772117257118,
      "grad_norm": 0.09966401010751724,
      "h_stock_valid_norm": 98.14542999267579,
      "invalid_tail_samples": 23.5,
      "learning_rate": 4.384818522938531e-05,
      "loss": 0.0294,
      "positive_teacher_pairs": 26917.1,
      "query_h_norm_ratio": 0.42852537631988524,
      "relation_loss": 0.003167901746928692,
      "selected_pairs": 26917.1,
      "step": 16870,
      "student_distance_mean": 24.42200469970703,
      "teacher_distance_mean": 0.8582234442234039,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.003167901746928692,
      "weighted_relation_loss": 0.003167901746928692
    },
    {
      "action_query_batch_std": 0.3685261160135269,
      "action_query_norm": 42.04681243896484,
      "action_relation_loss": 0.003265469241887331,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.790458372310571,
      "flow_matching_loss": 0.02851932980120182,
      "grad_norm": 0.12146008759737015,
      "h_stock_valid_norm": 98.19104614257813,
      "invalid_tail_samples": 21.5,
      "learning_rate": 4.379349207672695e-05,
      "loss": 0.0318,
      "positive_teacher_pairs": 27385.7,
      "query_h_norm_ratio": 0.4282166212797165,
      "relation_loss": 0.003265469241887331,
      "selected_pairs": 27385.7,
      "step": 16880,
      "student_distance_mean": 24.378634643554687,
      "teacher_distance_mean": 0.8574559986591339,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.003265469241887331,
      "weighted_relation_loss": 0.003265469241887331
    },
    {
      "action_query_batch_std": 0.36489345133304596,
      "action_query_norm": 42.02582931518555,
      "action_relation_loss": 0.0032836947357282043,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.799812909260991,
      "flow_matching_loss": 0.02660145740956068,
      "grad_norm": 0.10415925830602646,
      "h_stock_valid_norm": 98.27223510742188,
      "invalid_tail_samples": 22.9,
      "learning_rate": 4.373880646556098e-05,
      "loss": 0.0299,
      "positive_teacher_pairs": 27061.1,
      "query_h_norm_ratio": 0.42765040397644044,
      "relation_loss": 0.0032836947357282043,
      "selected_pairs": 27061.1,
      "step": 16890,
      "student_distance_mean": 24.180476570129393,
      "teacher_distance_mean": 0.8491452217102051,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0032836947357282043,
      "weighted_relation_loss": 0.0032836947357282043
    },
    {
      "action_query_batch_std": 0.37159368097782136,
      "action_query_norm": 42.05988388061523,
      "action_relation_loss": 0.0033980362117290498,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.809167446211413,
      "flow_matching_loss": 0.026958484575152396,
      "grad_norm": 0.1299481838941574,
      "h_stock_valid_norm": 98.55567245483398,
      "invalid_tail_samples": 24.5,
      "learning_rate": 4.3684128462335533e-05,
      "loss": 0.0304,
      "positive_teacher_pairs": 26692.7,
      "query_h_norm_ratio": 0.42676606476306916,
      "relation_loss": 0.0033980362117290498,
      "selected_pairs": 26692.7,
      "step": 16900,
      "student_distance_mean": 24.722865104675293,
      "teacher_distance_mean": 0.8651179194450378,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0033980362117290498,
      "weighted_relation_loss": 0.0033980362117290498
    },
    {
      "action_query_batch_std": 0.3719299852848053,
      "action_query_norm": 42.052412796020505,
      "action_relation_loss": 0.0031439410522580147,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.818521983161833,
      "flow_matching_loss": 0.025508719868957995,
      "grad_norm": 0.09921235591173172,
      "h_stock_valid_norm": 98.51444549560547,
      "invalid_tail_samples": 24.8,
      "learning_rate": 4.362945813348955e-05,
      "loss": 0.0287,
      "positive_teacher_pairs": 26625.6,
      "query_h_norm_ratio": 0.4268685936927795,
      "relation_loss": 0.0031439410522580147,
      "selected_pairs": 26625.6,
      "step": 16910,
      "student_distance_mean": 24.625323867797853,
      "teacher_distance_mean": 0.8649627506732941,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0031439410522580147,
      "weighted_relation_loss": 0.0031439410522580147
    },
    {
      "action_query_batch_std": 0.3654475033283234,
      "action_query_norm": 42.01293601989746,
      "action_relation_loss": 0.0032311371294781567,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.827876520112255,
      "flow_matching_loss": 0.026934385113418103,
      "grad_norm": 0.13076286017894745,
      "h_stock_valid_norm": 98.34842071533203,
      "invalid_tail_samples": 22.3,
      "learning_rate": 4.357479554545262e-05,
      "loss": 0.0302,
      "positive_teacher_pairs": 27200.8,
      "query_h_norm_ratio": 0.42718771696090696,
      "relation_loss": 0.0032311371294781567,
      "selected_pairs": 27200.8,
      "step": 16920,
      "student_distance_mean": 24.242015075683593,
      "teacher_distance_mean": 0.8537888944149017,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0032311371294781567,
      "weighted_relation_loss": 0.0032311371294781567
    },
    {
      "action_query_batch_std": 0.3612256973981857,
      "action_query_norm": 41.99066925048828,
      "action_relation_loss": 0.003287493484094739,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.837231057062676,
      "flow_matching_loss": 0.026683943159878255,
      "grad_norm": 0.11461582034826279,
      "h_stock_valid_norm": 98.0393051147461,
      "invalid_tail_samples": 23.6,
      "learning_rate": 4.352014076464499e-05,
      "loss": 0.03,
      "positive_teacher_pairs": 26894.3,
      "query_h_norm_ratio": 0.4283065438270569,
      "relation_loss": 0.003287493484094739,
      "selected_pairs": 26894.3,
      "step": 16930,
      "student_distance_mean": 23.972668647766113,
      "teacher_distance_mean": 0.8407330453395844,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.003287493484094739,
      "weighted_relation_loss": 0.003287493484094739
    },
    {
      "action_query_batch_std": 0.3748482584953308,
      "action_query_norm": 42.07253913879394,
      "action_relation_loss": 0.0033138214144855737,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.846585594013096,
      "flow_matching_loss": 0.026385098323225976,
      "grad_norm": 0.11004149913787842,
      "h_stock_valid_norm": 98.37392807006836,
      "invalid_tail_samples": 23.1,
      "learning_rate": 4.346549385747734e-05,
      "loss": 0.0297,
      "positive_teacher_pairs": 27008.5,
      "query_h_norm_ratio": 0.42768132984638213,
      "relation_loss": 0.0033138214144855737,
      "selected_pairs": 27008.5,
      "step": 16940,
      "student_distance_mean": 24.80527267456055,
      "teacher_distance_mean": 0.8637501716613769,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0033138214144855737,
      "weighted_relation_loss": 0.0033138214144855737
    },
    {
      "action_query_batch_std": 0.3690040737390518,
      "action_query_norm": 42.02049560546875,
      "action_relation_loss": 0.003070707828737795,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.855940130963518,
      "flow_matching_loss": 0.027202770672738553,
      "grad_norm": 0.10031701624393463,
      "h_stock_valid_norm": 98.50815505981446,
      "invalid_tail_samples": 23.0,
      "learning_rate": 4.34108548903508e-05,
      "loss": 0.0303,
      "positive_teacher_pairs": 27034.0,
      "query_h_norm_ratio": 0.42657290995121,
      "relation_loss": 0.003070707828737795,
      "selected_pairs": 27034.0,
      "step": 16950,
      "student_distance_mean": 24.344651794433595,
      "teacher_distance_mean": 0.854862904548645,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.003070707828737795,
      "weighted_relation_loss": 0.003070707828737795
    },
    {
      "action_query_batch_std": 0.36947105526924134,
      "action_query_norm": 42.019906234741214,
      "action_relation_loss": 0.003289383323863149,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.865294667913938,
      "flow_matching_loss": 0.027364262938499452,
      "grad_norm": 0.1768590658903122,
      "h_stock_valid_norm": 98.2958984375,
      "invalid_tail_samples": 24.5,
      "learning_rate": 4.335622392965689e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 26692.1,
      "query_h_norm_ratio": 0.42748761773109434,
      "relation_loss": 0.003289383323863149,
      "selected_pairs": 26692.1,
      "step": 16960,
      "student_distance_mean": 24.276632881164552,
      "teacher_distance_mean": 0.8532449066638946,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.003289383323863149,
      "weighted_relation_loss": 0.003289383323863149
    },
    {
      "action_query_batch_std": 0.36588907837867735,
      "action_query_norm": 42.025508117675784,
      "action_relation_loss": 0.003215408674441278,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.87464920486436,
      "flow_matching_loss": 0.02731127180159092,
      "grad_norm": 0.09887094050645828,
      "h_stock_valid_norm": 97.71240844726563,
      "invalid_tail_samples": 20.9,
      "learning_rate": 4.330160104177738e-05,
      "loss": 0.0305,
      "positive_teacher_pairs": 27527.9,
      "query_h_norm_ratio": 0.4300992161035538,
      "relation_loss": 0.003215408674441278,
      "selected_pairs": 27527.9,
      "step": 16970,
      "student_distance_mean": 24.28750228881836,
      "teacher_distance_mean": 0.8521814703941345,
      "valid_action_samples": 235.1,
      "weighted_auxiliary_loss": 0.003215408674441278,
      "weighted_relation_loss": 0.003215408674441278
    },
    {
      "action_query_batch_std": 0.3652422457933426,
      "action_query_norm": 41.99471855163574,
      "action_relation_loss": 0.0032463933108374476,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.88400374181478,
      "flow_matching_loss": 0.026385725662112235,
      "grad_norm": 0.10852163285017014,
      "h_stock_valid_norm": 97.20252075195313,
      "invalid_tail_samples": 24.0,
      "learning_rate": 4.32469862930842e-05,
      "loss": 0.0296,
      "positive_teacher_pairs": 26807.5,
      "query_h_norm_ratio": 0.4320370197296143,
      "relation_loss": 0.0032463933108374476,
      "selected_pairs": 26807.5,
      "step": 16980,
      "student_distance_mean": 24.218740463256836,
      "teacher_distance_mean": 0.8541272938251495,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0032463933108374476,
      "weighted_relation_loss": 0.0032463933108374476
    },
    {
      "action_query_batch_std": 0.363007190823555,
      "action_query_norm": 41.988909530639646,
      "action_relation_loss": 0.003340410045348108,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.8933582787652,
      "flow_matching_loss": 0.02813694179058075,
      "grad_norm": 0.13712909817695618,
      "h_stock_valid_norm": 97.4987663269043,
      "invalid_tail_samples": 22.2,
      "learning_rate": 4.319237974993946e-05,
      "loss": 0.0315,
      "positive_teacher_pairs": 27216.5,
      "query_h_norm_ratio": 0.43066360652446745,
      "relation_loss": 0.003340410045348108,
      "selected_pairs": 27216.5,
      "step": 16990,
      "student_distance_mean": 24.123449325561523,
      "teacher_distance_mean": 0.8507105946540833,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.003340410045348108,
      "weighted_relation_loss": 0.003340410045348108
    },
    {
      "action_query_batch_std": 0.35915294885635374,
      "action_query_norm": 41.96488647460937,
      "action_relation_loss": 0.0032634383533149958,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.902712815715622,
      "flow_matching_loss": 0.02802182547748089,
      "grad_norm": 0.12441303580999374,
      "h_stock_valid_norm": 97.8424461364746,
      "invalid_tail_samples": 24.8,
      "learning_rate": 4.3137781478695236e-05,
      "loss": 0.0313,
      "positive_teacher_pairs": 26617.3,
      "query_h_norm_ratio": 0.4289051562547684,
      "relation_loss": 0.0032634383533149958,
      "selected_pairs": 26617.3,
      "step": 17000,
      "student_distance_mean": 23.836292457580566,
      "teacher_distance_mean": 0.8458852469921112,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0032634383533149958,
      "weighted_relation_loss": 0.0032634383533149958
    },
    {
      "action_query_batch_std": 0.36505638659000395,
      "action_query_norm": 41.980230712890624,
      "action_relation_loss": 0.003184405341744423,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.912067352666043,
      "flow_matching_loss": 0.028089638240635394,
      "grad_norm": 0.13029205799102783,
      "h_stock_valid_norm": 98.0571907043457,
      "invalid_tail_samples": 23.2,
      "learning_rate": 4.308319154569358e-05,
      "loss": 0.0313,
      "positive_teacher_pairs": 26991.5,
      "query_h_norm_ratio": 0.42812507748603823,
      "relation_loss": 0.003184405341744423,
      "selected_pairs": 26991.5,
      "step": 17010,
      "student_distance_mean": 24.15136013031006,
      "teacher_distance_mean": 0.8530342161655426,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.003184405341744423,
      "weighted_relation_loss": 0.003184405341744423
    },
    {
      "action_query_batch_std": 0.36382563710212706,
      "action_query_norm": 41.9873348236084,
      "action_relation_loss": 0.0031879972666502,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.921421889616465,
      "flow_matching_loss": 0.02585524059832096,
      "grad_norm": 0.08725665509700775,
      "h_stock_valid_norm": 97.99227447509766,
      "invalid_tail_samples": 20.3,
      "learning_rate": 4.302861001726641e-05,
      "loss": 0.029,
      "positive_teacher_pairs": 27675.7,
      "query_h_norm_ratio": 0.4284781962633133,
      "relation_loss": 0.0031879972666502,
      "selected_pairs": 27675.7,
      "step": 17020,
      "student_distance_mean": 24.08716869354248,
      "teacher_distance_mean": 0.8510075032711029,
      "valid_action_samples": 235.7,
      "weighted_auxiliary_loss": 0.0031879972666502,
      "weighted_relation_loss": 0.0031879972666502
    },
    {
      "action_query_batch_std": 0.36498628854751586,
      "action_query_norm": 41.99607124328613,
      "action_relation_loss": 0.0033496481832116843,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.930776426566885,
      "flow_matching_loss": 0.02700221985578537,
      "grad_norm": 0.0832022950053215,
      "h_stock_valid_norm": 97.77594680786133,
      "invalid_tail_samples": 23.1,
      "learning_rate": 4.2974036959735415e-05,
      "loss": 0.0304,
      "positive_teacher_pairs": 27014.2,
      "query_h_norm_ratio": 0.4295163661241531,
      "relation_loss": 0.0033496481832116843,
      "selected_pairs": 27014.2,
      "step": 17030,
      "student_distance_mean": 24.233420181274415,
      "teacher_distance_mean": 0.8521780848503113,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0033496481832116843,
      "weighted_relation_loss": 0.0033496481832116843
    },
    {
      "action_query_batch_std": 0.368379744887352,
      "action_query_norm": 41.993289184570315,
      "action_relation_loss": 0.0032255973666906355,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.940130963517305,
      "flow_matching_loss": 0.02750904466956854,
      "grad_norm": 0.11728841066360474,
      "h_stock_valid_norm": 97.7331558227539,
      "invalid_tail_samples": 26.3,
      "learning_rate": 4.2919472439412034e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 26272.7,
      "query_h_norm_ratio": 0.42967867851257324,
      "relation_loss": 0.0032255973666906355,
      "selected_pairs": 26272.7,
      "step": 17040,
      "student_distance_mean": 24.481031608581542,
      "teacher_distance_mean": 0.8615656077861786,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.0032255973666906355,
      "weighted_relation_loss": 0.0032255973666906355
    },
    {
      "action_query_batch_std": 0.3642088443040848,
      "action_query_norm": 41.95189590454102,
      "action_relation_loss": 0.0033565038116648792,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.949485500467727,
      "flow_matching_loss": 0.02565418481826782,
      "grad_norm": 0.10670113563537598,
      "h_stock_valid_norm": 97.76915588378907,
      "invalid_tail_samples": 25.5,
      "learning_rate": 4.286491652259729e-05,
      "loss": 0.029,
      "positive_teacher_pairs": 26462.8,
      "query_h_norm_ratio": 0.42909630239009855,
      "relation_loss": 0.0033565038116648792,
      "selected_pairs": 26462.8,
      "step": 17050,
      "student_distance_mean": 24.13584213256836,
      "teacher_distance_mean": 0.849552047252655,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0033565038116648792,
      "weighted_relation_loss": 0.0033565038116648792
    },
    {
      "action_query_batch_std": 0.3671547442674637,
      "action_query_norm": 41.98171691894531,
      "action_relation_loss": 0.0032045414904132486,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.958840037418147,
      "flow_matching_loss": 0.029432503134012224,
      "grad_norm": 0.14376504719257355,
      "h_stock_valid_norm": 97.52385177612305,
      "invalid_tail_samples": 22.6,
      "learning_rate": 4.281036927558178e-05,
      "loss": 0.0326,
      "positive_teacher_pairs": 27129.1,
      "query_h_norm_ratio": 0.430480095744133,
      "relation_loss": 0.0032045414904132486,
      "selected_pairs": 27129.1,
      "step": 17060,
      "student_distance_mean": 24.342919921875,
      "teacher_distance_mean": 0.8592657625675202,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.0032045414904132486,
      "weighted_relation_loss": 0.0032045414904132486
    },
    {
      "action_query_batch_std": 0.3658141613006592,
      "action_query_norm": 41.9705322265625,
      "action_relation_loss": 0.0033603850984945892,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.96819457436857,
      "flow_matching_loss": 0.027173976600170135,
      "grad_norm": 0.14446976780891418,
      "h_stock_valid_norm": 97.39353713989257,
      "invalid_tail_samples": 25.6,
      "learning_rate": 4.275583076464552e-05,
      "loss": 0.0305,
      "positive_teacher_pairs": 26433.5,
      "query_h_norm_ratio": 0.43093882501125336,
      "relation_loss": 0.0033603850984945892,
      "selected_pairs": 26433.5,
      "step": 17070,
      "student_distance_mean": 24.24621868133545,
      "teacher_distance_mean": 0.8561467111110688,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0033603850984945892,
      "weighted_relation_loss": 0.0033603850984945892
    },
    {
      "action_query_batch_std": 0.3711793959140778,
      "action_query_norm": 41.99500579833985,
      "action_relation_loss": 0.003043233579955995,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.97754911131899,
      "flow_matching_loss": 0.026872001215815546,
      "grad_norm": 0.09584943950176239,
      "h_stock_valid_norm": 97.42610702514648,
      "invalid_tail_samples": 23.4,
      "learning_rate": 4.270130105605794e-05,
      "loss": 0.0299,
      "positive_teacher_pairs": 26940.6,
      "query_h_norm_ratio": 0.4310498148202896,
      "relation_loss": 0.003043233579955995,
      "selected_pairs": 26940.6,
      "step": 17080,
      "student_distance_mean": 24.655435180664064,
      "teacher_distance_mean": 0.8684553384780884,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.003043233579955995,
      "weighted_relation_loss": 0.003043233579955995
    },
    {
      "action_query_batch_std": 0.3665947675704956,
      "action_query_norm": 41.970154190063475,
      "action_relation_loss": 0.00334391922224313,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.986903648269411,
      "flow_matching_loss": 0.02776602916419506,
      "grad_norm": 0.11402900516986847,
      "h_stock_valid_norm": 98.04752883911132,
      "invalid_tail_samples": 23.7,
      "learning_rate": 4.264678021607782e-05,
      "loss": 0.0311,
      "positive_teacher_pairs": 26878.4,
      "query_h_norm_ratio": 0.4280643671751022,
      "relation_loss": 0.00334391922224313,
      "selected_pairs": 26878.4,
      "step": 17090,
      "student_distance_mean": 24.219041442871095,
      "teacher_distance_mean": 0.8525933623313904,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.00334391922224313,
      "weighted_relation_loss": 0.00334391922224313
    },
    {
      "action_query_batch_std": 0.3693688690662384,
      "action_query_norm": 41.99176292419433,
      "action_relation_loss": 0.0032595374388620256,
      "auxiliary_loss_weight": 1.0,
      "epoch": 15.996258185219832,
      "flow_matching_loss": 0.027588429488241674,
      "grad_norm": 0.12058251351118088,
      "h_stock_valid_norm": 98.06983489990235,
      "invalid_tail_samples": 24.1,
      "learning_rate": 4.25922683109531e-05,
      "loss": 0.0308,
      "positive_teacher_pairs": 26776.8,
      "query_h_norm_ratio": 0.42818998396396635,
      "relation_loss": 0.0032595374388620256,
      "selected_pairs": 26776.8,
      "step": 17100,
      "student_distance_mean": 24.474491691589357,
      "teacher_distance_mean": 0.8633120656013489,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0032595374388620256,
      "weighted_relation_loss": 0.0032595374388620256
    },
    {
      "action_query_batch_std": 0.3680676281452179,
      "action_query_norm": 41.99913063049316,
      "action_relation_loss": 0.0031197390984743834,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.00561272217025,
      "flow_matching_loss": 0.02601372543722391,
      "grad_norm": 0.1116943433880806,
      "h_stock_valid_norm": 97.0951171875,
      "invalid_tail_samples": 21.7,
      "learning_rate": 4.25377654069209e-05,
      "loss": 0.0291,
      "positive_teacher_pairs": 24332.7,
      "query_h_norm_ratio": 0.4325678139925003,
      "relation_loss": 0.0031197390984743834,
      "selected_pairs": 24332.7,
      "step": 17110,
      "student_distance_mean": 24.295380973815917,
      "teacher_distance_mean": 0.8570813536643982,
      "valid_action_samples": 214.4,
      "weighted_auxiliary_loss": 0.0031197390984743834,
      "weighted_relation_loss": 0.0031197390984743834
    },
    {
      "action_query_batch_std": 0.36382445096969607,
      "action_query_norm": 41.93940773010254,
      "action_relation_loss": 0.0032245668582618235,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.014967259120674,
      "flow_matching_loss": 0.02931549046188593,
      "grad_norm": 0.1223897635936737,
      "h_stock_valid_norm": 97.3090690612793,
      "invalid_tail_samples": 23.8,
      "learning_rate": 4.248327157020737e-05,
      "loss": 0.0325,
      "positive_teacher_pairs": 26858.2,
      "query_h_norm_ratio": 0.43099349141120913,
      "relation_loss": 0.0032245668582618235,
      "selected_pairs": 26858.2,
      "step": 17120,
      "student_distance_mean": 24.05671157836914,
      "teacher_distance_mean": 0.8511475265026093,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0032245668582618235,
      "weighted_relation_loss": 0.0032245668582618235
    },
    {
      "action_query_batch_std": 0.3681732714176178,
      "action_query_norm": 41.9673526763916,
      "action_relation_loss": 0.003221111395396292,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.024321796071096,
      "flow_matching_loss": 0.02558883298188448,
      "grad_norm": 0.11035604774951935,
      "h_stock_valid_norm": 97.96208953857422,
      "invalid_tail_samples": 21.2,
      "learning_rate": 4.242878686702763e-05,
      "loss": 0.0288,
      "positive_teacher_pairs": 27454.4,
      "query_h_norm_ratio": 0.4284081608057022,
      "relation_loss": 0.003221111395396292,
      "selected_pairs": 27454.4,
      "step": 17130,
      "student_distance_mean": 24.359886741638185,
      "teacher_distance_mean": 0.8564788401126862,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.003221111395396292,
      "weighted_relation_loss": 0.003221111395396292
    },
    {
      "action_query_batch_std": 0.36289921402931213,
      "action_query_norm": 41.9553939819336,
      "action_relation_loss": 0.003176649077795446,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.033676333021514,
      "flow_matching_loss": 0.024918177165091038,
      "grad_norm": 0.08676442503929138,
      "h_stock_valid_norm": 98.10144271850587,
      "invalid_tail_samples": 24.7,
      "learning_rate": 4.23743113635858e-05,
      "loss": 0.0281,
      "positive_teacher_pairs": 26645.1,
      "query_h_norm_ratio": 0.42767471373081206,
      "relation_loss": 0.003176649077795446,
      "selected_pairs": 26645.1,
      "step": 17140,
      "student_distance_mean": 24.121196746826172,
      "teacher_distance_mean": 0.8488469600677491,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.003176649077795446,
      "weighted_relation_loss": 0.003176649077795446
    },
    {
      "action_query_batch_std": 0.3696667939424515,
      "action_query_norm": 41.95759201049805,
      "action_relation_loss": 0.003144976496696472,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.043030869971936,
      "flow_matching_loss": 0.02623444255441427,
      "grad_norm": 0.09468937665224075,
      "h_stock_valid_norm": 97.99939575195313,
      "invalid_tail_samples": 24.1,
      "learning_rate": 4.2319845126074705e-05,
      "loss": 0.0294,
      "positive_teacher_pairs": 26788.6,
      "query_h_norm_ratio": 0.4281433284282684,
      "relation_loss": 0.003144976496696472,
      "selected_pairs": 26788.6,
      "step": 17150,
      "student_distance_mean": 24.50972900390625,
      "teacher_distance_mean": 0.8625443637371063,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.003144976496696472,
      "weighted_relation_loss": 0.003144976496696472
    },
    {
      "action_query_batch_std": 0.37219208776950835,
      "action_query_norm": 41.987556838989256,
      "action_relation_loss": 0.003011045162566006,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.052385406922358,
      "flow_matching_loss": 0.02438405193388462,
      "grad_norm": 0.09830217063426971,
      "h_stock_valid_norm": 97.90581512451172,
      "invalid_tail_samples": 23.8,
      "learning_rate": 4.226538822067598e-05,
      "loss": 0.0274,
      "positive_teacher_pairs": 26862.8,
      "query_h_norm_ratio": 0.42885901927948,
      "relation_loss": 0.003011045162566006,
      "selected_pairs": 26862.8,
      "step": 17160,
      "student_distance_mean": 24.62722225189209,
      "teacher_distance_mean": 0.8643251538276673,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.003011045162566006,
      "weighted_relation_loss": 0.003011045162566006
    },
    {
      "action_query_batch_std": 0.3676888674497604,
      "action_query_norm": 41.95850791931152,
      "action_relation_loss": 0.003165285615250468,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.061739943872777,
      "flow_matching_loss": 0.026474994048476218,
      "grad_norm": 0.10192230343818665,
      "h_stock_valid_norm": 98.09973754882813,
      "invalid_tail_samples": 23.7,
      "learning_rate": 4.2210940713559896e-05,
      "loss": 0.0296,
      "positive_teacher_pairs": 26880.3,
      "query_h_norm_ratio": 0.42771499156951903,
      "relation_loss": 0.003165285615250468,
      "selected_pairs": 26880.3,
      "step": 17170,
      "student_distance_mean": 24.42229232788086,
      "teacher_distance_mean": 0.8598803102970123,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.003165285615250468,
      "weighted_relation_loss": 0.003165285615250468
    },
    {
      "action_query_batch_std": 0.37074785828590395,
      "action_query_norm": 41.96892890930176,
      "action_relation_loss": 0.0030689466511830688,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.0710944808232,
      "flow_matching_loss": 0.026152774319052698,
      "grad_norm": 0.13187475502490997,
      "h_stock_valid_norm": 97.97794723510742,
      "invalid_tail_samples": 25.8,
      "learning_rate": 4.21565026708853e-05,
      "loss": 0.0292,
      "positive_teacher_pairs": 26399.7,
      "query_h_norm_ratio": 0.4283510833978653,
      "relation_loss": 0.0030689466511830688,
      "selected_pairs": 26399.7,
      "step": 17180,
      "student_distance_mean": 24.593318939208984,
      "teacher_distance_mean": 0.865208488702774,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0030689466511830688,
      "weighted_relation_loss": 0.0030689466511830688
    },
    {
      "action_query_batch_std": 0.36858310401439665,
      "action_query_norm": 41.970868301391604,
      "action_relation_loss": 0.0030324880965054033,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.08044901777362,
      "flow_matching_loss": 0.024083041958510874,
      "grad_norm": 0.09662675857543945,
      "h_stock_valid_norm": 97.9842643737793,
      "invalid_tail_samples": 24.0,
      "learning_rate": 4.2102074158799535e-05,
      "loss": 0.0271,
      "positive_teacher_pairs": 26803.0,
      "query_h_norm_ratio": 0.42834922671318054,
      "relation_loss": 0.0030324880965054033,
      "selected_pairs": 26803.0,
      "step": 17190,
      "student_distance_mean": 24.47895622253418,
      "teacher_distance_mean": 0.8599130868911743,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0030324880965054033,
      "weighted_relation_loss": 0.0030324880965054033
    },
    {
      "action_query_batch_std": 0.3650091469287872,
      "action_query_norm": 41.93795471191406,
      "action_relation_loss": 0.0031341500114649536,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.089803554724043,
      "flow_matching_loss": 0.025833838246762754,
      "grad_norm": 0.10079768300056458,
      "h_stock_valid_norm": 98.07135772705078,
      "invalid_tail_samples": 23.8,
      "learning_rate": 4.204765524343841e-05,
      "loss": 0.029,
      "positive_teacher_pairs": 26847.7,
      "query_h_norm_ratio": 0.4276291251182556,
      "relation_loss": 0.0031341500114649536,
      "selected_pairs": 26847.7,
      "step": 17200,
      "student_distance_mean": 24.250105476379396,
      "teacher_distance_mean": 0.8514027655124664,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0031341500114649536,
      "weighted_relation_loss": 0.0031341500114649536
    },
    {
      "action_query_batch_std": 0.36565932631492615,
      "action_query_norm": 41.92339096069336,
      "action_relation_loss": 0.003163681854493916,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.09915809167446,
      "flow_matching_loss": 0.025703944079577924,
      "grad_norm": 0.12152322381734848,
      "h_stock_valid_norm": 97.9575210571289,
      "invalid_tail_samples": 26.3,
      "learning_rate": 4.1993245990926036e-05,
      "loss": 0.0289,
      "positive_teacher_pairs": 26280.7,
      "query_h_norm_ratio": 0.4279785960912704,
      "relation_loss": 0.003163681854493916,
      "selected_pairs": 26280.7,
      "step": 17210,
      "student_distance_mean": 24.22556266784668,
      "teacher_distance_mean": 0.8561035752296448,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.003163681854493916,
      "weighted_relation_loss": 0.003163681854493916
    },
    {
      "action_query_batch_std": 0.36401407420635223,
      "action_query_norm": 41.928279113769534,
      "action_relation_loss": 0.0030619077384471892,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.108512628624883,
      "flow_matching_loss": 0.026489145308732986,
      "grad_norm": 0.10032356530427933,
      "h_stock_valid_norm": 98.02488174438477,
      "invalid_tail_samples": 25.7,
      "learning_rate": 4.193884646737474e-05,
      "loss": 0.0296,
      "positive_teacher_pairs": 26422.3,
      "query_h_norm_ratio": 0.427732390165329,
      "relation_loss": 0.0030619077384471892,
      "selected_pairs": 26422.3,
      "step": 17220,
      "student_distance_mean": 24.11034564971924,
      "teacher_distance_mean": 0.8504813849925995,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0030619077384471892,
      "weighted_relation_loss": 0.0030619077384471892
    },
    {
      "action_query_batch_std": 0.36731659173965453,
      "action_query_norm": 41.927407836914064,
      "action_relation_loss": 0.0031516948947682977,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.117867165575305,
      "flow_matching_loss": 0.026381073892116545,
      "grad_norm": 0.09615977853536606,
      "h_stock_valid_norm": 97.79233169555664,
      "invalid_tail_samples": 26.5,
      "learning_rate": 4.1884456738885124e-05,
      "loss": 0.0295,
      "positive_teacher_pairs": 26233.6,
      "query_h_norm_ratio": 0.42874405682086947,
      "relation_loss": 0.0031516948947682977,
      "selected_pairs": 26233.6,
      "step": 17230,
      "student_distance_mean": 24.295383644104003,
      "teacher_distance_mean": 0.8577429115772247,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.0031516948947682977,
      "weighted_relation_loss": 0.0031516948947682977
    },
    {
      "action_query_batch_std": 0.3698242396116257,
      "action_query_norm": 41.951026916503906,
      "action_relation_loss": 0.003110753791406751,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.127221702525723,
      "flow_matching_loss": 0.02629667893052101,
      "grad_norm": 0.09517259895801544,
      "h_stock_valid_norm": 97.78477630615234,
      "invalid_tail_samples": 22.8,
      "learning_rate": 4.1830076871545795e-05,
      "loss": 0.0294,
      "positive_teacher_pairs": 27080.3,
      "query_h_norm_ratio": 0.4290157169103622,
      "relation_loss": 0.003110753791406751,
      "selected_pairs": 27080.3,
      "step": 17240,
      "student_distance_mean": 24.55178565979004,
      "teacher_distance_mean": 0.864785087108612,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.003110753791406751,
      "weighted_relation_loss": 0.003110753791406751
    },
    {
      "action_query_batch_std": 0.3674052327871323,
      "action_query_norm": 41.95559120178223,
      "action_relation_loss": 0.003111426532268524,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.136576239476145,
      "flow_matching_loss": 0.027563577331602574,
      "grad_norm": 0.24971038103103638,
      "h_stock_valid_norm": 97.51268157958984,
      "invalid_tail_samples": 23.1,
      "learning_rate": 4.177570693143346e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 27015.2,
      "query_h_norm_ratio": 0.43026070296764374,
      "relation_loss": 0.003111426532268524,
      "selected_pairs": 27015.2,
      "step": 17250,
      "student_distance_mean": 24.390302085876463,
      "teacher_distance_mean": 0.8588145077228546,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.003111426532268524,
      "weighted_relation_loss": 0.003111426532268524
    },
    {
      "action_query_batch_std": 0.36422196626663206,
      "action_query_norm": 41.91635360717773,
      "action_relation_loss": 0.003273768210783601,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.145930776426567,
      "flow_matching_loss": 0.027742170356214048,
      "grad_norm": 0.12317784130573273,
      "h_stock_valid_norm": 97.67315216064453,
      "invalid_tail_samples": 25.1,
      "learning_rate": 4.172134698461271e-05,
      "loss": 0.031,
      "positive_teacher_pairs": 26547.1,
      "query_h_norm_ratio": 0.4291512191295624,
      "relation_loss": 0.003273768210783601,
      "selected_pairs": 26547.1,
      "step": 17260,
      "student_distance_mean": 24.10801753997803,
      "teacher_distance_mean": 0.8534925818443299,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.003273768210783601,
      "weighted_relation_loss": 0.003273768210783601
    },
    {
      "action_query_batch_std": 0.36581377387046815,
      "action_query_norm": 41.915754318237305,
      "action_relation_loss": 0.003138229111209512,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.15528531337699,
      "flow_matching_loss": 0.025104889646172523,
      "grad_norm": 0.11950474977493286,
      "h_stock_valid_norm": 97.8416877746582,
      "invalid_tail_samples": 21.8,
      "learning_rate": 4.166699709713599e-05,
      "loss": 0.0282,
      "positive_teacher_pairs": 27318.2,
      "query_h_norm_ratio": 0.42840576767921446,
      "relation_loss": 0.003138229111209512,
      "selected_pairs": 27318.2,
      "step": 17270,
      "student_distance_mean": 24.122270774841308,
      "teacher_distance_mean": 0.8549350798130035,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.003138229111209512,
      "weighted_relation_loss": 0.003138229111209512
    },
    {
      "action_query_batch_std": 0.36548295319080354,
      "action_query_norm": 41.92957229614258,
      "action_relation_loss": 0.0030848425114527343,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.164639850327408,
      "flow_matching_loss": 0.02777951043099165,
      "grad_norm": 0.10330569744110107,
      "h_stock_valid_norm": 98.21938629150391,
      "invalid_tail_samples": 23.3,
      "learning_rate": 4.1612657335043555e-05,
      "loss": 0.0309,
      "positive_teacher_pairs": 26962.5,
      "query_h_norm_ratio": 0.42690223157405854,
      "relation_loss": 0.0030848425114527343,
      "selected_pairs": 26962.5,
      "step": 17280,
      "student_distance_mean": 24.21493453979492,
      "teacher_distance_mean": 0.8547489404678345,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0030848425114527343,
      "weighted_relation_loss": 0.0030848425114527343
    },
    {
      "action_query_batch_std": 0.3663867890834808,
      "action_query_norm": 41.929853439331055,
      "action_relation_loss": 0.0031245991121977568,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.17399438727783,
      "flow_matching_loss": 0.02623597178608179,
      "grad_norm": 0.13603293895721436,
      "h_stock_valid_norm": 98.18525390625,
      "invalid_tail_samples": 23.4,
      "learning_rate": 4.155832776436331e-05,
      "loss": 0.0294,
      "positive_teacher_pairs": 26946.3,
      "query_h_norm_ratio": 0.4270515441894531,
      "relation_loss": 0.0031245991121977568,
      "selected_pairs": 26946.3,
      "step": 17290,
      "student_distance_mean": 24.29937629699707,
      "teacher_distance_mean": 0.8559456944465638,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0031245991121977568,
      "weighted_relation_loss": 0.0031245991121977568
    },
    {
      "action_query_batch_std": 0.36236355304718015,
      "action_query_norm": 41.92666091918945,
      "action_relation_loss": 0.003148294216953218,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.18334892422825,
      "flow_matching_loss": 0.02797895222902298,
      "grad_norm": 0.15500091016292572,
      "h_stock_valid_norm": 98.5371940612793,
      "invalid_tail_samples": 24.0,
      "learning_rate": 4.1504008451110845e-05,
      "loss": 0.0311,
      "positive_teacher_pairs": 26807.3,
      "query_h_norm_ratio": 0.4254956543445587,
      "relation_loss": 0.003148294216953218,
      "selected_pairs": 26807.3,
      "step": 17300,
      "student_distance_mean": 24.044929122924806,
      "teacher_distance_mean": 0.8482035756111145,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.003148294216953218,
      "weighted_relation_loss": 0.003148294216953218
    },
    {
      "action_query_batch_std": 0.36408066749572754,
      "action_query_norm": 41.916337203979495,
      "action_relation_loss": 0.0031121418811380862,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.19270346117867,
      "flow_matching_loss": 0.02788108102977276,
      "grad_norm": 0.19235242903232574,
      "h_stock_valid_norm": 98.56543426513672,
      "invalid_tail_samples": 24.6,
      "learning_rate": 4.1449699461289225e-05,
      "loss": 0.031,
      "positive_teacher_pairs": 26669.4,
      "query_h_norm_ratio": 0.4252674371004105,
      "relation_loss": 0.0031121418811380862,
      "selected_pairs": 26669.4,
      "step": 17310,
      "student_distance_mean": 24.12104606628418,
      "teacher_distance_mean": 0.8501830220222473,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0031121418811380862,
      "weighted_relation_loss": 0.0031121418811380862
    },
    {
      "action_query_batch_std": 0.36259677112102506,
      "action_query_norm": 41.9200252532959,
      "action_relation_loss": 0.003001907700672746,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.202057998129092,
      "flow_matching_loss": 0.02617100365459919,
      "grad_norm": 0.09921032935380936,
      "h_stock_valid_norm": 98.28469619750976,
      "invalid_tail_samples": 23.6,
      "learning_rate": 4.139540086088901e-05,
      "loss": 0.0292,
      "positive_teacher_pairs": 26893.1,
      "query_h_norm_ratio": 0.426517853140831,
      "relation_loss": 0.003001907700672746,
      "selected_pairs": 26893.1,
      "step": 17320,
      "student_distance_mean": 23.998980712890624,
      "teacher_distance_mean": 0.8471957564353942,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.003001907700672746,
      "weighted_relation_loss": 0.003001907700672746
    },
    {
      "action_query_batch_std": 0.36498725712299346,
      "action_query_norm": 41.9127082824707,
      "action_relation_loss": 0.003214668622240424,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.211412535079514,
      "flow_matching_loss": 0.02699556890875101,
      "grad_norm": 0.10164370387792587,
      "h_stock_valid_norm": 98.28042221069336,
      "invalid_tail_samples": 24.3,
      "learning_rate": 4.1341112715888106e-05,
      "loss": 0.0302,
      "positive_teacher_pairs": 26734.6,
      "query_h_norm_ratio": 0.42646178901195525,
      "relation_loss": 0.003214668622240424,
      "selected_pairs": 26734.6,
      "step": 17330,
      "student_distance_mean": 24.145501327514648,
      "teacher_distance_mean": 0.8549201250076294,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.003214668622240424,
      "weighted_relation_loss": 0.003214668622240424
    },
    {
      "action_query_batch_std": 0.36366071105003356,
      "action_query_norm": 41.89881629943848,
      "action_relation_loss": 0.003151540388353169,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.220767072029936,
      "flow_matching_loss": 0.024918292835354804,
      "grad_norm": 0.11515635997056961,
      "h_stock_valid_norm": 98.23547973632813,
      "invalid_tail_samples": 23.7,
      "learning_rate": 4.128683509225172e-05,
      "loss": 0.0281,
      "positive_teacher_pairs": 26870.3,
      "query_h_norm_ratio": 0.4265167325735092,
      "relation_loss": 0.003151540388353169,
      "selected_pairs": 26870.3,
      "step": 17340,
      "student_distance_mean": 24.053332138061524,
      "teacher_distance_mean": 0.8460904955863953,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.003151540388353169,
      "weighted_relation_loss": 0.003151540388353169
    },
    {
      "action_query_batch_std": 0.36812600791454314,
      "action_query_norm": 41.934590911865236,
      "action_relation_loss": 0.0031762792728841304,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.230121608980355,
      "flow_matching_loss": 0.027275016345083714,
      "grad_norm": 0.1195056363940239,
      "h_stock_valid_norm": 98.03758392333984,
      "invalid_tail_samples": 20.6,
      "learning_rate": 4.123256805593231e-05,
      "loss": 0.0305,
      "positive_teacher_pairs": 27598.8,
      "query_h_norm_ratio": 0.42774165272712705,
      "relation_loss": 0.0031762792728841304,
      "selected_pairs": 27598.8,
      "step": 17350,
      "student_distance_mean": 24.27874889373779,
      "teacher_distance_mean": 0.8548086225986481,
      "valid_action_samples": 235.4,
      "weighted_auxiliary_loss": 0.0031762792728841304,
      "weighted_relation_loss": 0.0031762792728841304
    },
    {
      "action_query_batch_std": 0.3604351818561554,
      "action_query_norm": 41.894187927246094,
      "action_relation_loss": 0.003239778196439147,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.239476145930777,
      "flow_matching_loss": 0.0262816259637475,
      "grad_norm": 0.09463831037282944,
      "h_stock_valid_norm": 97.57243270874024,
      "invalid_tail_samples": 25.0,
      "learning_rate": 4.117831167286943e-05,
      "loss": 0.0295,
      "positive_teacher_pairs": 26575.5,
      "query_h_norm_ratio": 0.42936747074127196,
      "relation_loss": 0.003239778196439147,
      "selected_pairs": 26575.5,
      "step": 17360,
      "student_distance_mean": 24.017046546936037,
      "teacher_distance_mean": 0.844481474161148,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.003239778196439147,
      "weighted_relation_loss": 0.003239778196439147
    },
    {
      "action_query_batch_std": 0.36690548360347747,
      "action_query_norm": 41.91179275512695,
      "action_relation_loss": 0.0031905085081234573,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.2488306828812,
      "flow_matching_loss": 0.027258407883346082,
      "grad_norm": 0.12592622637748718,
      "h_stock_valid_norm": 97.73043823242188,
      "invalid_tail_samples": 25.0,
      "learning_rate": 4.1124066008989685e-05,
      "loss": 0.0304,
      "positive_teacher_pairs": 26571.2,
      "query_h_norm_ratio": 0.4288547694683075,
      "relation_loss": 0.0031905085081234573,
      "selected_pairs": 26571.2,
      "step": 17370,
      "student_distance_mean": 24.310586357116698,
      "teacher_distance_mean": 0.8568577289581298,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0031905085081234573,
      "weighted_relation_loss": 0.0031905085081234573
    },
    {
      "action_query_batch_std": 0.3639717519283295,
      "action_query_norm": 41.88122978210449,
      "action_relation_loss": 0.003119536559097469,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.258185219831617,
      "flow_matching_loss": 0.027313876152038574,
      "grad_norm": 0.09718438982963562,
      "h_stock_valid_norm": 97.69840621948242,
      "invalid_tail_samples": 22.9,
      "learning_rate": 4.106983113020669e-05,
      "loss": 0.0304,
      "positive_teacher_pairs": 27059.1,
      "query_h_norm_ratio": 0.42868092060089114,
      "relation_loss": 0.003119536559097469,
      "selected_pairs": 27059.1,
      "step": 17380,
      "student_distance_mean": 24.107417488098143,
      "teacher_distance_mean": 0.8541369199752807,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.003119536559097469,
      "weighted_relation_loss": 0.003119536559097469
    },
    {
      "action_query_batch_std": 0.3661016821861267,
      "action_query_norm": 41.904965209960935,
      "action_relation_loss": 0.003124875295907259,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.26753975678204,
      "flow_matching_loss": 0.02591728214174509,
      "grad_norm": 0.10927657783031464,
      "h_stock_valid_norm": 98.04547576904297,
      "invalid_tail_samples": 21.5,
      "learning_rate": 4.101560710242094e-05,
      "loss": 0.029,
      "positive_teacher_pairs": 27393.7,
      "query_h_norm_ratio": 0.4274071305990219,
      "relation_loss": 0.003124875295907259,
      "selected_pairs": 27393.7,
      "step": 17390,
      "student_distance_mean": 24.23587131500244,
      "teacher_distance_mean": 0.8613944351673126,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.003124875295907259,
      "weighted_relation_loss": 0.003124875295907259
    },
    {
      "action_query_batch_std": 0.3667057007551193,
      "action_query_norm": 41.91171531677246,
      "action_relation_loss": 0.0030506602488458157,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.27689429373246,
      "flow_matching_loss": 0.02760252133011818,
      "grad_norm": 0.11170661449432373,
      "h_stock_valid_norm": 98.14351196289063,
      "invalid_tail_samples": 24.8,
      "learning_rate": 4.096139399151971e-05,
      "loss": 0.0307,
      "positive_teacher_pairs": 26617.4,
      "query_h_norm_ratio": 0.42704866230487826,
      "relation_loss": 0.0030506602488458157,
      "selected_pairs": 26617.4,
      "step": 17400,
      "student_distance_mean": 24.34137954711914,
      "teacher_distance_mean": 0.8568037748336792,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0030506602488458157,
      "weighted_relation_loss": 0.0030506602488458157
    },
    {
      "action_query_batch_std": 0.36545341908931733,
      "action_query_norm": 41.89191513061523,
      "action_relation_loss": 0.0031363992020487785,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.28624883068288,
      "flow_matching_loss": 0.026046392880380152,
      "grad_norm": 0.09864538162946701,
      "h_stock_valid_norm": 98.16024398803711,
      "invalid_tail_samples": 25.6,
      "learning_rate": 4.090719186337709e-05,
      "loss": 0.0292,
      "positive_teacher_pairs": 26439.1,
      "query_h_norm_ratio": 0.42677244544029236,
      "relation_loss": 0.0031363992020487785,
      "selected_pairs": 26439.1,
      "step": 17410,
      "student_distance_mean": 24.21858940124512,
      "teacher_distance_mean": 0.8542625844478607,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0031363992020487785,
      "weighted_relation_loss": 0.0031363992020487785
    },
    {
      "action_query_batch_std": 0.3653177797794342,
      "action_query_norm": 41.90021896362305,
      "action_relation_loss": 0.0031038772081956266,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.2956033676333,
      "flow_matching_loss": 0.025701785832643508,
      "grad_norm": 0.10088653117418289,
      "h_stock_valid_norm": 97.67443618774413,
      "invalid_tail_samples": 21.8,
      "learning_rate": 4.0853000783853746e-05,
      "loss": 0.0288,
      "positive_teacher_pairs": 27312.6,
      "query_h_norm_ratio": 0.42898178696632383,
      "relation_loss": 0.0031038772081956266,
      "selected_pairs": 27312.6,
      "step": 17420,
      "student_distance_mean": 24.27343578338623,
      "teacher_distance_mean": 0.8604182839393616,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.0031038772081956266,
      "weighted_relation_loss": 0.0031038772081956266
    },
    {
      "action_query_batch_std": 0.36781682074069977,
      "action_query_norm": 41.89937553405762,
      "action_relation_loss": 0.0032129669794812798,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.304957904583723,
      "flow_matching_loss": 0.025710492953658105,
      "grad_norm": 0.1459834724664688,
      "h_stock_valid_norm": 97.40051040649413,
      "invalid_tail_samples": 23.9,
      "learning_rate": 4.0798820818796956e-05,
      "loss": 0.0289,
      "positive_teacher_pairs": 26836.7,
      "query_h_norm_ratio": 0.43017966449260714,
      "relation_loss": 0.0032129669794812798,
      "selected_pairs": 26836.7,
      "step": 17430,
      "student_distance_mean": 24.26955451965332,
      "teacher_distance_mean": 0.8572149395942688,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0032129669794812798,
      "weighted_relation_loss": 0.0032129669794812798
    },
    {
      "action_query_batch_std": 0.36645298898220063,
      "action_query_norm": 41.89771842956543,
      "action_relation_loss": 0.0031340714311227204,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.314312441534145,
      "flow_matching_loss": 0.024458727240562438,
      "grad_norm": 0.11486497521400452,
      "h_stock_valid_norm": 97.26886215209962,
      "invalid_tail_samples": 23.7,
      "learning_rate": 4.074465203404048e-05,
      "loss": 0.0276,
      "positive_teacher_pairs": 26873.4,
      "query_h_norm_ratio": 0.43074322640895846,
      "relation_loss": 0.0031340714311227204,
      "selected_pairs": 26873.4,
      "step": 17440,
      "student_distance_mean": 24.354191398620607,
      "teacher_distance_mean": 0.8573591947555542,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0031340714311227204,
      "weighted_relation_loss": 0.0031340714311227204
    },
    {
      "action_query_batch_std": 0.3703224629163742,
      "action_query_norm": 41.91523399353027,
      "action_relation_loss": 0.003098669298924506,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.323666978484564,
      "flow_matching_loss": 0.02797345947474241,
      "grad_norm": 0.09786399453878403,
      "h_stock_valid_norm": 97.13909072875977,
      "invalid_tail_samples": 22.5,
      "learning_rate": 4.0690494495404495e-05,
      "loss": 0.0311,
      "positive_teacher_pairs": 27151.7,
      "query_h_norm_ratio": 0.43149946331977845,
      "relation_loss": 0.003098669298924506,
      "selected_pairs": 27151.7,
      "step": 17450,
      "student_distance_mean": 24.488203620910646,
      "teacher_distance_mean": 0.863102114200592,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.003098669298924506,
      "weighted_relation_loss": 0.003098669298924506
    },
    {
      "action_query_batch_std": 0.36398559510707856,
      "action_query_norm": 41.886479187011716,
      "action_relation_loss": 0.003096578037366271,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.333021515434986,
      "flow_matching_loss": 0.024550039134919644,
      "grad_norm": 0.10648374259471893,
      "h_stock_valid_norm": 97.2248321533203,
      "invalid_tail_samples": 22.2,
      "learning_rate": 4.0636348268695535e-05,
      "loss": 0.0276,
      "positive_teacher_pairs": 27220.3,
      "query_h_norm_ratio": 0.43082385659217837,
      "relation_loss": 0.003096578037366271,
      "selected_pairs": 27220.3,
      "step": 17460,
      "student_distance_mean": 24.090818405151367,
      "teacher_distance_mean": 0.8551287651062012,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.003096578037366271,
      "weighted_relation_loss": 0.003096578037366271
    },
    {
      "action_query_batch_std": 0.3658820688724518,
      "action_query_norm": 41.87659759521485,
      "action_relation_loss": 0.003001537034288049,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.342376052385408,
      "flow_matching_loss": 0.02622213363647461,
      "grad_norm": 0.10635647922754288,
      "h_stock_valid_norm": 97.59863052368163,
      "invalid_tail_samples": 23.0,
      "learning_rate": 4.058221341970637e-05,
      "loss": 0.0292,
      "positive_teacher_pairs": 27046.9,
      "query_h_norm_ratio": 0.4290732264518738,
      "relation_loss": 0.003001537034288049,
      "selected_pairs": 27046.9,
      "step": 17470,
      "student_distance_mean": 24.223387908935546,
      "teacher_distance_mean": 0.8535286247730255,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.003001537034288049,
      "weighted_relation_loss": 0.003001537034288049
    },
    {
      "action_query_batch_std": 0.3688449144363403,
      "action_query_norm": 41.904252243041995,
      "action_relation_loss": 0.0031770744128152726,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.351730589335826,
      "flow_matching_loss": 0.02742045819759369,
      "grad_norm": 0.08838436752557755,
      "h_stock_valid_norm": 97.61866302490235,
      "invalid_tail_samples": 25.8,
      "learning_rate": 4.0528090014215945e-05,
      "loss": 0.0306,
      "positive_teacher_pairs": 26392.7,
      "query_h_norm_ratio": 0.42926557660102843,
      "relation_loss": 0.0031770744128152726,
      "selected_pairs": 26392.7,
      "step": 17480,
      "student_distance_mean": 24.4197566986084,
      "teacher_distance_mean": 0.8562084376811981,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0031770744128152726,
      "weighted_relation_loss": 0.0031770744128152726
    },
    {
      "action_query_batch_std": 0.36636578738689424,
      "action_query_norm": 41.89793663024902,
      "action_relation_loss": 0.0031530884094536304,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.36108512628625,
      "flow_matching_loss": 0.027119150571525096,
      "grad_norm": 0.12736789882183075,
      "h_stock_valid_norm": 97.58908386230469,
      "invalid_tail_samples": 18.9,
      "learning_rate": 4.047397811798929e-05,
      "loss": 0.0303,
      "positive_teacher_pairs": 27994.1,
      "query_h_norm_ratio": 0.42933391630649564,
      "relation_loss": 0.0031530884094536304,
      "selected_pairs": 27994.1,
      "step": 17490,
      "student_distance_mean": 24.288808822631836,
      "teacher_distance_mean": 0.8554233133792877,
      "valid_action_samples": 237.1,
      "weighted_auxiliary_loss": 0.0031530884094536304,
      "weighted_relation_loss": 0.0031530884094536304
    },
    {
      "action_query_batch_std": 0.3638909548521042,
      "action_query_norm": 41.865017318725585,
      "action_relation_loss": 0.0030853079864755274,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.37043966323667,
      "flow_matching_loss": 0.026693208515644072,
      "grad_norm": 0.10664457082748413,
      "h_stock_valid_norm": 97.22102127075195,
      "invalid_tail_samples": 22.3,
      "learning_rate": 4.041987779677745e-05,
      "loss": 0.0298,
      "positive_teacher_pairs": 27208.8,
      "query_h_norm_ratio": 0.43061899542808535,
      "relation_loss": 0.0030853079864755274,
      "selected_pairs": 27208.8,
      "step": 17500,
      "student_distance_mean": 24.083186912536622,
      "teacher_distance_mean": 0.8513366878032684,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0030853079864755274,
      "weighted_relation_loss": 0.0030853079864755274
    },
    {
      "action_query_batch_std": 0.3640672147274017,
      "action_query_norm": 41.87414131164551,
      "action_relation_loss": 0.0030928786611184476,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.379794200187092,
      "flow_matching_loss": 0.026927389949560166,
      "grad_norm": 0.09299055486917496,
      "h_stock_valid_norm": 96.81748504638672,
      "invalid_tail_samples": 21.6,
      "learning_rate": 4.036578911631745e-05,
      "loss": 0.03,
      "positive_teacher_pairs": 27363.2,
      "query_h_norm_ratio": 0.43250848054885865,
      "relation_loss": 0.0030928786611184476,
      "selected_pairs": 27363.2,
      "step": 17510,
      "student_distance_mean": 24.116258811950683,
      "teacher_distance_mean": 0.8496860504150391,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.0030928786611184476,
      "weighted_relation_loss": 0.0030928786611184476
    },
    {
      "action_query_batch_std": 0.3640671014785767,
      "action_query_norm": 41.866065216064456,
      "action_relation_loss": 0.0031180620193481446,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.38914873713751,
      "flow_matching_loss": 0.026434120908379554,
      "grad_norm": 0.10792960971593857,
      "h_stock_valid_norm": 96.65558700561523,
      "invalid_tail_samples": 26.1,
      "learning_rate": 4.031171214233211e-05,
      "loss": 0.0296,
      "positive_teacher_pairs": 26322.3,
      "query_h_norm_ratio": 0.4331502377986908,
      "relation_loss": 0.0031180620193481446,
      "selected_pairs": 26322.3,
      "step": 17520,
      "student_distance_mean": 24.143199920654297,
      "teacher_distance_mean": 0.8555191874504089,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.0031180620193481446,
      "weighted_relation_loss": 0.0031180620193481446
    },
    {
      "action_query_batch_std": 0.36133474111557007,
      "action_query_norm": 41.849007034301756,
      "action_relation_loss": 0.003165944176726043,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.398503274087933,
      "flow_matching_loss": 0.024828110262751578,
      "grad_norm": 0.12028717249631882,
      "h_stock_valid_norm": 96.58046875,
      "invalid_tail_samples": 21.5,
      "learning_rate": 4.025764694053008e-05,
      "loss": 0.028,
      "positive_teacher_pairs": 27383.6,
      "query_h_norm_ratio": 0.4333102643489838,
      "relation_loss": 0.003165944176726043,
      "selected_pairs": 27383.6,
      "step": 17530,
      "student_distance_mean": 23.88134002685547,
      "teacher_distance_mean": 0.8437394976615906,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.003165944176726043,
      "weighted_relation_loss": 0.003165944176726043
    },
    {
      "action_query_batch_std": 0.36316114366054536,
      "action_query_norm": 41.85910415649414,
      "action_relation_loss": 0.003103844332508743,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.407857811038355,
      "flow_matching_loss": 0.02355650272220373,
      "grad_norm": 0.10396828502416611,
      "h_stock_valid_norm": 97.06051254272461,
      "invalid_tail_samples": 25.1,
      "learning_rate": 4.020359357660566e-05,
      "loss": 0.0267,
      "positive_teacher_pairs": 26552.7,
      "query_h_norm_ratio": 0.4312721937894821,
      "relation_loss": 0.003103844332508743,
      "selected_pairs": 26552.7,
      "step": 17540,
      "student_distance_mean": 24.071358108520506,
      "teacher_distance_mean": 0.8488347232341766,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.003103844332508743,
      "weighted_relation_loss": 0.003103844332508743
    },
    {
      "action_query_batch_std": 0.36319049298763273,
      "action_query_norm": 41.86145248413086,
      "action_relation_loss": 0.0032287958776578305,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.417212347988773,
      "flow_matching_loss": 0.026319318264722825,
      "grad_norm": 0.11978873610496521,
      "h_stock_valid_norm": 97.77754821777344,
      "invalid_tail_samples": 22.5,
      "learning_rate": 4.014955211623875e-05,
      "loss": 0.0295,
      "positive_teacher_pairs": 27151.7,
      "query_h_norm_ratio": 0.4281391084194183,
      "relation_loss": 0.0032287958776578305,
      "selected_pairs": 27151.7,
      "step": 17550,
      "student_distance_mean": 24.018187522888184,
      "teacher_distance_mean": 0.8491820096969604,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0032287958776578305,
      "weighted_relation_loss": 0.0032287958776578305
    },
    {
      "action_query_batch_std": 0.3677927523851395,
      "action_query_norm": 41.87786102294922,
      "action_relation_loss": 0.003048319416120648,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.426566884939195,
      "flow_matching_loss": 0.02465586308389902,
      "grad_norm": 0.1143946722149849,
      "h_stock_valid_norm": 98.19182586669922,
      "invalid_tail_samples": 23.3,
      "learning_rate": 4.0095522625094895e-05,
      "loss": 0.0277,
      "positive_teacher_pairs": 26967.4,
      "query_h_norm_ratio": 0.42649149894714355,
      "relation_loss": 0.003048319416120648,
      "selected_pairs": 26967.4,
      "step": 17560,
      "student_distance_mean": 24.47495918273926,
      "teacher_distance_mean": 0.8689408242702484,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.003048319416120648,
      "weighted_relation_loss": 0.003048319416120648
    },
    {
      "action_query_batch_std": 0.3663850069046021,
      "action_query_norm": 41.86897506713867,
      "action_relation_loss": 0.0030356869101524353,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.435921421889617,
      "flow_matching_loss": 0.025955912843346595,
      "grad_norm": 0.08685460686683655,
      "h_stock_valid_norm": 98.454345703125,
      "invalid_tail_samples": 24.3,
      "learning_rate": 4.004150516882497e-05,
      "loss": 0.029,
      "positive_teacher_pairs": 26732.0,
      "query_h_norm_ratio": 0.42526633739471437,
      "relation_loss": 0.0030356869101524353,
      "selected_pairs": 26732.0,
      "step": 17570,
      "student_distance_mean": 24.31289138793945,
      "teacher_distance_mean": 0.8614676773548127,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0030356869101524353,
      "weighted_relation_loss": 0.0030356869101524353
    },
    {
      "action_query_batch_std": 0.3632864743471146,
      "action_query_norm": 41.8540542602539,
      "action_relation_loss": 0.0030807594303041697,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.44527595884004,
      "flow_matching_loss": 0.02643325738608837,
      "grad_norm": 0.11590554565191269,
      "h_stock_valid_norm": 98.53430252075195,
      "invalid_tail_samples": 23.5,
      "learning_rate": 3.99874998130653e-05,
      "loss": 0.0295,
      "positive_teacher_pairs": 26926.2,
      "query_h_norm_ratio": 0.4247687757015228,
      "relation_loss": 0.0030807594303041697,
      "selected_pairs": 26926.2,
      "step": 17580,
      "student_distance_mean": 24.14454116821289,
      "teacher_distance_mean": 0.8564611375331879,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0030807594303041697,
      "weighted_relation_loss": 0.0030807594303041697
    },
    {
      "action_query_batch_std": 0.3641200244426727,
      "action_query_norm": 41.839748001098634,
      "action_relation_loss": 0.003017020458355546,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.454630495790457,
      "flow_matching_loss": 0.026435060240328313,
      "grad_norm": 0.1897086352109909,
      "h_stock_valid_norm": 98.1531364440918,
      "invalid_tail_samples": 22.9,
      "learning_rate": 3.9933506623437455e-05,
      "loss": 0.0295,
      "positive_teacher_pairs": 27055.6,
      "query_h_norm_ratio": 0.42627396881580354,
      "relation_loss": 0.003017020458355546,
      "selected_pairs": 27055.6,
      "step": 17590,
      "student_distance_mean": 24.006973266601562,
      "teacher_distance_mean": 0.8524772465229035,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.003017020458355546,
      "weighted_relation_loss": 0.003017020458355546
    },
    {
      "action_query_batch_std": 0.3658294349908829,
      "action_query_norm": 41.843838119506835,
      "action_relation_loss": 0.0029381419299170375,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.46398503274088,
      "flow_matching_loss": 0.024802624806761743,
      "grad_norm": 0.10823189467191696,
      "h_stock_valid_norm": 98.2711784362793,
      "invalid_tail_samples": 27.4,
      "learning_rate": 3.9879525665548276e-05,
      "loss": 0.0277,
      "positive_teacher_pairs": 26026.3,
      "query_h_norm_ratio": 0.42580113410949705,
      "relation_loss": 0.0029381419299170375,
      "selected_pairs": 26026.3,
      "step": 17600,
      "student_distance_mean": 24.29018077850342,
      "teacher_distance_mean": 0.859050965309143,
      "valid_action_samples": 228.6,
      "weighted_auxiliary_loss": 0.0029381419299170375,
      "weighted_relation_loss": 0.0029381419299170375
    },
    {
      "action_query_batch_std": 0.36584576070308683,
      "action_query_norm": 41.85316696166992,
      "action_relation_loss": 0.0030172571074217557,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.4733395696913,
      "flow_matching_loss": 0.02582656629383564,
      "grad_norm": 0.10984818637371063,
      "h_stock_valid_norm": 97.7346092224121,
      "invalid_tail_samples": 22.5,
      "learning_rate": 3.9825557004989704e-05,
      "loss": 0.0288,
      "positive_teacher_pairs": 27162.2,
      "query_h_norm_ratio": 0.42823803424835205,
      "relation_loss": 0.0030172571074217557,
      "selected_pairs": 27162.2,
      "step": 17610,
      "student_distance_mean": 24.178360748291016,
      "teacher_distance_mean": 0.857507973909378,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0030172571074217557,
      "weighted_relation_loss": 0.0030172571074217557
    },
    {
      "action_query_batch_std": 0.3702441602945328,
      "action_query_norm": 41.89466438293457,
      "action_relation_loss": 0.002937189885415137,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.48269410664172,
      "flow_matching_loss": 0.02477260660380125,
      "grad_norm": 0.10130681097507477,
      "h_stock_valid_norm": 97.63937606811524,
      "invalid_tail_samples": 23.4,
      "learning_rate": 3.977160070733877e-05,
      "loss": 0.0277,
      "positive_teacher_pairs": 26940.3,
      "query_h_norm_ratio": 0.4290765464305878,
      "relation_loss": 0.002937189885415137,
      "selected_pairs": 26940.3,
      "step": 17620,
      "student_distance_mean": 24.536430740356444,
      "teacher_distance_mean": 0.8682787239551544,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002937189885415137,
      "weighted_relation_loss": 0.002937189885415137
    },
    {
      "action_query_batch_std": 0.3710533887147903,
      "action_query_norm": 41.881150817871095,
      "action_relation_loss": 0.0029426554683595896,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.492048643592142,
      "flow_matching_loss": 0.02763692829757929,
      "grad_norm": 0.11950229108333588,
      "h_stock_valid_norm": 97.81566772460937,
      "invalid_tail_samples": 21.1,
      "learning_rate": 3.971765683815746e-05,
      "loss": 0.0306,
      "positive_teacher_pairs": 27475.4,
      "query_h_norm_ratio": 0.4281662106513977,
      "relation_loss": 0.0029426554683595896,
      "selected_pairs": 27475.4,
      "step": 17630,
      "student_distance_mean": 24.666936111450195,
      "teacher_distance_mean": 0.8706569969654083,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.0029426554683595896,
      "weighted_relation_loss": 0.0029426554683595896
    },
    {
      "action_query_batch_std": 0.3622362524271011,
      "action_query_norm": 41.80278358459473,
      "action_relation_loss": 0.0031285519944503903,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.501403180542564,
      "flow_matching_loss": 0.025441940873861313,
      "grad_norm": 0.14500947296619415,
      "h_stock_valid_norm": 98.13388595581054,
      "invalid_tail_samples": 24.0,
      "learning_rate": 3.966372546299265e-05,
      "loss": 0.0286,
      "positive_teacher_pairs": 26815.7,
      "query_h_norm_ratio": 0.4259778082370758,
      "relation_loss": 0.0031285519944503903,
      "selected_pairs": 26815.7,
      "step": 17640,
      "student_distance_mean": 23.863302612304686,
      "teacher_distance_mean": 0.8479695558547974,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0031285519944503903,
      "weighted_relation_loss": 0.0031285519944503903
    },
    {
      "action_query_batch_std": 0.3616950958967209,
      "action_query_norm": 41.81387786865234,
      "action_relation_loss": 0.0029911293648183348,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.510757717492986,
      "flow_matching_loss": 0.02486909069120884,
      "grad_norm": 0.1130906194448471,
      "h_stock_valid_norm": 98.18755416870117,
      "invalid_tail_samples": 24.6,
      "learning_rate": 3.960980664737604e-05,
      "loss": 0.0279,
      "positive_teacher_pairs": 26658.6,
      "query_h_norm_ratio": 0.42585994899272916,
      "relation_loss": 0.0029911293648183348,
      "selected_pairs": 26658.6,
      "step": 17650,
      "student_distance_mean": 23.92206268310547,
      "teacher_distance_mean": 0.8497118353843689,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0029911293648183348,
      "weighted_relation_loss": 0.0029911293648183348
    },
    {
      "action_query_batch_std": 0.36621021628379824,
      "action_query_norm": 41.85984840393066,
      "action_relation_loss": 0.003106690174899995,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.520112254443404,
      "flow_matching_loss": 0.02620091810822487,
      "grad_norm": 0.12521164119243622,
      "h_stock_valid_norm": 97.79216613769532,
      "invalid_tail_samples": 23.9,
      "learning_rate": 3.955590045682408e-05,
      "loss": 0.0293,
      "positive_teacher_pairs": 26832.6,
      "query_h_norm_ratio": 0.42805390655994413,
      "relation_loss": 0.003106690174899995,
      "selected_pairs": 26832.6,
      "step": 17660,
      "student_distance_mean": 24.26533031463623,
      "teacher_distance_mean": 0.8580476880073548,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.003106690174899995,
      "weighted_relation_loss": 0.003106690174899995
    },
    {
      "action_query_batch_std": 0.36761722564697263,
      "action_query_norm": 41.83264656066895,
      "action_relation_loss": 0.003162581939250231,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.529466791393826,
      "flow_matching_loss": 0.02522515580058098,
      "grad_norm": 0.08615896850824356,
      "h_stock_valid_norm": 98.0402618408203,
      "invalid_tail_samples": 22.3,
      "learning_rate": 3.950200695683788e-05,
      "loss": 0.0284,
      "positive_teacher_pairs": 27194.1,
      "query_h_norm_ratio": 0.4266931265592575,
      "relation_loss": 0.003162581939250231,
      "selected_pairs": 27194.1,
      "step": 17670,
      "student_distance_mean": 24.24310302734375,
      "teacher_distance_mean": 0.861275565624237,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.003162581939250231,
      "weighted_relation_loss": 0.003162581939250231
    },
    {
      "action_query_batch_std": 0.3620946168899536,
      "action_query_norm": 41.802920532226565,
      "action_relation_loss": 0.002968078153207898,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.53882132834425,
      "flow_matching_loss": 0.025122398510575294,
      "grad_norm": 0.1173444613814354,
      "h_stock_valid_norm": 98.06318435668945,
      "invalid_tail_samples": 22.7,
      "learning_rate": 3.944812621290314e-05,
      "loss": 0.0281,
      "positive_teacher_pairs": 27108.0,
      "query_h_norm_ratio": 0.4262888222932816,
      "relation_loss": 0.002968078153207898,
      "selected_pairs": 27108.0,
      "step": 17680,
      "student_distance_mean": 23.940709114074707,
      "teacher_distance_mean": 0.8533590197563171,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.002968078153207898,
      "weighted_relation_loss": 0.002968078153207898
    },
    {
      "action_query_batch_std": 0.3638834714889526,
      "action_query_norm": 41.82448310852051,
      "action_relation_loss": 0.002966220863163471,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.548175865294667,
      "flow_matching_loss": 0.025919174216687678,
      "grad_norm": 0.12657392024993896,
      "h_stock_valid_norm": 97.66450881958008,
      "invalid_tail_samples": 21.7,
      "learning_rate": 3.939425829049002e-05,
      "loss": 0.0289,
      "positive_teacher_pairs": 27337.5,
      "query_h_norm_ratio": 0.428253236413002,
      "relation_loss": 0.002966220863163471,
      "selected_pairs": 27337.5,
      "step": 17690,
      "student_distance_mean": 24.014466285705566,
      "teacher_distance_mean": 0.852530962228775,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.002966220863163471,
      "weighted_relation_loss": 0.002966220863163471
    },
    {
      "action_query_batch_std": 0.36720393002033236,
      "action_query_norm": 41.8383731842041,
      "action_relation_loss": 0.0031001008115708827,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.55753040224509,
      "flow_matching_loss": 0.025816874206066133,
      "grad_norm": 0.11919927597045898,
      "h_stock_valid_norm": 97.6539291381836,
      "invalid_tail_samples": 25.5,
      "learning_rate": 3.9340403255053135e-05,
      "loss": 0.0289,
      "positive_teacher_pairs": 26460.6,
      "query_h_norm_ratio": 0.4284366279840469,
      "relation_loss": 0.0031001008115708827,
      "selected_pairs": 26460.6,
      "step": 17700,
      "student_distance_mean": 24.27661266326904,
      "teacher_distance_mean": 0.8591692566871643,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0031001008115708827,
      "weighted_relation_loss": 0.0031001008115708827
    },
    {
      "action_query_batch_std": 0.36593453884124755,
      "action_query_norm": 41.837039947509766,
      "action_relation_loss": 0.0031587603036314247,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.56688493919551,
      "flow_matching_loss": 0.02427775114774704,
      "grad_norm": 0.10975391417741776,
      "h_stock_valid_norm": 98.17750396728516,
      "invalid_tail_samples": 23.3,
      "learning_rate": 3.9286561172031407e-05,
      "loss": 0.0274,
      "positive_teacher_pairs": 26971.5,
      "query_h_norm_ratio": 0.4261397272348404,
      "relation_loss": 0.0031587603036314247,
      "selected_pairs": 26971.5,
      "step": 17710,
      "student_distance_mean": 24.115812683105467,
      "teacher_distance_mean": 0.8551708638668061,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0031587603036314247,
      "weighted_relation_loss": 0.0031587603036314247
    },
    {
      "action_query_batch_std": 0.35685949623584745,
      "action_query_norm": 41.778255462646484,
      "action_relation_loss": 0.00306708428543061,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.57623947614593,
      "flow_matching_loss": 0.025196470506489276,
      "grad_norm": 0.10886174440383911,
      "h_stock_valid_norm": 98.61302947998047,
      "invalid_tail_samples": 24.2,
      "learning_rate": 3.923273210684809e-05,
      "loss": 0.0283,
      "positive_teacher_pairs": 26762.8,
      "query_h_norm_ratio": 0.4236633837223053,
      "relation_loss": 0.00306708428543061,
      "selected_pairs": 26762.8,
      "step": 17720,
      "student_distance_mean": 23.68249454498291,
      "teacher_distance_mean": 0.8409530520439148,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.00306708428543061,
      "weighted_relation_loss": 0.00306708428543061
    },
    {
      "action_query_batch_std": 0.36225478947162626,
      "action_query_norm": 41.80605239868164,
      "action_relation_loss": 0.0030964088859036566,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.58559401309635,
      "flow_matching_loss": 0.025052732415497304,
      "grad_norm": 0.10543258488178253,
      "h_stock_valid_norm": 98.63198471069336,
      "invalid_tail_samples": 25.1,
      "learning_rate": 3.9178916124910555e-05,
      "loss": 0.0281,
      "positive_teacher_pairs": 26547.8,
      "query_h_norm_ratio": 0.4238634467124939,
      "relation_loss": 0.0030964088859036566,
      "selected_pairs": 26547.8,
      "step": 17730,
      "student_distance_mean": 23.935150718688966,
      "teacher_distance_mean": 0.8509639143943787,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0030964088859036566,
      "weighted_relation_loss": 0.0030964088859036566
    },
    {
      "action_query_batch_std": 0.36283089518547057,
      "action_query_norm": 41.8284107208252,
      "action_relation_loss": 0.0029930011369287968,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.594948550046773,
      "flow_matching_loss": 0.025013707764446735,
      "grad_norm": 0.11695661395788193,
      "h_stock_valid_norm": 98.21988067626953,
      "invalid_tail_samples": 22.1,
      "learning_rate": 3.9125113291610274e-05,
      "loss": 0.028,
      "positive_teacher_pairs": 27244.7,
      "query_h_norm_ratio": 0.4258726239204407,
      "relation_loss": 0.0029930011369287968,
      "selected_pairs": 27244.7,
      "step": 17740,
      "student_distance_mean": 24.113753700256346,
      "teacher_distance_mean": 0.8592665255069732,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0029930011369287968,
      "weighted_relation_loss": 0.0029930011369287968
    },
    {
      "action_query_batch_std": 0.3627009689807892,
      "action_query_norm": 41.7903491973877,
      "action_relation_loss": 0.003151170676574111,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.604303086997195,
      "flow_matching_loss": 0.0264131810516119,
      "grad_norm": 0.15818330645561218,
      "h_stock_valid_norm": 98.24383850097657,
      "invalid_tail_samples": 24.3,
      "learning_rate": 3.907132367232278e-05,
      "loss": 0.0296,
      "positive_teacher_pairs": 26736.8,
      "query_h_norm_ratio": 0.42537638545036316,
      "relation_loss": 0.003151170676574111,
      "selected_pairs": 26736.8,
      "step": 17750,
      "student_distance_mean": 24.039092445373534,
      "teacher_distance_mean": 0.855676805973053,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.003151170676574111,
      "weighted_relation_loss": 0.003151170676574111
    },
    {
      "action_query_batch_std": 0.36382476389408114,
      "action_query_norm": 41.80844955444336,
      "action_relation_loss": 0.0031528843799605967,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.613657623947613,
      "flow_matching_loss": 0.025310979597270487,
      "grad_norm": 0.11046094447374344,
      "h_stock_valid_norm": 98.37871475219727,
      "invalid_tail_samples": 24.8,
      "learning_rate": 3.901754733240752e-05,
      "loss": 0.0285,
      "positive_teacher_pairs": 26618.2,
      "query_h_norm_ratio": 0.42497549653053285,
      "relation_loss": 0.0031528843799605967,
      "selected_pairs": 26618.2,
      "step": 17760,
      "student_distance_mean": 23.974919509887695,
      "teacher_distance_mean": 0.8492497026920318,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0031528843799605967,
      "weighted_relation_loss": 0.0031528843799605967
    },
    {
      "action_query_batch_std": 0.36533697247505187,
      "action_query_norm": 41.825436401367185,
      "action_relation_loss": 0.0030999201349914073,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.623012160898035,
      "flow_matching_loss": 0.023739235103130342,
      "grad_norm": 0.09414985030889511,
      "h_stock_valid_norm": 98.36456832885742,
      "invalid_tail_samples": 24.7,
      "learning_rate": 3.896378433720785e-05,
      "loss": 0.0268,
      "positive_teacher_pairs": 26640.1,
      "query_h_norm_ratio": 0.425211226940155,
      "relation_loss": 0.0030999201349914073,
      "selected_pairs": 26640.1,
      "step": 17770,
      "student_distance_mean": 24.150882148742674,
      "teacher_distance_mean": 0.8544222950935364,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0030999201349914073,
      "weighted_relation_loss": 0.0030999201349914073
    },
    {
      "action_query_batch_std": 0.3601236790418625,
      "action_query_norm": 41.801449584960935,
      "action_relation_loss": 0.0029158115154132246,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.632366697848457,
      "flow_matching_loss": 0.025079657696187495,
      "grad_norm": 0.11043331027030945,
      "h_stock_valid_norm": 98.42343521118164,
      "invalid_tail_samples": 22.8,
      "learning_rate": 3.891003475205086e-05,
      "loss": 0.028,
      "positive_teacher_pairs": 27084.0,
      "query_h_norm_ratio": 0.42471333742141726,
      "relation_loss": 0.0029158115154132246,
      "selected_pairs": 27084.0,
      "step": 17780,
      "student_distance_mean": 23.93996658325195,
      "teacher_distance_mean": 0.8513096988201141,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0029158115154132246,
      "weighted_relation_loss": 0.0029158115154132246
    },
    {
      "action_query_batch_std": 0.36359548568725586,
      "action_query_norm": 41.81550559997559,
      "action_relation_loss": 0.0031398461665958165,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.641721234798876,
      "flow_matching_loss": 0.026494507491588593,
      "grad_norm": 0.1340351104736328,
      "h_stock_valid_norm": 97.89896163940429,
      "invalid_tail_samples": 23.6,
      "learning_rate": 3.885629864224736e-05,
      "loss": 0.0296,
      "positive_teacher_pairs": 26903.4,
      "query_h_norm_ratio": 0.42713718116283417,
      "relation_loss": 0.0031398461665958165,
      "selected_pairs": 26903.4,
      "step": 17790,
      "student_distance_mean": 23.978601264953614,
      "teacher_distance_mean": 0.8515890419483185,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0031398461665958165,
      "weighted_relation_loss": 0.0031398461665958165
    },
    {
      "action_query_batch_std": 0.36903453469276426,
      "action_query_norm": 41.82637062072754,
      "action_relation_loss": 0.0031005822122097014,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.651075771749298,
      "flow_matching_loss": 0.02569964248687029,
      "grad_norm": 0.1285027116537094,
      "h_stock_valid_norm": 97.60898208618164,
      "invalid_tail_samples": 23.7,
      "learning_rate": 3.880257607309178e-05,
      "loss": 0.0288,
      "positive_teacher_pairs": 26876.4,
      "query_h_norm_ratio": 0.4285116225481033,
      "relation_loss": 0.0031005822122097014,
      "selected_pairs": 26876.4,
      "step": 17800,
      "student_distance_mean": 24.42831802368164,
      "teacher_distance_mean": 0.8674292385578155,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0031005822122097014,
      "weighted_relation_loss": 0.0031005822122097014
    },
    {
      "action_query_batch_std": 0.3635925084352493,
      "action_query_norm": 41.804751586914065,
      "action_relation_loss": 0.003043102379888296,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.66043030869972,
      "flow_matching_loss": 0.02608245052397251,
      "grad_norm": 0.14501848816871643,
      "h_stock_valid_norm": 97.36815567016602,
      "invalid_tail_samples": 22.5,
      "learning_rate": 3.874886710986213e-05,
      "loss": 0.0291,
      "positive_teacher_pairs": 27157.5,
      "query_h_norm_ratio": 0.4293520271778107,
      "relation_loss": 0.003043102379888296,
      "selected_pairs": 27157.5,
      "step": 17810,
      "student_distance_mean": 24.09978828430176,
      "teacher_distance_mean": 0.8526267766952514,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.003043102379888296,
      "weighted_relation_loss": 0.003043102379888296
    },
    {
      "action_query_batch_std": 0.36412598490715026,
      "action_query_norm": 41.795949172973636,
      "action_relation_loss": 0.0030480217887088657,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.669784845650142,
      "flow_matching_loss": 0.025718321464955805,
      "grad_norm": 0.09897751361131668,
      "h_stock_valid_norm": 97.95742645263672,
      "invalid_tail_samples": 23.3,
      "learning_rate": 3.8695171817819834e-05,
      "loss": 0.0288,
      "positive_teacher_pairs": 26962.6,
      "query_h_norm_ratio": 0.42668184638023376,
      "relation_loss": 0.0030480217887088657,
      "selected_pairs": 26962.6,
      "step": 17820,
      "student_distance_mean": 24.183358001708985,
      "teacher_distance_mean": 0.8608235418796539,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0030480217887088657,
      "weighted_relation_loss": 0.0030480217887088657
    },
    {
      "action_query_batch_std": 0.36321933269500734,
      "action_query_norm": 41.78700218200684,
      "action_relation_loss": 0.003201076271943748,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.67913938260056,
      "flow_matching_loss": 0.025935114920139314,
      "grad_norm": 0.12506677210330963,
      "h_stock_valid_norm": 98.2833465576172,
      "invalid_tail_samples": 24.4,
      "learning_rate": 3.8641490262209774e-05,
      "loss": 0.0291,
      "positive_teacher_pairs": 26708.2,
      "query_h_norm_ratio": 0.4251712799072266,
      "relation_loss": 0.003201076271943748,
      "selected_pairs": 26708.2,
      "step": 17830,
      "student_distance_mean": 23.996290016174317,
      "teacher_distance_mean": 0.8512551367282868,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.003201076271943748,
      "weighted_relation_loss": 0.003201076271943748
    },
    {
      "action_query_batch_std": 0.36441974341869354,
      "action_query_norm": 41.78622817993164,
      "action_relation_loss": 0.0029573758132755757,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.688493919550982,
      "flow_matching_loss": 0.024032716825604438,
      "grad_norm": 0.10145038366317749,
      "h_stock_valid_norm": 98.16840133666992,
      "invalid_tail_samples": 23.6,
      "learning_rate": 3.8587822508260084e-05,
      "loss": 0.027,
      "positive_teacher_pairs": 26896.2,
      "query_h_norm_ratio": 0.4256603330373764,
      "relation_loss": 0.0029573758132755757,
      "selected_pairs": 26896.2,
      "step": 17840,
      "student_distance_mean": 24.044185638427734,
      "teacher_distance_mean": 0.8506120026111603,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0029573758132755757,
      "weighted_relation_loss": 0.0029573758132755757
    },
    {
      "action_query_batch_std": 0.3581685870885849,
      "action_query_norm": 41.75669479370117,
      "action_relation_loss": 0.0031539607560262086,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.697848456501404,
      "flow_matching_loss": 0.025415512174367903,
      "grad_norm": 0.09055536985397339,
      "h_stock_valid_norm": 97.71227264404297,
      "invalid_tail_samples": 25.0,
      "learning_rate": 3.853416862118214e-05,
      "loss": 0.0286,
      "positive_teacher_pairs": 26569.2,
      "query_h_norm_ratio": 0.42734751403331755,
      "relation_loss": 0.0031539607560262086,
      "selected_pairs": 26569.2,
      "step": 17850,
      "student_distance_mean": 23.620727157592775,
      "teacher_distance_mean": 0.8406519770622254,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0031539607560262086,
      "weighted_relation_loss": 0.0031539607560262086
    },
    {
      "action_query_batch_std": 0.3585384666919708,
      "action_query_norm": 41.766166305541994,
      "action_relation_loss": 0.0030201033689081667,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.707202993451823,
      "flow_matching_loss": 0.025662812776863574,
      "grad_norm": 0.10343842953443527,
      "h_stock_valid_norm": 97.40045700073242,
      "invalid_tail_samples": 24.5,
      "learning_rate": 3.848052866617049e-05,
      "loss": 0.0287,
      "positive_teacher_pairs": 26688.0,
      "query_h_norm_ratio": 0.4288140803575516,
      "relation_loss": 0.0030201033689081667,
      "selected_pairs": 26688.0,
      "step": 17860,
      "student_distance_mean": 23.791986083984376,
      "teacher_distance_mean": 0.8482271313667298,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0030201033689081667,
      "weighted_relation_loss": 0.0030201033689081667
    },
    {
      "action_query_batch_std": 0.3637063354253769,
      "action_query_norm": 41.78544654846191,
      "action_relation_loss": 0.00319047118537128,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.716557530402245,
      "flow_matching_loss": 0.02714582532644272,
      "grad_norm": 0.11301974952220917,
      "h_stock_valid_norm": 97.3875862121582,
      "invalid_tail_samples": 22.6,
      "learning_rate": 3.84269027084027e-05,
      "loss": 0.0303,
      "positive_teacher_pairs": 27124.5,
      "query_h_norm_ratio": 0.4290667325258255,
      "relation_loss": 0.00319047118537128,
      "selected_pairs": 27124.5,
      "step": 17870,
      "student_distance_mean": 24.150678634643555,
      "teacher_distance_mean": 0.8563400566577911,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.00319047118537128,
      "weighted_relation_loss": 0.00319047118537128
    },
    {
      "action_query_batch_std": 0.36423035264015197,
      "action_query_norm": 41.777720260620114,
      "action_relation_loss": 0.0030049745459109546,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.725912067352667,
      "flow_matching_loss": 0.02523074299097061,
      "grad_norm": 0.12791715562343597,
      "h_stock_valid_norm": 97.61908569335938,
      "invalid_tail_samples": 26.8,
      "learning_rate": 3.83732908130394e-05,
      "loss": 0.0282,
      "positive_teacher_pairs": 26163.1,
      "query_h_norm_ratio": 0.42796880900859835,
      "relation_loss": 0.0030049745459109546,
      "selected_pairs": 26163.1,
      "step": 17880,
      "student_distance_mean": 24.07731685638428,
      "teacher_distance_mean": 0.852934592962265,
      "valid_action_samples": 229.2,
      "weighted_auxiliary_loss": 0.0030049745459109546,
      "weighted_relation_loss": 0.0030049745459109546
    },
    {
      "action_query_batch_std": 0.36623784005641935,
      "action_query_norm": 41.79116668701172,
      "action_relation_loss": 0.0030508416704833506,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.73526660430309,
      "flow_matching_loss": 0.02616732455790043,
      "grad_norm": 0.12273628264665604,
      "h_stock_valid_norm": 97.88842010498047,
      "invalid_tail_samples": 22.6,
      "learning_rate": 3.831969304522409e-05,
      "loss": 0.0292,
      "positive_teacher_pairs": 27135.7,
      "query_h_norm_ratio": 0.426928973197937,
      "relation_loss": 0.0030508416704833506,
      "selected_pairs": 27135.7,
      "step": 17890,
      "student_distance_mean": 24.21074981689453,
      "teacher_distance_mean": 0.8601294696331024,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.0030508416704833506,
      "weighted_relation_loss": 0.0030508416704833506
    },
    {
      "action_query_batch_std": 0.3667760580778122,
      "action_query_norm": 41.80298843383789,
      "action_relation_loss": 0.0031315698986873032,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.744621141253507,
      "flow_matching_loss": 0.024319344013929368,
      "grad_norm": 0.08478996157646179,
      "h_stock_valid_norm": 97.85088806152343,
      "invalid_tail_samples": 24.0,
      "learning_rate": 3.826610947008312e-05,
      "loss": 0.0275,
      "positive_teacher_pairs": 26800.8,
      "query_h_norm_ratio": 0.4272152692079544,
      "relation_loss": 0.0031315698986873032,
      "selected_pairs": 26800.8,
      "step": 17900,
      "student_distance_mean": 24.16290225982666,
      "teacher_distance_mean": 0.8561338603496551,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0031315698986873032,
      "weighted_relation_loss": 0.0031315698986873032
    },
    {
      "action_query_batch_std": 0.36110295355319977,
      "action_query_norm": 41.75442237854004,
      "action_relation_loss": 0.0030934088164940475,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.75397567820393,
      "flow_matching_loss": 0.025132245570421218,
      "grad_norm": 0.0952167958021164,
      "h_stock_valid_norm": 98.23930587768555,
      "invalid_tail_samples": 21.5,
      "learning_rate": 3.821254015272559e-05,
      "loss": 0.0282,
      "positive_teacher_pairs": 27391.1,
      "query_h_norm_ratio": 0.42503617107868197,
      "relation_loss": 0.0030934088164940475,
      "selected_pairs": 27391.1,
      "step": 17910,
      "student_distance_mean": 23.891579246520998,
      "teacher_distance_mean": 0.8473030984401703,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.0030934088164940475,
      "weighted_relation_loss": 0.0030934088164940475
    },
    {
      "action_query_batch_std": 0.36389979124069216,
      "action_query_norm": 41.76711730957031,
      "action_relation_loss": 0.0030241879168897865,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.76333021515435,
      "flow_matching_loss": 0.023127206601202488,
      "grad_norm": 0.09245281666517258,
      "h_stock_valid_norm": 98.20584182739258,
      "invalid_tail_samples": 22.7,
      "learning_rate": 3.8158985158243214e-05,
      "loss": 0.0262,
      "positive_teacher_pairs": 27115.2,
      "query_h_norm_ratio": 0.4253045529127121,
      "relation_loss": 0.0030241879168897865,
      "selected_pairs": 27115.2,
      "step": 17920,
      "student_distance_mean": 23.99544563293457,
      "teacher_distance_mean": 0.8513485372066498,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0030241879168897865,
      "weighted_relation_loss": 0.0030241879168897865
    },
    {
      "action_query_batch_std": 0.3608736634254456,
      "action_query_norm": 41.76316337585449,
      "action_relation_loss": 0.0030597822973504663,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.77268475210477,
      "flow_matching_loss": 0.025972026772797106,
      "grad_norm": 0.13848286867141724,
      "h_stock_valid_norm": 98.0023582458496,
      "invalid_tail_samples": 23.6,
      "learning_rate": 3.810544455171044e-05,
      "loss": 0.029,
      "positive_teacher_pairs": 26901.2,
      "query_h_norm_ratio": 0.42614734172821045,
      "relation_loss": 0.0030597822973504663,
      "selected_pairs": 26901.2,
      "step": 17930,
      "student_distance_mean": 23.98266258239746,
      "teacher_distance_mean": 0.8522762060165405,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0030597822973504663,
      "weighted_relation_loss": 0.0030597822973504663
    },
    {
      "action_query_batch_std": 0.36494689583778384,
      "action_query_norm": 41.78352813720703,
      "action_relation_loss": 0.0030338175129145384,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.78203928905519,
      "flow_matching_loss": 0.02560958731919527,
      "grad_norm": 0.10023251920938492,
      "h_stock_valid_norm": 98.10760803222657,
      "invalid_tail_samples": 24.6,
      "learning_rate": 3.8051918398184114e-05,
      "loss": 0.0286,
      "positive_teacher_pairs": 26666.4,
      "query_h_norm_ratio": 0.42589688301086426,
      "relation_loss": 0.0030338175129145384,
      "selected_pairs": 26666.4,
      "step": 17940,
      "student_distance_mean": 24.158586311340333,
      "teacher_distance_mean": 0.85942223072052,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0030338175129145384,
      "weighted_relation_loss": 0.0030338175129145384
    },
    {
      "action_query_batch_std": 0.36168830990791323,
      "action_query_norm": 41.744828414916995,
      "action_relation_loss": 0.003017959906719625,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.791393826005613,
      "flow_matching_loss": 0.024913938716053963,
      "grad_norm": 0.09762091189622879,
      "h_stock_valid_norm": 98.39828338623047,
      "invalid_tail_samples": 23.5,
      "learning_rate": 3.799840676270356e-05,
      "loss": 0.0279,
      "positive_teacher_pairs": 26915.6,
      "query_h_norm_ratio": 0.4242468535900116,
      "relation_loss": 0.003017959906719625,
      "selected_pairs": 26915.6,
      "step": 17950,
      "student_distance_mean": 23.97698230743408,
      "teacher_distance_mean": 0.8521566092967987,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.003017959906719625,
      "weighted_relation_loss": 0.003017959906719625
    },
    {
      "action_query_batch_std": 0.36353620886802673,
      "action_query_norm": 41.7545711517334,
      "action_relation_loss": 0.003119643428362906,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.800748362956035,
      "flow_matching_loss": 0.024634006060659887,
      "grad_norm": 0.12402661144733429,
      "h_stock_valid_norm": 98.3925552368164,
      "invalid_tail_samples": 24.4,
      "learning_rate": 3.794490971029048e-05,
      "loss": 0.0278,
      "positive_teacher_pairs": 26718.6,
      "query_h_norm_ratio": 0.4243703305721283,
      "relation_loss": 0.003119643428362906,
      "selected_pairs": 26718.6,
      "step": 17960,
      "student_distance_mean": 24.067522621154787,
      "teacher_distance_mean": 0.8551423490047455,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.003119643428362906,
      "weighted_relation_loss": 0.003119643428362906
    },
    {
      "action_query_batch_std": 0.3640559077262878,
      "action_query_norm": 41.76143684387207,
      "action_relation_loss": 0.0030974817229434846,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.810102899906454,
      "flow_matching_loss": 0.02326465379446745,
      "grad_norm": 0.08518488705158234,
      "h_stock_valid_norm": 98.22330780029297,
      "invalid_tail_samples": 22.7,
      "learning_rate": 3.789142730594881e-05,
      "loss": 0.0264,
      "positive_teacher_pairs": 27102.6,
      "query_h_norm_ratio": 0.42517215609550474,
      "relation_loss": 0.0030974817229434846,
      "selected_pairs": 27102.6,
      "step": 17970,
      "student_distance_mean": 24.119810104370117,
      "teacher_distance_mean": 0.8575285196304321,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0030974817229434846,
      "weighted_relation_loss": 0.0030974817229434846
    },
    {
      "action_query_batch_std": 0.3653372496366501,
      "action_query_norm": 41.76680030822754,
      "action_relation_loss": 0.003184578032232821,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.819457436856876,
      "flow_matching_loss": 0.026353003829717635,
      "grad_norm": 0.1240948960185051,
      "h_stock_valid_norm": 97.878076171875,
      "invalid_tail_samples": 22.8,
      "learning_rate": 3.783795961466471e-05,
      "loss": 0.0295,
      "positive_teacher_pairs": 27081.8,
      "query_h_norm_ratio": 0.42672932147979736,
      "relation_loss": 0.003184578032232821,
      "selected_pairs": 27081.8,
      "step": 17980,
      "student_distance_mean": 24.091166687011718,
      "teacher_distance_mean": 0.8577366948127747,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.003184578032232821,
      "weighted_relation_loss": 0.003184578032232821
    },
    {
      "action_query_batch_std": 0.3651263952255249,
      "action_query_norm": 41.756945037841795,
      "action_relation_loss": 0.003040272439830005,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.828811973807298,
      "flow_matching_loss": 0.024795405380427838,
      "grad_norm": 0.13188683986663818,
      "h_stock_valid_norm": 97.77663040161133,
      "invalid_tail_samples": 22.9,
      "learning_rate": 3.778450670140651e-05,
      "loss": 0.0278,
      "positive_teacher_pairs": 27055.1,
      "query_h_norm_ratio": 0.42706628441810607,
      "relation_loss": 0.003040272439830005,
      "selected_pairs": 27055.1,
      "step": 17990,
      "student_distance_mean": 24.146992874145507,
      "teacher_distance_mean": 0.8596073925495148,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.003040272439830005,
      "weighted_relation_loss": 0.003040272439830005
    },
    {
      "action_query_batch_std": 0.36569435596466066,
      "action_query_norm": 41.761348724365234,
      "action_relation_loss": 0.0030257651349529626,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.838166510757716,
      "flow_matching_loss": 0.023961673863232134,
      "grad_norm": 0.10875026881694794,
      "h_stock_valid_norm": 97.88146057128907,
      "invalid_tail_samples": 25.9,
      "learning_rate": 3.773106863112451e-05,
      "loss": 0.027,
      "positive_teacher_pairs": 26365.2,
      "query_h_norm_ratio": 0.4266556054353714,
      "relation_loss": 0.0030257651349529626,
      "selected_pairs": 26365.2,
      "step": 18000,
      "student_distance_mean": 24.174310874938964,
      "teacher_distance_mean": 0.8596365213394165,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.0030257651349529626,
      "weighted_relation_loss": 0.0030257651349529626
    },
    {
      "action_query_batch_std": 0.36188589930534365,
      "action_query_norm": 41.726511001586914,
      "action_relation_loss": 0.003123543178662658,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.84752104770814,
      "flow_matching_loss": 0.027375403232872485,
      "grad_norm": 0.11549235135316849,
      "h_stock_valid_norm": 98.14436950683594,
      "invalid_tail_samples": 23.0,
      "learning_rate": 3.7677645468751e-05,
      "loss": 0.0305,
      "positive_teacher_pairs": 27034.3,
      "query_h_norm_ratio": 0.42515631318092345,
      "relation_loss": 0.003123543178662658,
      "selected_pairs": 27034.3,
      "step": 18010,
      "student_distance_mean": 23.92507858276367,
      "teacher_distance_mean": 0.850350534915924,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.003123543178662658,
      "weighted_relation_loss": 0.003123543178662658
    },
    {
      "action_query_batch_std": 0.36352703273296355,
      "action_query_norm": 41.7309326171875,
      "action_relation_loss": 0.0028699932154268026,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.85687558465856,
      "flow_matching_loss": 0.026638811640441417,
      "grad_norm": 0.11529260128736496,
      "h_stock_valid_norm": 98.30764923095703,
      "invalid_tail_samples": 23.2,
      "learning_rate": 3.762423727920018e-05,
      "loss": 0.0295,
      "positive_teacher_pairs": 26987.7,
      "query_h_norm_ratio": 0.4244953006505966,
      "relation_loss": 0.0028699932154268026,
      "selected_pairs": 26987.7,
      "step": 18020,
      "student_distance_mean": 24.085844612121583,
      "teacher_distance_mean": 0.8598633885383606,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0028699932154268026,
      "weighted_relation_loss": 0.0028699932154268026
    },
    {
      "action_query_batch_std": 0.3611134350299835,
      "action_query_norm": 41.71107215881348,
      "action_relation_loss": 0.0030232404824346304,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.86623012160898,
      "flow_matching_loss": 0.02420760504901409,
      "grad_norm": 0.09131027013063431,
      "h_stock_valid_norm": 98.36788864135742,
      "invalid_tail_samples": 26.1,
      "learning_rate": 3.7570844127367995e-05,
      "loss": 0.0272,
      "positive_teacher_pairs": 26320.8,
      "query_h_norm_ratio": 0.42403591275215147,
      "relation_loss": 0.0030232404824346304,
      "selected_pairs": 26320.8,
      "step": 18030,
      "student_distance_mean": 23.99421672821045,
      "teacher_distance_mean": 0.8531546771526337,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.0030232404824346304,
      "weighted_relation_loss": 0.0030232404824346304
    },
    {
      "action_query_batch_std": 0.36079617738723757,
      "action_query_norm": 41.727745819091794,
      "action_relation_loss": 0.0030693101696670057,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.8755846585594,
      "flow_matching_loss": 0.026305234245955943,
      "grad_norm": 0.08836756646633148,
      "h_stock_valid_norm": 97.94544830322266,
      "invalid_tail_samples": 24.7,
      "learning_rate": 3.751746607813221e-05,
      "loss": 0.0294,
      "positive_teacher_pairs": 26644.7,
      "query_h_norm_ratio": 0.4260342657566071,
      "relation_loss": 0.0030693101696670057,
      "selected_pairs": 26644.7,
      "step": 18040,
      "student_distance_mean": 23.908216667175292,
      "teacher_distance_mean": 0.8497032642364502,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0030693101696670057,
      "weighted_relation_loss": 0.0030693101696670057
    },
    {
      "action_query_batch_std": 0.35849824249744416,
      "action_query_norm": 41.69039726257324,
      "action_relation_loss": 0.0030533257639035584,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.884939195509823,
      "flow_matching_loss": 0.024206959269940855,
      "grad_norm": 0.09222115576267242,
      "h_stock_valid_norm": 97.34554824829101,
      "invalid_tail_samples": 26.9,
      "learning_rate": 3.746410319635217e-05,
      "loss": 0.0273,
      "positive_teacher_pairs": 26136.2,
      "query_h_norm_ratio": 0.4282746732234955,
      "relation_loss": 0.0030533257639035584,
      "selected_pairs": 26136.2,
      "step": 18050,
      "student_distance_mean": 23.743530654907225,
      "teacher_distance_mean": 0.8476254224777222,
      "valid_action_samples": 229.1,
      "weighted_auxiliary_loss": 0.0030533257639035584,
      "weighted_relation_loss": 0.0030533257639035584
    },
    {
      "action_query_batch_std": 0.3635400921106339,
      "action_query_norm": 41.73268127441406,
      "action_relation_loss": 0.0028634793125092983,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.894293732460245,
      "flow_matching_loss": 0.02429398149251938,
      "grad_norm": 0.11779965460300446,
      "h_stock_valid_norm": 97.3287727355957,
      "invalid_tail_samples": 24.6,
      "learning_rate": 3.7410755546868806e-05,
      "loss": 0.0272,
      "positive_teacher_pairs": 26662.5,
      "query_h_norm_ratio": 0.4287839025259018,
      "relation_loss": 0.0028634793125092983,
      "selected_pairs": 26662.5,
      "step": 18060,
      "student_distance_mean": 24.087239837646486,
      "teacher_distance_mean": 0.8561741054058075,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0028634793125092983,
      "weighted_relation_loss": 0.0028634793125092983
    },
    {
      "action_query_batch_std": 0.3623620718717575,
      "action_query_norm": 41.72572555541992,
      "action_relation_loss": 0.0030342989834025503,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.903648269410663,
      "flow_matching_loss": 0.025045400112867357,
      "grad_norm": 0.0938742533326149,
      "h_stock_valid_norm": 98.11281356811523,
      "invalid_tail_samples": 23.5,
      "learning_rate": 3.735742319450454e-05,
      "loss": 0.0281,
      "positive_teacher_pairs": 26920.2,
      "query_h_norm_ratio": 0.4252855688333511,
      "relation_loss": 0.0030342989834025503,
      "selected_pairs": 26920.2,
      "step": 18070,
      "student_distance_mean": 24.014798164367676,
      "teacher_distance_mean": 0.8519567549228668,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0030342989834025503,
      "weighted_relation_loss": 0.0030342989834025503
    },
    {
      "action_query_batch_std": 0.36480668783187864,
      "action_query_norm": 41.72474670410156,
      "action_relation_loss": 0.0030107477447018026,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.913002806361085,
      "flow_matching_loss": 0.026067092642188074,
      "grad_norm": 0.10214141756296158,
      "h_stock_valid_norm": 98.57879180908203,
      "invalid_tail_samples": 23.6,
      "learning_rate": 3.7304106204063184e-05,
      "loss": 0.0291,
      "positive_teacher_pairs": 26907.8,
      "query_h_norm_ratio": 0.42326479256153104,
      "relation_loss": 0.0030107477447018026,
      "selected_pairs": 26907.8,
      "step": 18080,
      "student_distance_mean": 24.059994316101076,
      "teacher_distance_mean": 0.853370064496994,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0030107477447018026,
      "weighted_relation_loss": 0.0030107477447018026
    },
    {
      "action_query_batch_std": 0.3622566550970078,
      "action_query_norm": 41.70991897583008,
      "action_relation_loss": 0.0029932038392871617,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.922357343311507,
      "flow_matching_loss": 0.026174492575228213,
      "grad_norm": 0.12561523914337158,
      "h_stock_valid_norm": 98.72129898071289,
      "invalid_tail_samples": 21.8,
      "learning_rate": 3.725080464032996e-05,
      "loss": 0.0292,
      "positive_teacher_pairs": 27315.1,
      "query_h_norm_ratio": 0.42250754237174987,
      "relation_loss": 0.0029932038392871617,
      "selected_pairs": 27315.1,
      "step": 18090,
      "student_distance_mean": 23.904355430603026,
      "teacher_distance_mean": 0.8496035516262055,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.0029932038392871617,
      "weighted_relation_loss": 0.0029932038392871617
    },
    {
      "action_query_batch_std": 0.36110522747039797,
      "action_query_norm": 41.69643440246582,
      "action_relation_loss": 0.0029591638362035156,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.931711880261926,
      "flow_matching_loss": 0.023776398040354252,
      "grad_norm": 0.12945249676704407,
      "h_stock_valid_norm": 98.34738082885742,
      "invalid_tail_samples": 24.9,
      "learning_rate": 3.719751856807125e-05,
      "loss": 0.0267,
      "positive_teacher_pairs": 26604.6,
      "query_h_norm_ratio": 0.42397373616695405,
      "relation_loss": 0.0029591638362035156,
      "selected_pairs": 26604.6,
      "step": 18100,
      "student_distance_mean": 23.7261531829834,
      "teacher_distance_mean": 0.8436450481414794,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0029591638362035156,
      "weighted_relation_loss": 0.0029591638362035156
    },
    {
      "action_query_batch_std": 0.36248416602611544,
      "action_query_norm": 41.70823211669922,
      "action_relation_loss": 0.002944376436062157,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.941066417212348,
      "flow_matching_loss": 0.025452737510204316,
      "grad_norm": 0.1396999955177307,
      "h_stock_valid_norm": 98.01184997558593,
      "invalid_tail_samples": 26.8,
      "learning_rate": 3.71442480520347e-05,
      "loss": 0.0284,
      "positive_teacher_pairs": 26166.6,
      "query_h_norm_ratio": 0.42554361522197726,
      "relation_loss": 0.002944376436062157,
      "selected_pairs": 26166.6,
      "step": 18110,
      "student_distance_mean": 24.05455150604248,
      "teacher_distance_mean": 0.855136114358902,
      "valid_action_samples": 229.2,
      "weighted_auxiliary_loss": 0.002944376436062157,
      "weighted_relation_loss": 0.002944376436062157
    },
    {
      "action_query_batch_std": 0.3627426475286484,
      "action_query_norm": 41.71983528137207,
      "action_relation_loss": 0.003071920480579138,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.95042095416277,
      "flow_matching_loss": 0.02635882403701544,
      "grad_norm": 0.09250419586896896,
      "h_stock_valid_norm": 97.82924728393554,
      "invalid_tail_samples": 24.6,
      "learning_rate": 3.709099315694897e-05,
      "loss": 0.0294,
      "positive_teacher_pairs": 26676.7,
      "query_h_norm_ratio": 0.4264575481414795,
      "relation_loss": 0.003071920480579138,
      "selected_pairs": 26676.7,
      "step": 18120,
      "student_distance_mean": 24.084455680847167,
      "teacher_distance_mean": 0.8545642793178558,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.003071920480579138,
      "weighted_relation_loss": 0.003071920480579138
    },
    {
      "action_query_batch_std": 0.3650729775428772,
      "action_query_norm": 41.7307804107666,
      "action_relation_loss": 0.0028829564340412616,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.95977549111319,
      "flow_matching_loss": 0.026037717051804066,
      "grad_norm": 0.11154354363679886,
      "h_stock_valid_norm": 97.93746795654297,
      "invalid_tail_samples": 25.8,
      "learning_rate": 3.703775394752378e-05,
      "loss": 0.0289,
      "positive_teacher_pairs": 26386.2,
      "query_h_norm_ratio": 0.4260986179113388,
      "relation_loss": 0.0028829564340412616,
      "selected_pairs": 26386.2,
      "step": 18130,
      "student_distance_mean": 24.204815673828126,
      "teacher_distance_mean": 0.8648471117019654,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0028829564340412616,
      "weighted_relation_loss": 0.0028829564340412616
    },
    {
      "action_query_batch_std": 0.3565308749675751,
      "action_query_norm": 41.67600288391113,
      "action_relation_loss": 0.0031146782683208587,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.96913002806361,
      "flow_matching_loss": 0.023781098984181882,
      "grad_norm": 0.0902317613363266,
      "h_stock_valid_norm": 98.08640670776367,
      "invalid_tail_samples": 25.1,
      "learning_rate": 3.698453048844983e-05,
      "loss": 0.0269,
      "positive_teacher_pairs": 26548.9,
      "query_h_norm_ratio": 0.42489463090896606,
      "relation_loss": 0.0031146782683208587,
      "selected_pairs": 26548.9,
      "step": 18140,
      "student_distance_mean": 23.698288917541504,
      "teacher_distance_mean": 0.8471561431884765,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0031146782683208587,
      "weighted_relation_loss": 0.0031146782683208587
    },
    {
      "action_query_batch_std": 0.3667699694633484,
      "action_query_norm": 41.717752075195314,
      "action_relation_loss": 0.0030469123274087908,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.978484565014032,
      "flow_matching_loss": 0.026194930635392667,
      "grad_norm": 0.12900060415267944,
      "h_stock_valid_norm": 98.30652465820313,
      "invalid_tail_samples": 24.8,
      "learning_rate": 3.693132284439861e-05,
      "loss": 0.0292,
      "positive_teacher_pairs": 26621.1,
      "query_h_norm_ratio": 0.4243648201227188,
      "relation_loss": 0.0030469123274087908,
      "selected_pairs": 26621.1,
      "step": 18150,
      "student_distance_mean": 24.19867820739746,
      "teacher_distance_mean": 0.8641947209835052,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0030469123274087908,
      "weighted_relation_loss": 0.0030469123274087908
    },
    {
      "action_query_batch_std": 0.36200471222400665,
      "action_query_norm": 41.71583251953125,
      "action_relation_loss": 0.0031021406408399343,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.987839101964454,
      "flow_matching_loss": 0.025592537224292757,
      "grad_norm": 0.12976840138435364,
      "h_stock_valid_norm": 98.25661087036133,
      "invalid_tail_samples": 23.0,
      "learning_rate": 3.6878131080022414e-05,
      "loss": 0.0287,
      "positive_teacher_pairs": 27038.6,
      "query_h_norm_ratio": 0.42456182539463044,
      "relation_loss": 0.0031021406408399343,
      "selected_pairs": 27038.6,
      "step": 18160,
      "student_distance_mean": 23.961035537719727,
      "teacher_distance_mean": 0.8513702929019928,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0031021406408399343,
      "weighted_relation_loss": 0.0031021406408399343
    },
    {
      "action_query_batch_std": 0.3612460732460022,
      "action_query_norm": 41.69169692993164,
      "action_relation_loss": 0.003010177635587752,
      "auxiliary_loss_weight": 1.0,
      "epoch": 16.997193638914872,
      "flow_matching_loss": 0.02644976507872343,
      "grad_norm": 0.12896575033664703,
      "h_stock_valid_norm": 97.81660919189453,
      "invalid_tail_samples": 22.5,
      "learning_rate": 3.682495525995429e-05,
      "loss": 0.0295,
      "positive_teacher_pairs": 27161.7,
      "query_h_norm_ratio": 0.4262283146381378,
      "relation_loss": 0.003010177635587752,
      "selected_pairs": 27161.7,
      "step": 18170,
      "student_distance_mean": 23.952720832824706,
      "teacher_distance_mean": 0.8492787301540374,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.003010177635587752,
      "weighted_relation_loss": 0.003010177635587752
    },
    {
      "action_query_batch_std": 0.360585430264473,
      "action_query_norm": 41.68551521301269,
      "action_relation_loss": 0.0030724145472049713,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.006548175865294,
      "flow_matching_loss": 0.02749968096613884,
      "grad_norm": 0.14033108949661255,
      "h_stock_valid_norm": 97.58946685791015,
      "invalid_tail_samples": 21.6,
      "learning_rate": 3.6771795448807846e-05,
      "loss": 0.0306,
      "positive_teacher_pairs": 24369.6,
      "query_h_norm_ratio": 0.4271548897027969,
      "relation_loss": 0.0030724145472049713,
      "selected_pairs": 24369.6,
      "step": 18180,
      "student_distance_mean": 23.845263481140137,
      "teacher_distance_mean": 0.8496906578540802,
      "valid_action_samples": 214.5,
      "weighted_auxiliary_loss": 0.0030724145472049713,
      "weighted_relation_loss": 0.0030724145472049713
    },
    {
      "action_query_batch_std": 0.3643040329217911,
      "action_query_norm": 41.71104049682617,
      "action_relation_loss": 0.002972286194562912,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.015902712815716,
      "flow_matching_loss": 0.023052923567593097,
      "grad_norm": 0.09835197776556015,
      "h_stock_valid_norm": 97.27781448364257,
      "invalid_tail_samples": 23.1,
      "learning_rate": 3.671865171117724e-05,
      "loss": 0.026,
      "positive_teacher_pairs": 27020.0,
      "query_h_norm_ratio": 0.4287847399711609,
      "relation_loss": 0.002972286194562912,
      "selected_pairs": 27020.0,
      "step": 18190,
      "student_distance_mean": 24.160652923583985,
      "teacher_distance_mean": 0.858939653635025,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002972286194562912,
      "weighted_relation_loss": 0.002972286194562912
    },
    {
      "action_query_batch_std": 0.3612354040145874,
      "action_query_norm": 41.68938217163086,
      "action_relation_loss": 0.002922644722275436,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.02525724976614,
      "flow_matching_loss": 0.02519606128334999,
      "grad_norm": 0.10802722722291946,
      "h_stock_valid_norm": 97.72183456420899,
      "invalid_tail_samples": 26.3,
      "learning_rate": 3.666552411163718e-05,
      "loss": 0.0281,
      "positive_teacher_pairs": 26274.7,
      "query_h_norm_ratio": 0.4266165763139725,
      "relation_loss": 0.002922644722275436,
      "selected_pairs": 26274.7,
      "step": 18200,
      "student_distance_mean": 24.004984283447264,
      "teacher_distance_mean": 0.8541640400886535,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.002922644722275436,
      "weighted_relation_loss": 0.002922644722275436
    },
    {
      "action_query_batch_std": 0.36914218962192535,
      "action_query_norm": 41.72914772033691,
      "action_relation_loss": 0.0028691526502370836,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.034611786716557,
      "flow_matching_loss": 0.02500540539622307,
      "grad_norm": 0.10675127059221268,
      "h_stock_valid_norm": 98.5427032470703,
      "invalid_tail_samples": 21.2,
      "learning_rate": 3.661241271474269e-05,
      "loss": 0.0279,
      "positive_teacher_pairs": 27469.4,
      "query_h_norm_ratio": 0.42346576452255247,
      "relation_loss": 0.0028691526502370836,
      "selected_pairs": 27469.4,
      "step": 18210,
      "student_distance_mean": 24.359095764160156,
      "teacher_distance_mean": 0.872089010477066,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.0028691526502370836,
      "weighted_relation_loss": 0.0028691526502370836
    },
    {
      "action_query_batch_std": 0.3642656534910202,
      "action_query_norm": 41.68134384155273,
      "action_relation_loss": 0.003001631237566471,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.04396632366698,
      "flow_matching_loss": 0.023001698963344096,
      "grad_norm": 0.11553601920604706,
      "h_stock_valid_norm": 98.60444412231445,
      "invalid_tail_samples": 22.3,
      "learning_rate": 3.6559317585029116e-05,
      "loss": 0.026,
      "positive_teacher_pairs": 27208.8,
      "query_h_norm_ratio": 0.4227137416601181,
      "relation_loss": 0.003001631237566471,
      "selected_pairs": 27208.8,
      "step": 18220,
      "student_distance_mean": 24.16193332672119,
      "teacher_distance_mean": 0.8588692128658295,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.003001631237566471,
      "weighted_relation_loss": 0.003001631237566471
    },
    {
      "action_query_batch_std": 0.3609608203172684,
      "action_query_norm": 41.68962936401367,
      "action_relation_loss": 0.002990655950270593,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.0533208606174,
      "flow_matching_loss": 0.02378982901573181,
      "grad_norm": 0.11578278988599777,
      "h_stock_valid_norm": 98.39960174560547,
      "invalid_tail_samples": 22.2,
      "learning_rate": 3.650623878701204e-05,
      "loss": 0.0268,
      "positive_teacher_pairs": 27229.6,
      "query_h_norm_ratio": 0.42367938756942747,
      "relation_loss": 0.002990655950270593,
      "selected_pairs": 27229.6,
      "step": 18230,
      "student_distance_mean": 23.934638214111327,
      "teacher_distance_mean": 0.8506628930568695,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.002990655950270593,
      "weighted_relation_loss": 0.002990655950270593
    },
    {
      "action_query_batch_std": 0.36518117487430574,
      "action_query_norm": 41.709906005859374,
      "action_relation_loss": 0.002934729913249612,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.06267539756782,
      "flow_matching_loss": 0.024838145077228545,
      "grad_norm": 0.08129821717739105,
      "h_stock_valid_norm": 98.07554626464844,
      "invalid_tail_samples": 23.4,
      "learning_rate": 3.645317638518721e-05,
      "loss": 0.0278,
      "positive_teacher_pairs": 26947.0,
      "query_h_norm_ratio": 0.4252849996089935,
      "relation_loss": 0.002934729913249612,
      "selected_pairs": 26947.0,
      "step": 18240,
      "student_distance_mean": 24.091669273376464,
      "teacher_distance_mean": 0.8589485883712769,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002934729913249612,
      "weighted_relation_loss": 0.002934729913249612
    },
    {
      "action_query_batch_std": 0.36142740249633787,
      "action_query_norm": 41.68663902282715,
      "action_relation_loss": 0.0028446570038795473,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.07202993451824,
      "flow_matching_loss": 0.02487682644277811,
      "grad_norm": 0.11305883526802063,
      "h_stock_valid_norm": 97.89657821655274,
      "invalid_tail_samples": 25.3,
      "learning_rate": 3.640013044403046e-05,
      "loss": 0.0277,
      "positive_teacher_pairs": 26506.6,
      "query_h_norm_ratio": 0.425825235247612,
      "relation_loss": 0.0028446570038795473,
      "selected_pairs": 26506.6,
      "step": 18250,
      "student_distance_mean": 23.964305877685547,
      "teacher_distance_mean": 0.8517720997333527,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0028446570038795473,
      "weighted_relation_loss": 0.0028446570038795473
    },
    {
      "action_query_batch_std": 0.3650973796844482,
      "action_query_norm": 41.6997932434082,
      "action_relation_loss": 0.0029986944515258073,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.081384471468663,
      "flow_matching_loss": 0.02530914917588234,
      "grad_norm": 0.09654460102319717,
      "h_stock_valid_norm": 98.14366989135742,
      "invalid_tail_samples": 24.3,
      "learning_rate": 3.634710102799761e-05,
      "loss": 0.0283,
      "positive_teacher_pairs": 26735.2,
      "query_h_norm_ratio": 0.4248879700899124,
      "relation_loss": 0.0029986944515258073,
      "selected_pairs": 26735.2,
      "step": 18260,
      "student_distance_mean": 24.04827461242676,
      "teacher_distance_mean": 0.8515469789505005,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0029986944515258073,
      "weighted_relation_loss": 0.0029986944515258073
    },
    {
      "action_query_batch_std": 0.36536299586296084,
      "action_query_norm": 41.71296768188476,
      "action_relation_loss": 0.002717973734252155,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.09073900841908,
      "flow_matching_loss": 0.023853123374283313,
      "grad_norm": 0.10402727127075195,
      "h_stock_valid_norm": 98.47124938964843,
      "invalid_tail_samples": 23.3,
      "learning_rate": 3.6294088201524394e-05,
      "loss": 0.0266,
      "positive_teacher_pairs": 26966.8,
      "query_h_norm_ratio": 0.4236080676317215,
      "relation_loss": 0.002717973734252155,
      "selected_pairs": 26966.8,
      "step": 18270,
      "student_distance_mean": 24.21724510192871,
      "teacher_distance_mean": 0.8632014870643616,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.002717973734252155,
      "weighted_relation_loss": 0.002717973734252155
    },
    {
      "action_query_batch_std": 0.36519127190113065,
      "action_query_norm": 41.6826889038086,
      "action_relation_loss": 0.0027465389342978598,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.100093545369504,
      "flow_matching_loss": 0.02528014462441206,
      "grad_norm": 0.11391937732696533,
      "h_stock_valid_norm": 98.33525772094727,
      "invalid_tail_samples": 24.8,
      "learning_rate": 3.62410920290264e-05,
      "loss": 0.028,
      "positive_teacher_pairs": 26620.9,
      "query_h_norm_ratio": 0.4238863468170166,
      "relation_loss": 0.0027465389342978598,
      "selected_pairs": 26620.9,
      "step": 18280,
      "student_distance_mean": 23.968448638916016,
      "teacher_distance_mean": 0.8540900886058808,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0027465389342978598,
      "weighted_relation_loss": 0.0027465389342978598
    },
    {
      "action_query_batch_std": 0.36223839223384857,
      "action_query_norm": 41.68694305419922,
      "action_relation_loss": 0.002850855770520866,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.109448082319926,
      "flow_matching_loss": 0.024626548402011395,
      "grad_norm": 0.097684845328331,
      "h_stock_valid_norm": 98.30697402954101,
      "invalid_tail_samples": 25.0,
      "learning_rate": 3.6188112574898954e-05,
      "loss": 0.0275,
      "positive_teacher_pairs": 26571.9,
      "query_h_norm_ratio": 0.42405077517032624,
      "relation_loss": 0.002850855770520866,
      "selected_pairs": 26571.9,
      "step": 18290,
      "student_distance_mean": 23.997918701171876,
      "teacher_distance_mean": 0.8533148527145386,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.002850855770520866,
      "weighted_relation_loss": 0.002850855770520866
    },
    {
      "action_query_batch_std": 0.35809951424598696,
      "action_query_norm": 41.66551322937012,
      "action_relation_loss": 0.002977875014767051,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.118802619270348,
      "flow_matching_loss": 0.02614122647792101,
      "grad_norm": 0.11779092252254486,
      "h_stock_valid_norm": 98.26112442016601,
      "invalid_tail_samples": 23.8,
      "learning_rate": 3.6135149903517115e-05,
      "loss": 0.0291,
      "positive_teacher_pairs": 26864.4,
      "query_h_norm_ratio": 0.42403114438056944,
      "relation_loss": 0.002977875014767051,
      "selected_pairs": 26864.4,
      "step": 18300,
      "student_distance_mean": 23.738253784179687,
      "teacher_distance_mean": 0.8460357964038849,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.002977875014767051,
      "weighted_relation_loss": 0.002977875014767051
    },
    {
      "action_query_batch_std": 0.36427724957466123,
      "action_query_norm": 41.70809783935547,
      "action_relation_loss": 0.0028905817307531834,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.128157156220766,
      "flow_matching_loss": 0.026361642591655254,
      "grad_norm": 0.17282699048519135,
      "h_stock_valid_norm": 98.01910400390625,
      "invalid_tail_samples": 25.0,
      "learning_rate": 3.608220407923552e-05,
      "loss": 0.0293,
      "positive_teacher_pairs": 26574.4,
      "query_h_norm_ratio": 0.42551229894161224,
      "relation_loss": 0.0028905817307531834,
      "selected_pairs": 26574.4,
      "step": 18310,
      "student_distance_mean": 24.211251258850098,
      "teacher_distance_mean": 0.8614311277866363,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0028905817307531834,
      "weighted_relation_loss": 0.0028905817307531834
    },
    {
      "action_query_batch_std": 0.36465119421482084,
      "action_query_norm": 41.695021057128905,
      "action_relation_loss": 0.002816159254871309,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.137511693171188,
      "flow_matching_loss": 0.02517856229096651,
      "grad_norm": 0.12203172594308853,
      "h_stock_valid_norm": 97.51689682006835,
      "invalid_tail_samples": 22.1,
      "learning_rate": 3.602927516638833e-05,
      "loss": 0.028,
      "positive_teacher_pairs": 27247.5,
      "query_h_norm_ratio": 0.42756868004798887,
      "relation_loss": 0.002816159254871309,
      "selected_pairs": 27247.5,
      "step": 18320,
      "student_distance_mean": 24.051160430908205,
      "teacher_distance_mean": 0.8562282681465149,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.002816159254871309,
      "weighted_relation_loss": 0.002816159254871309
    },
    {
      "action_query_batch_std": 0.36026560962200166,
      "action_query_norm": 41.672864532470705,
      "action_relation_loss": 0.0029940373031422498,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.14686623012161,
      "flow_matching_loss": 0.02238449901342392,
      "grad_norm": 0.0960397720336914,
      "h_stock_valid_norm": 97.55297393798828,
      "invalid_tail_samples": 23.0,
      "learning_rate": 3.5976363229289165e-05,
      "loss": 0.0254,
      "positive_teacher_pairs": 27034.1,
      "query_h_norm_ratio": 0.4271842449903488,
      "relation_loss": 0.0029940373031422498,
      "selected_pairs": 27034.1,
      "step": 18330,
      "student_distance_mean": 23.80372714996338,
      "teacher_distance_mean": 0.8467571496963501,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0029940373031422498,
      "weighted_relation_loss": 0.0029940373031422498
    },
    {
      "action_query_batch_std": 0.36443125903606416,
      "action_query_norm": 41.67204666137695,
      "action_relation_loss": 0.002976235281676054,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.15622076707203,
      "flow_matching_loss": 0.02356984782963991,
      "grad_norm": 0.11420316249132156,
      "h_stock_valid_norm": 97.7720115661621,
      "invalid_tail_samples": 23.8,
      "learning_rate": 3.5923468332231e-05,
      "loss": 0.0265,
      "positive_teacher_pairs": 26848.3,
      "query_h_norm_ratio": 0.4262185126543045,
      "relation_loss": 0.002976235281676054,
      "selected_pairs": 26848.3,
      "step": 18340,
      "student_distance_mean": 24.046290016174318,
      "teacher_distance_mean": 0.8564862966537475,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.002976235281676054,
      "weighted_relation_loss": 0.002976235281676054
    },
    {
      "action_query_batch_std": 0.36238149404525755,
      "action_query_norm": 41.68513984680176,
      "action_relation_loss": 0.0028290603077039123,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.16557530402245,
      "flow_matching_loss": 0.024013211391866206,
      "grad_norm": 0.09999830275774002,
      "h_stock_valid_norm": 98.0023094177246,
      "invalid_tail_samples": 23.1,
      "learning_rate": 3.587059053948616e-05,
      "loss": 0.0268,
      "positive_teacher_pairs": 27015.3,
      "query_h_norm_ratio": 0.42534978687763214,
      "relation_loss": 0.0028290603077039123,
      "selected_pairs": 27015.3,
      "step": 18350,
      "student_distance_mean": 23.959156036376953,
      "teacher_distance_mean": 0.8527961492538452,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0028290603077039123,
      "weighted_relation_loss": 0.0028290603077039123
    },
    {
      "action_query_batch_std": 0.3639482170343399,
      "action_query_norm": 41.683793258666995,
      "action_relation_loss": 0.0027932507218793035,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.174929840972872,
      "flow_matching_loss": 0.02620099950581789,
      "grad_norm": 0.1112297773361206,
      "h_stock_valid_norm": 98.04197235107422,
      "invalid_tail_samples": 25.0,
      "learning_rate": 3.5817729915306134e-05,
      "loss": 0.029,
      "positive_teacher_pairs": 26570.7,
      "query_h_norm_ratio": 0.42516433000564574,
      "relation_loss": 0.0027932507218793035,
      "selected_pairs": 26570.7,
      "step": 18360,
      "student_distance_mean": 24.059262847900392,
      "teacher_distance_mean": 0.854728764295578,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0027932507218793035,
      "weighted_relation_loss": 0.0027932507218793035
    },
    {
      "action_query_batch_std": 0.3611216813325882,
      "action_query_norm": 41.67618103027344,
      "action_relation_loss": 0.0028831642121076584,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.184284377923294,
      "flow_matching_loss": 0.02303969133645296,
      "grad_norm": 0.12999185919761658,
      "h_stock_valid_norm": 98.51090621948242,
      "invalid_tail_samples": 24.5,
      "learning_rate": 3.5764886523921566e-05,
      "loss": 0.0259,
      "positive_teacher_pairs": 26692.0,
      "query_h_norm_ratio": 0.42306545078754426,
      "relation_loss": 0.0028831642121076584,
      "selected_pairs": 26692.0,
      "step": 18370,
      "student_distance_mean": 23.8848237991333,
      "teacher_distance_mean": 0.8485233366489411,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0028831642121076584,
      "weighted_relation_loss": 0.0028831642121076584
    },
    {
      "action_query_batch_std": 0.3630338340997696,
      "action_query_norm": 41.67736625671387,
      "action_relation_loss": 0.002829675143584609,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.193638914873713,
      "flow_matching_loss": 0.025218763388693333,
      "grad_norm": 0.10477457195520401,
      "h_stock_valid_norm": 98.65071029663086,
      "invalid_tail_samples": 22.9,
      "learning_rate": 3.571206042954214e-05,
      "loss": 0.028,
      "positive_teacher_pairs": 27058.9,
      "query_h_norm_ratio": 0.42247456312179565,
      "relation_loss": 0.002829675143584609,
      "selected_pairs": 27058.9,
      "step": 18380,
      "student_distance_mean": 24.117329025268553,
      "teacher_distance_mean": 0.8597762644290924,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.002829675143584609,
      "weighted_relation_loss": 0.002829675143584609
    },
    {
      "action_query_batch_std": 0.36303224265575407,
      "action_query_norm": 41.6546127319336,
      "action_relation_loss": 0.002875361847691238,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.202993451824135,
      "flow_matching_loss": 0.02370724957436323,
      "grad_norm": 0.10584235936403275,
      "h_stock_valid_norm": 98.8154510498047,
      "invalid_tail_samples": 23.5,
      "learning_rate": 3.5659251696356566e-05,
      "loss": 0.0266,
      "positive_teacher_pairs": 26925.7,
      "query_h_norm_ratio": 0.42154479026794434,
      "relation_loss": 0.002875361847691238,
      "selected_pairs": 26925.7,
      "step": 18390,
      "student_distance_mean": 23.986343383789062,
      "teacher_distance_mean": 0.852979975938797,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.002875361847691238,
      "weighted_relation_loss": 0.002875361847691238
    },
    {
      "action_query_batch_std": 0.36517757177352905,
      "action_query_norm": 41.69330596923828,
      "action_relation_loss": 0.0027343094116076827,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.212347988774557,
      "flow_matching_loss": 0.023292243666946887,
      "grad_norm": 0.09211388975381851,
      "h_stock_valid_norm": 98.64687271118164,
      "invalid_tail_samples": 24.0,
      "learning_rate": 3.56064603885324e-05,
      "loss": 0.026,
      "positive_teacher_pairs": 26805.2,
      "query_h_norm_ratio": 0.4226580262184143,
      "relation_loss": 0.0027343094116076827,
      "selected_pairs": 26805.2,
      "step": 18400,
      "student_distance_mean": 24.23832778930664,
      "teacher_distance_mean": 0.8671962976455688,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0027343094116076827,
      "weighted_relation_loss": 0.0027343094116076827
    },
    {
      "action_query_batch_std": 0.3635497331619263,
      "action_query_norm": 41.64541893005371,
      "action_relation_loss": 0.0027664586901664733,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.221702525724975,
      "flow_matching_loss": 0.02356854286044836,
      "grad_norm": 0.11996409296989441,
      "h_stock_valid_norm": 98.50231323242187,
      "invalid_tail_samples": 28.2,
      "learning_rate": 3.5553686570216115e-05,
      "loss": 0.0263,
      "positive_teacher_pairs": 25846.2,
      "query_h_norm_ratio": 0.4227897137403488,
      "relation_loss": 0.0027664586901664733,
      "selected_pairs": 25846.2,
      "step": 18410,
      "student_distance_mean": 24.071712303161622,
      "teacher_distance_mean": 0.8588769912719727,
      "valid_action_samples": 227.8,
      "weighted_auxiliary_loss": 0.0027664586901664733,
      "weighted_relation_loss": 0.0027664586901664733
    },
    {
      "action_query_batch_std": 0.36164352893829343,
      "action_query_norm": 41.635607147216795,
      "action_relation_loss": 0.002882314007729292,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.231057062675397,
      "flow_matching_loss": 0.025124416872859,
      "grad_norm": 0.08165622502565384,
      "h_stock_valid_norm": 98.36162414550782,
      "invalid_tail_samples": 23.2,
      "learning_rate": 3.550093030553284e-05,
      "loss": 0.028,
      "positive_teacher_pairs": 26986.4,
      "query_h_norm_ratio": 0.4232939392328262,
      "relation_loss": 0.002882314007729292,
      "selected_pairs": 26986.4,
      "step": 18420,
      "student_distance_mean": 24.013101005554198,
      "teacher_distance_mean": 0.8565915644168853,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.002882314007729292,
      "weighted_relation_loss": 0.002882314007729292
    },
    {
      "action_query_batch_std": 0.36510559022426603,
      "action_query_norm": 41.667772674560545,
      "action_relation_loss": 0.002765445318073034,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.24041159962582,
      "flow_matching_loss": 0.02428321596235037,
      "grad_norm": 0.08317086845636368,
      "h_stock_valid_norm": 98.95100173950195,
      "invalid_tail_samples": 25.4,
      "learning_rate": 3.544819165858642e-05,
      "loss": 0.027,
      "positive_teacher_pairs": 26486.2,
      "query_h_norm_ratio": 0.4210997223854065,
      "relation_loss": 0.002765445318073034,
      "selected_pairs": 26486.2,
      "step": 18430,
      "student_distance_mean": 24.155502319335938,
      "teacher_distance_mean": 0.8600064158439636,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.002765445318073034,
      "weighted_relation_loss": 0.002765445318073034
    },
    {
      "action_query_batch_std": 0.36692692041397096,
      "action_query_norm": 41.6858268737793,
      "action_relation_loss": 0.0027458623051643373,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.24976613657624,
      "flow_matching_loss": 0.024834264442324637,
      "grad_norm": 0.08955556899309158,
      "h_stock_valid_norm": 99.0350227355957,
      "invalid_tail_samples": 22.7,
      "learning_rate": 3.539547069345926e-05,
      "loss": 0.0276,
      "positive_teacher_pairs": 27104.0,
      "query_h_norm_ratio": 0.42092489898204805,
      "relation_loss": 0.0027458623051643373,
      "selected_pairs": 27104.0,
      "step": 18440,
      "student_distance_mean": 24.2669620513916,
      "teacher_distance_mean": 0.8662505090236664,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0027458623051643373,
      "weighted_relation_loss": 0.0027458623051643373
    },
    {
      "action_query_batch_std": 0.35844548642635343,
      "action_query_norm": 41.63329162597656,
      "action_relation_loss": 0.0029074735241010783,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.25912067352666,
      "flow_matching_loss": 0.024262231215834616,
      "grad_norm": 0.0797182247042656,
      "h_stock_valid_norm": 98.62799301147462,
      "invalid_tail_samples": 20.4,
      "learning_rate": 3.5342767474212344e-05,
      "loss": 0.0272,
      "positive_teacher_pairs": 27639.2,
      "query_h_norm_ratio": 0.4221262723207474,
      "relation_loss": 0.0029074735241010783,
      "selected_pairs": 27639.2,
      "step": 18450,
      "student_distance_mean": 23.73101234436035,
      "teacher_distance_mean": 0.8445306777954101,
      "valid_action_samples": 235.6,
      "weighted_auxiliary_loss": 0.0029074735241010783,
      "weighted_relation_loss": 0.0029074735241010783
    },
    {
      "action_query_batch_std": 0.3604997217655182,
      "action_query_norm": 41.639194107055665,
      "action_relation_loss": 0.0028171985410153865,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.26847521047708,
      "flow_matching_loss": 0.023791921325027943,
      "grad_norm": 0.09455300122499466,
      "h_stock_valid_norm": 98.2068717956543,
      "invalid_tail_samples": 22.1,
      "learning_rate": 3.529008206488502e-05,
      "loss": 0.0266,
      "positive_teacher_pairs": 27262.2,
      "query_h_norm_ratio": 0.4239968091249466,
      "relation_loss": 0.0028171985410153865,
      "selected_pairs": 27262.2,
      "step": 18460,
      "student_distance_mean": 23.945039176940917,
      "teacher_distance_mean": 0.8555312275886535,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0028171985410153865,
      "weighted_relation_loss": 0.0028171985410153865
    },
    {
      "action_query_batch_std": 0.35833426713943484,
      "action_query_norm": 41.62038230895996,
      "action_relation_loss": 0.002878964738920331,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.277829747427504,
      "flow_matching_loss": 0.023578140884637833,
      "grad_norm": 0.09516565501689911,
      "h_stock_valid_norm": 98.1826286315918,
      "invalid_tail_samples": 24.1,
      "learning_rate": 3.5237414529495055e-05,
      "loss": 0.0265,
      "positive_teacher_pairs": 26790.1,
      "query_h_norm_ratio": 0.4239092439413071,
      "relation_loss": 0.002878964738920331,
      "selected_pairs": 26790.1,
      "step": 18470,
      "student_distance_mean": 23.762664604187012,
      "teacher_distance_mean": 0.8484342038631439,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.002878964738920331,
      "weighted_relation_loss": 0.002878964738920331
    },
    {
      "action_query_batch_std": 0.3611570566892624,
      "action_query_norm": 41.62482948303223,
      "action_relation_loss": 0.002775999438017607,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.287184284377922,
      "flow_matching_loss": 0.021956039778888225,
      "grad_norm": 0.09433170408010483,
      "h_stock_valid_norm": 98.52874755859375,
      "invalid_tail_samples": 23.4,
      "learning_rate": 3.5184764932038454e-05,
      "loss": 0.0247,
      "positive_teacher_pairs": 26943.2,
      "query_h_norm_ratio": 0.42246816456317904,
      "relation_loss": 0.002775999438017607,
      "selected_pairs": 26943.2,
      "step": 18480,
      "student_distance_mean": 23.941471099853516,
      "teacher_distance_mean": 0.8564285695552826,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002775999438017607,
      "weighted_relation_loss": 0.002775999438017607
    },
    {
      "action_query_batch_std": 0.3608788222074509,
      "action_query_norm": 41.63911323547363,
      "action_relation_loss": 0.002829153183847666,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.296538821328344,
      "flow_matching_loss": 0.025318429991602897,
      "grad_norm": 0.07437030971050262,
      "h_stock_valid_norm": 98.55301971435547,
      "invalid_tail_samples": 23.9,
      "learning_rate": 3.513213333648945e-05,
      "loss": 0.0281,
      "positive_teacher_pairs": 26824.5,
      "query_h_norm_ratio": 0.4225051313638687,
      "relation_loss": 0.002829153183847666,
      "selected_pairs": 26824.5,
      "step": 18490,
      "student_distance_mean": 23.901860427856445,
      "teacher_distance_mean": 0.85255047082901,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.002829153183847666,
      "weighted_relation_loss": 0.002829153183847666
    },
    {
      "action_query_batch_std": 0.3669454246759415,
      "action_query_norm": 41.67475776672363,
      "action_relation_loss": 0.002831372735090554,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.305893358278766,
      "flow_matching_loss": 0.023530954495072365,
      "grad_norm": 0.1008560061454773,
      "h_stock_valid_norm": 99.1867904663086,
      "invalid_tail_samples": 23.3,
      "learning_rate": 3.5079519806800374e-05,
      "loss": 0.0264,
      "positive_teacher_pairs": 26961.5,
      "query_h_norm_ratio": 0.4201669543981552,
      "relation_loss": 0.002831372735090554,
      "selected_pairs": 26961.5,
      "step": 18500,
      "student_distance_mean": 24.266297149658204,
      "teacher_distance_mean": 0.8605756461620331,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.002831372735090554,
      "weighted_relation_loss": 0.002831372735090554
    },
    {
      "action_query_batch_std": 0.36296802163124087,
      "action_query_norm": 41.65305061340332,
      "action_relation_loss": 0.002848120266571641,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.315247895229184,
      "flow_matching_loss": 0.02543763369321823,
      "grad_norm": 0.09820523858070374,
      "h_stock_valid_norm": 99.1900016784668,
      "invalid_tail_samples": 24.4,
      "learning_rate": 3.502692440690165e-05,
      "loss": 0.0283,
      "positive_teacher_pairs": 26711.1,
      "query_h_norm_ratio": 0.4199352920055389,
      "relation_loss": 0.002848120266571641,
      "selected_pairs": 26711.1,
      "step": 18510,
      "student_distance_mean": 23.99757137298584,
      "teacher_distance_mean": 0.8524319231510162,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.002848120266571641,
      "weighted_relation_loss": 0.002848120266571641
    },
    {
      "action_query_batch_std": 0.3633705168962479,
      "action_query_norm": 41.643355178833005,
      "action_relation_loss": 0.002822555531747639,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.324602432179606,
      "flow_matching_loss": 0.02480455543845892,
      "grad_norm": 0.10668966919183731,
      "h_stock_valid_norm": 98.44186553955078,
      "invalid_tail_samples": 22.6,
      "learning_rate": 3.497434720070165e-05,
      "loss": 0.0276,
      "positive_teacher_pairs": 27124.8,
      "query_h_norm_ratio": 0.42302840650081636,
      "relation_loss": 0.002822555531747639,
      "selected_pairs": 27124.8,
      "step": 18520,
      "student_distance_mean": 24.06101016998291,
      "teacher_distance_mean": 0.8591370105743408,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.002822555531747639,
      "weighted_relation_loss": 0.002822555531747639
    },
    {
      "action_query_batch_std": 0.36173452734947203,
      "action_query_norm": 41.61268577575684,
      "action_relation_loss": 0.0027660041116178034,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.33395696913003,
      "flow_matching_loss": 0.02425333671271801,
      "grad_norm": 0.09740934520959854,
      "h_stock_valid_norm": 98.16307754516602,
      "invalid_tail_samples": 24.6,
      "learning_rate": 3.4921788252086616e-05,
      "loss": 0.027,
      "positive_teacher_pairs": 26673.5,
      "query_h_norm_ratio": 0.42391647696495055,
      "relation_loss": 0.0027660041116178034,
      "selected_pairs": 26673.5,
      "step": 18530,
      "student_distance_mean": 23.91819496154785,
      "teacher_distance_mean": 0.8537682354450226,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0027660041116178034,
      "weighted_relation_loss": 0.0027660041116178034
    },
    {
      "action_query_batch_std": 0.35927637815475466,
      "action_query_norm": 41.616434478759764,
      "action_relation_loss": 0.002887378679588437,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.34331150608045,
      "flow_matching_loss": 0.024470832757651807,
      "grad_norm": 0.10318230092525482,
      "h_stock_valid_norm": 97.93116836547851,
      "invalid_tail_samples": 21.4,
      "learning_rate": 3.486924762492065e-05,
      "loss": 0.0274,
      "positive_teacher_pairs": 27406.1,
      "query_h_norm_ratio": 0.42496044635772706,
      "relation_loss": 0.002887378679588437,
      "selected_pairs": 27406.1,
      "step": 18540,
      "student_distance_mean": 23.800197029113768,
      "teacher_distance_mean": 0.8477123320102692,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.002887378679588437,
      "weighted_relation_loss": 0.002887378679588437
    },
    {
      "action_query_batch_std": 0.36372252702713015,
      "action_query_norm": 41.642604446411134,
      "action_relation_loss": 0.002774538565427065,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.35266604303087,
      "flow_matching_loss": 0.023960239440202712,
      "grad_norm": 0.09664306044578552,
      "h_stock_valid_norm": 98.23893356323242,
      "invalid_tail_samples": 23.1,
      "learning_rate": 3.4816725383045534e-05,
      "loss": 0.0267,
      "positive_teacher_pairs": 27017.8,
      "query_h_norm_ratio": 0.42389459908008575,
      "relation_loss": 0.002774538565427065,
      "selected_pairs": 27017.8,
      "step": 18550,
      "student_distance_mean": 24.10980091094971,
      "teacher_distance_mean": 0.8618335843086242,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002774538565427065,
      "weighted_relation_loss": 0.002774538565427065
    },
    {
      "action_query_batch_std": 0.36055498719215395,
      "action_query_norm": 41.61036148071289,
      "action_relation_loss": 0.002861997974105179,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.36202057998129,
      "flow_matching_loss": 0.024960286729037762,
      "grad_norm": 0.10856472700834274,
      "h_stock_valid_norm": 98.25688705444335,
      "invalid_tail_samples": 23.9,
      "learning_rate": 3.476422159028079e-05,
      "loss": 0.0278,
      "positive_teacher_pairs": 26832.8,
      "query_h_norm_ratio": 0.4234928458929062,
      "relation_loss": 0.002861997974105179,
      "selected_pairs": 26832.8,
      "step": 18560,
      "student_distance_mean": 23.86324348449707,
      "teacher_distance_mean": 0.8548852801322937,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.002861997974105179,
      "weighted_relation_loss": 0.002861997974105179
    },
    {
      "action_query_batch_std": 0.3642189711332321,
      "action_query_norm": 41.637538528442384,
      "action_relation_loss": 0.0028490638360381126,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.371375116931713,
      "flow_matching_loss": 0.02328752875328064,
      "grad_norm": 0.1096220389008522,
      "h_stock_valid_norm": 98.14099502563477,
      "invalid_tail_samples": 22.1,
      "learning_rate": 3.471173631042345e-05,
      "loss": 0.0261,
      "positive_teacher_pairs": 27248.1,
      "query_h_norm_ratio": 0.424266853928566,
      "relation_loss": 0.0028490638360381126,
      "selected_pairs": 27248.1,
      "step": 18570,
      "student_distance_mean": 24.061820411682127,
      "teacher_distance_mean": 0.8571873545646668,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0028490638360381126,
      "weighted_relation_loss": 0.0028490638360381126
    },
    {
      "action_query_batch_std": 0.36060247421264646,
      "action_query_norm": 41.60705757141113,
      "action_relation_loss": 0.002938175364397466,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.38072965388213,
      "flow_matching_loss": 0.0241843581199646,
      "grad_norm": 0.127560093998909,
      "h_stock_valid_norm": 98.0178436279297,
      "invalid_tail_samples": 23.4,
      "learning_rate": 3.465926960724808e-05,
      "loss": 0.0271,
      "positive_teacher_pairs": 26944.1,
      "query_h_norm_ratio": 0.42448765933513644,
      "relation_loss": 0.002938175364397466,
      "selected_pairs": 26944.1,
      "step": 18580,
      "student_distance_mean": 23.843624687194826,
      "teacher_distance_mean": 0.8488309144973755,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002938175364397466,
      "weighted_relation_loss": 0.002938175364397466
    },
    {
      "action_query_batch_std": 0.3616303414106369,
      "action_query_norm": 41.61185417175293,
      "action_relation_loss": 0.002841659029945731,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.390084190832553,
      "flow_matching_loss": 0.022663427516818047,
      "grad_norm": 0.11365131288766861,
      "h_stock_valid_norm": 98.08361434936523,
      "invalid_tail_samples": 23.4,
      "learning_rate": 3.460682154450666e-05,
      "loss": 0.0255,
      "positive_teacher_pairs": 26937.2,
      "query_h_norm_ratio": 0.42424941062927246,
      "relation_loss": 0.002841659029945731,
      "selected_pairs": 26937.2,
      "step": 18590,
      "student_distance_mean": 23.90051403045654,
      "teacher_distance_mean": 0.8517044067382813,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002841659029945731,
      "weighted_relation_loss": 0.002841659029945731
    },
    {
      "action_query_batch_std": 0.3629885971546173,
      "action_query_norm": 41.600585556030275,
      "action_relation_loss": 0.0027940648607909678,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.399438727782975,
      "flow_matching_loss": 0.024764124676585196,
      "grad_norm": 0.12513862550258636,
      "h_stock_valid_norm": 98.04924087524414,
      "invalid_tail_samples": 24.9,
      "learning_rate": 3.4554392185928564e-05,
      "loss": 0.0276,
      "positive_teacher_pairs": 26597.6,
      "query_h_norm_ratio": 0.4242845743894577,
      "relation_loss": 0.0027940648607909678,
      "selected_pairs": 26597.6,
      "step": 18600,
      "student_distance_mean": 23.969414901733398,
      "teacher_distance_mean": 0.8560985207557679,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0027940648607909678,
      "weighted_relation_loss": 0.0027940648607909678
    },
    {
      "action_query_batch_std": 0.3581430524587631,
      "action_query_norm": 41.56569595336914,
      "action_relation_loss": 0.002832869067788124,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.408793264733397,
      "flow_matching_loss": 0.025299837067723274,
      "grad_norm": 0.13399386405944824,
      "h_stock_valid_norm": 98.21967010498047,
      "invalid_tail_samples": 24.8,
      "learning_rate": 3.450198159522037e-05,
      "loss": 0.0281,
      "positive_teacher_pairs": 26620.9,
      "query_h_norm_ratio": 0.42319696843624116,
      "relation_loss": 0.002832869067788124,
      "selected_pairs": 26620.9,
      "step": 18610,
      "student_distance_mean": 23.599951934814452,
      "teacher_distance_mean": 0.8441881775856018,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.002832869067788124,
      "weighted_relation_loss": 0.002832869067788124
    },
    {
      "action_query_batch_std": 0.36112105250358584,
      "action_query_norm": 41.58848609924316,
      "action_relation_loss": 0.002838453371077776,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.418147801683816,
      "flow_matching_loss": 0.022957101836800574,
      "grad_norm": 0.09916279464960098,
      "h_stock_valid_norm": 98.14603042602539,
      "invalid_tail_samples": 22.0,
      "learning_rate": 3.444958983606592e-05,
      "loss": 0.0258,
      "positive_teacher_pairs": 27264.3,
      "query_h_norm_ratio": 0.4237449079751968,
      "relation_loss": 0.002838453371077776,
      "selected_pairs": 27264.3,
      "step": 18620,
      "student_distance_mean": 23.814281463623047,
      "teacher_distance_mean": 0.8499880075454712,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.002838453371077776,
      "weighted_relation_loss": 0.002838453371077776
    },
    {
      "action_query_batch_std": 0.3608220398426056,
      "action_query_norm": 41.59285621643066,
      "action_relation_loss": 0.0026986805256456136,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.427502338634238,
      "flow_matching_loss": 0.024164142645895482,
      "grad_norm": 0.14986006915569305,
      "h_stock_valid_norm": 98.26900863647461,
      "invalid_tail_samples": 25.4,
      "learning_rate": 3.439721697212612e-05,
      "loss": 0.0269,
      "positive_teacher_pairs": 26484.2,
      "query_h_norm_ratio": 0.4232565999031067,
      "relation_loss": 0.0026986805256456136,
      "selected_pairs": 26484.2,
      "step": 18630,
      "student_distance_mean": 23.905772590637206,
      "teacher_distance_mean": 0.8525522589683533,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0026986805256456136,
      "weighted_relation_loss": 0.0026986805256456136
    },
    {
      "action_query_batch_std": 0.3589614540338516,
      "action_query_norm": 41.57743377685547,
      "action_relation_loss": 0.002901098178699613,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.43685687558466,
      "flow_matching_loss": 0.022947430983185767,
      "grad_norm": 0.1007683053612709,
      "h_stock_valid_norm": 98.64135818481445,
      "invalid_tail_samples": 23.1,
      "learning_rate": 3.4344863067038954e-05,
      "loss": 0.0258,
      "positive_teacher_pairs": 27013.4,
      "query_h_norm_ratio": 0.42150221168994906,
      "relation_loss": 0.002901098178699613,
      "selected_pairs": 27013.4,
      "step": 18640,
      "student_distance_mean": 23.712784385681154,
      "teacher_distance_mean": 0.8482330143451691,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002901098178699613,
      "weighted_relation_loss": 0.002901098178699613
    },
    {
      "action_query_batch_std": 0.35947910249233245,
      "action_query_norm": 41.58600959777832,
      "action_relation_loss": 0.0029167930362746118,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.446211412535078,
      "flow_matching_loss": 0.0242327980697155,
      "grad_norm": 0.10410035401582718,
      "h_stock_valid_norm": 98.60058059692383,
      "invalid_tail_samples": 22.5,
      "learning_rate": 3.429252818441935e-05,
      "loss": 0.0271,
      "positive_teacher_pairs": 27164.1,
      "query_h_norm_ratio": 0.42176732122898103,
      "relation_loss": 0.0029167930362746118,
      "selected_pairs": 27164.1,
      "step": 18650,
      "student_distance_mean": 23.788075637817382,
      "teacher_distance_mean": 0.8537384569644928,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0029167930362746118,
      "weighted_relation_loss": 0.0029167930362746118
    },
    {
      "action_query_batch_std": 0.35782996714115145,
      "action_query_norm": 41.58056411743164,
      "action_relation_loss": 0.0028418405214324593,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.4555659494855,
      "flow_matching_loss": 0.02464668620377779,
      "grad_norm": 0.09223771840333939,
      "h_stock_valid_norm": 98.68634338378907,
      "invalid_tail_samples": 23.8,
      "learning_rate": 3.4240212387859094e-05,
      "loss": 0.0275,
      "positive_teacher_pairs": 26848.0,
      "query_h_norm_ratio": 0.42134317457675935,
      "relation_loss": 0.0028418405214324593,
      "selected_pairs": 26848.0,
      "step": 18660,
      "student_distance_mean": 23.688100624084473,
      "teacher_distance_mean": 0.8476728141307831,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0028418405214324593,
      "weighted_relation_loss": 0.0028418405214324593
    },
    {
      "action_query_batch_std": 0.3659472346305847,
      "action_query_norm": 41.62601776123047,
      "action_relation_loss": 0.002728116558864713,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.464920486435922,
      "flow_matching_loss": 0.025280851125717162,
      "grad_norm": 0.08615841716527939,
      "h_stock_valid_norm": 98.29812774658203,
      "invalid_tail_samples": 21.6,
      "learning_rate": 3.418791574092686e-05,
      "loss": 0.028,
      "positive_teacher_pairs": 27369.4,
      "query_h_norm_ratio": 0.42347000241279603,
      "relation_loss": 0.002728116558864713,
      "selected_pairs": 27369.4,
      "step": 18670,
      "student_distance_mean": 24.093918228149413,
      "teacher_distance_mean": 0.8611822426319122,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.002728116558864713,
      "weighted_relation_loss": 0.002728116558864713
    },
    {
      "action_query_batch_std": 0.3614562153816223,
      "action_query_norm": 41.605272674560545,
      "action_relation_loss": 0.00274677702691406,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.474275023386344,
      "flow_matching_loss": 0.023336385563015937,
      "grad_norm": 0.10779755562543869,
      "h_stock_valid_norm": 97.7720115661621,
      "invalid_tail_samples": 23.2,
      "learning_rate": 3.413563830716796e-05,
      "loss": 0.0261,
      "positive_teacher_pairs": 26993.8,
      "query_h_norm_ratio": 0.42553913295269014,
      "relation_loss": 0.00274677702691406,
      "selected_pairs": 26993.8,
      "step": 18680,
      "student_distance_mean": 23.901690673828124,
      "teacher_distance_mean": 0.8535219192504883,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.00274677702691406,
      "weighted_relation_loss": 0.00274677702691406
    },
    {
      "action_query_batch_std": 0.35869724452495577,
      "action_query_norm": 41.549415588378906,
      "action_relation_loss": 0.002940490213222802,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.483629560336762,
      "flow_matching_loss": 0.02512795738875866,
      "grad_norm": 0.1204582005739212,
      "h_stock_valid_norm": 97.79364700317383,
      "invalid_tail_samples": 25.7,
      "learning_rate": 3.408338015010445e-05,
      "loss": 0.0281,
      "positive_teacher_pairs": 26410.1,
      "query_h_norm_ratio": 0.42487089037895204,
      "relation_loss": 0.002940490213222802,
      "selected_pairs": 26410.1,
      "step": 18690,
      "student_distance_mean": 23.674489784240723,
      "teacher_distance_mean": 0.8454993963241577,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.002940490213222802,
      "weighted_relation_loss": 0.002940490213222802
    },
    {
      "action_query_batch_std": 0.36287951469421387,
      "action_query_norm": 41.604713439941406,
      "action_relation_loss": 0.0027508421102538706,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.492984097287184,
      "flow_matching_loss": 0.024662415124475957,
      "grad_norm": 0.09169971197843552,
      "h_stock_valid_norm": 98.48234634399414,
      "invalid_tail_samples": 21.6,
      "learning_rate": 3.40311413332349e-05,
      "loss": 0.0274,
      "positive_teacher_pairs": 27367.3,
      "query_h_norm_ratio": 0.4224604606628418,
      "relation_loss": 0.0027508421102538706,
      "selected_pairs": 27367.3,
      "step": 18700,
      "student_distance_mean": 24.08653030395508,
      "teacher_distance_mean": 0.8630158305168152,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.0027508421102538706,
      "weighted_relation_loss": 0.0027508421102538706
    },
    {
      "action_query_batch_std": 0.3586602836847305,
      "action_query_norm": 41.564535522460936,
      "action_relation_loss": 0.0028812123695388437,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.502338634237606,
      "flow_matching_loss": 0.023108762875199317,
      "grad_norm": 0.08950021862983704,
      "h_stock_valid_norm": 98.35720825195312,
      "invalid_tail_samples": 24.7,
      "learning_rate": 3.3978921920034365e-05,
      "loss": 0.026,
      "positive_teacher_pairs": 26653.3,
      "query_h_norm_ratio": 0.422590509057045,
      "relation_loss": 0.0028812123695388437,
      "selected_pairs": 26653.3,
      "step": 18710,
      "student_distance_mean": 23.697336196899414,
      "teacher_distance_mean": 0.847247987985611,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0028812123695388437,
      "weighted_relation_loss": 0.0028812123695388437
    },
    {
      "action_query_batch_std": 0.3623656094074249,
      "action_query_norm": 41.584794998168945,
      "action_relation_loss": 0.0028200376778841018,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.511693171188025,
      "flow_matching_loss": 0.022454961203038692,
      "grad_norm": 0.10146541893482208,
      "h_stock_valid_norm": 98.1075439453125,
      "invalid_tail_samples": 23.8,
      "learning_rate": 3.392672197395441e-05,
      "loss": 0.0253,
      "positive_teacher_pairs": 26851.6,
      "query_h_norm_ratio": 0.4238730549812317,
      "relation_loss": 0.0028200376778841018,
      "selected_pairs": 26851.6,
      "step": 18720,
      "student_distance_mean": 23.917364883422852,
      "teacher_distance_mean": 0.8552358031272889,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0028200376778841018,
      "weighted_relation_loss": 0.0028200376778841018
    },
    {
      "action_query_batch_std": 0.3574276179075241,
      "action_query_norm": 41.54058265686035,
      "action_relation_loss": 0.002848169533535838,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.521047708138447,
      "flow_matching_loss": 0.0227958669885993,
      "grad_norm": 0.1589346081018448,
      "h_stock_valid_norm": 97.89476699829102,
      "invalid_tail_samples": 24.7,
      "learning_rate": 3.387454155842287e-05,
      "loss": 0.0256,
      "positive_teacher_pairs": 26643.4,
      "query_h_norm_ratio": 0.42434251606464385,
      "relation_loss": 0.002848169533535838,
      "selected_pairs": 26643.4,
      "step": 18730,
      "student_distance_mean": 23.72008361816406,
      "teacher_distance_mean": 0.8541475236415863,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.002848169533535838,
      "weighted_relation_loss": 0.002848169533535838
    },
    {
      "action_query_batch_std": 0.35850814580917356,
      "action_query_norm": 41.54294357299805,
      "action_relation_loss": 0.002687619859352708,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.53040224508887,
      "flow_matching_loss": 0.023535108752548696,
      "grad_norm": 0.11354735493659973,
      "h_stock_valid_norm": 98.02016525268554,
      "invalid_tail_samples": 23.3,
      "learning_rate": 3.382238073684386e-05,
      "loss": 0.0262,
      "positive_teacher_pairs": 26969.5,
      "query_h_norm_ratio": 0.4238245815038681,
      "relation_loss": 0.002687619859352708,
      "selected_pairs": 26969.5,
      "step": 18740,
      "student_distance_mean": 23.71334590911865,
      "teacher_distance_mean": 0.8529600441455841,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.002687619859352708,
      "weighted_relation_loss": 0.002687619859352708
    },
    {
      "action_query_batch_std": 0.3621318221092224,
      "action_query_norm": 41.57981910705566,
      "action_relation_loss": 0.0027621957240626218,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.53975678203929,
      "flow_matching_loss": 0.022880923189222812,
      "grad_norm": 0.10485169291496277,
      "h_stock_valid_norm": 98.0818260192871,
      "invalid_tail_samples": 22.8,
      "learning_rate": 3.377023957259771e-05,
      "loss": 0.0256,
      "positive_teacher_pairs": 27088.0,
      "query_h_norm_ratio": 0.423934206366539,
      "relation_loss": 0.0027621957240626218,
      "selected_pairs": 27088.0,
      "step": 18750,
      "student_distance_mean": 24.038321495056152,
      "teacher_distance_mean": 0.8624837458133697,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0027621957240626218,
      "weighted_relation_loss": 0.0027621957240626218
    },
    {
      "action_query_batch_std": 0.3642215430736542,
      "action_query_norm": 41.59193153381348,
      "action_relation_loss": 0.0028109503677114845,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.54911131898971,
      "flow_matching_loss": 0.024035608768463133,
      "grad_norm": 0.10714223980903625,
      "h_stock_valid_norm": 97.7531120300293,
      "invalid_tail_samples": 22.4,
      "learning_rate": 3.3718118129040835e-05,
      "loss": 0.0268,
      "positive_teacher_pairs": 27180.2,
      "query_h_norm_ratio": 0.4254811525344849,
      "relation_loss": 0.0028109503677114845,
      "selected_pairs": 27180.2,
      "step": 18760,
      "student_distance_mean": 24.142616081237794,
      "teacher_distance_mean": 0.8644888520240783,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.0028109503677114845,
      "weighted_relation_loss": 0.0028109503677114845
    },
    {
      "action_query_batch_std": 0.36516263484954836,
      "action_query_norm": 41.58696670532227,
      "action_relation_loss": 0.002811044082045555,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.55846585594013,
      "flow_matching_loss": 0.022263696789741515,
      "grad_norm": 0.11166641116142273,
      "h_stock_valid_norm": 97.86922988891601,
      "invalid_tail_samples": 23.4,
      "learning_rate": 3.3666016469505724e-05,
      "loss": 0.0251,
      "positive_teacher_pairs": 26943.6,
      "query_h_norm_ratio": 0.42492546439170836,
      "relation_loss": 0.002811044082045555,
      "selected_pairs": 26943.6,
      "step": 18770,
      "student_distance_mean": 23.993774604797363,
      "teacher_distance_mean": 0.8584478259086609,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002811044082045555,
      "weighted_relation_loss": 0.002811044082045555
    },
    {
      "action_query_batch_std": 0.3610899567604065,
      "action_query_norm": 41.56824264526367,
      "action_relation_loss": 0.002824081317521632,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.567820392890553,
      "flow_matching_loss": 0.023259949497878553,
      "grad_norm": 0.08908072859048843,
      "h_stock_valid_norm": 97.5889503479004,
      "invalid_tail_samples": 23.7,
      "learning_rate": 3.361393465730079e-05,
      "loss": 0.0261,
      "positive_teacher_pairs": 26875.8,
      "query_h_norm_ratio": 0.42595831155776975,
      "relation_loss": 0.002824081317521632,
      "selected_pairs": 26875.8,
      "step": 18780,
      "student_distance_mean": 23.795072174072267,
      "teacher_distance_mean": 0.8527871072292328,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.002824081317521632,
      "weighted_relation_loss": 0.002824081317521632
    },
    {
      "action_query_batch_std": 0.36403191089630127,
      "action_query_norm": 41.567373275756836,
      "action_relation_loss": 0.0028736092615872623,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.57717492984097,
      "flow_matching_loss": 0.025668614357709885,
      "grad_norm": 0.11198552697896957,
      "h_stock_valid_norm": 97.82281951904297,
      "invalid_tail_samples": 23.5,
      "learning_rate": 3.3561872755710366e-05,
      "loss": 0.0285,
      "positive_teacher_pairs": 26922.0,
      "query_h_norm_ratio": 0.4249305218458176,
      "relation_loss": 0.0028736092615872623,
      "selected_pairs": 26922.0,
      "step": 18790,
      "student_distance_mean": 24.049661254882814,
      "teacher_distance_mean": 0.8570180416107178,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0028736092615872623,
      "weighted_relation_loss": 0.0028736092615872623
    },
    {
      "action_query_batch_std": 0.36192681193351744,
      "action_query_norm": 41.56192588806152,
      "action_relation_loss": 0.0027842733543366194,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.586529466791394,
      "flow_matching_loss": 0.02386308629065752,
      "grad_norm": 0.09120997786521912,
      "h_stock_valid_norm": 98.14227600097657,
      "invalid_tail_samples": 24.1,
      "learning_rate": 3.350983082799456e-05,
      "loss": 0.0266,
      "positive_teacher_pairs": 26780.2,
      "query_h_norm_ratio": 0.4234915554523468,
      "relation_loss": 0.0027842733543366194,
      "selected_pairs": 26780.2,
      "step": 18800,
      "student_distance_mean": 23.9873327255249,
      "teacher_distance_mean": 0.8574211001396179,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0027842733543366194,
      "weighted_relation_loss": 0.0027842733543366194
    },
    {
      "action_query_batch_std": 0.35703770220279696,
      "action_query_norm": 41.52278518676758,
      "action_relation_loss": 0.0028549466980621217,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.595884003741816,
      "flow_matching_loss": 0.022050132043659687,
      "grad_norm": 0.10431592911481857,
      "h_stock_valid_norm": 97.95033340454101,
      "invalid_tail_samples": 24.2,
      "learning_rate": 3.34578089373892e-05,
      "loss": 0.0249,
      "positive_teacher_pairs": 26756.7,
      "query_h_norm_ratio": 0.4239173620939255,
      "relation_loss": 0.0028549466980621217,
      "selected_pairs": 26756.7,
      "step": 18810,
      "student_distance_mean": 23.591683387756348,
      "teacher_distance_mean": 0.848661071062088,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0028549466980621217,
      "weighted_relation_loss": 0.0028549466980621217
    },
    {
      "action_query_batch_std": 0.3622562050819397,
      "action_query_norm": 41.564910125732425,
      "action_relation_loss": 0.0027363076340407135,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.605238540692234,
      "flow_matching_loss": 0.021788351982831956,
      "grad_norm": 0.12547966837882996,
      "h_stock_valid_norm": 98.03889541625976,
      "invalid_tail_samples": 24.0,
      "learning_rate": 3.340580714710581e-05,
      "loss": 0.0245,
      "positive_teacher_pairs": 26800.5,
      "query_h_norm_ratio": 0.4239652931690216,
      "relation_loss": 0.0027363076340407135,
      "selected_pairs": 26800.5,
      "step": 18820,
      "student_distance_mean": 23.91604881286621,
      "teacher_distance_mean": 0.8630214631557465,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0027363076340407135,
      "weighted_relation_loss": 0.0027363076340407135
    },
    {
      "action_query_batch_std": 0.362293952703476,
      "action_query_norm": 41.58187599182129,
      "action_relation_loss": 0.002822795184329152,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.614593077642656,
      "flow_matching_loss": 0.023399336077272893,
      "grad_norm": 0.13126789033412933,
      "h_stock_valid_norm": 98.08432235717774,
      "invalid_tail_samples": 20.5,
      "learning_rate": 3.3353825520331466e-05,
      "loss": 0.0262,
      "positive_teacher_pairs": 27627.5,
      "query_h_norm_ratio": 0.4239425450563431,
      "relation_loss": 0.002822795184329152,
      "selected_pairs": 27627.5,
      "step": 18830,
      "student_distance_mean": 23.933966064453124,
      "teacher_distance_mean": 0.8551330327987671,
      "valid_action_samples": 235.5,
      "weighted_auxiliary_loss": 0.002822795184329152,
      "weighted_relation_loss": 0.002822795184329152
    },
    {
      "action_query_batch_std": 0.3566432327032089,
      "action_query_norm": 41.532597351074216,
      "action_relation_loss": 0.002809474105015397,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.623947614593078,
      "flow_matching_loss": 0.02296545747667551,
      "grad_norm": 0.08761792629957199,
      "h_stock_valid_norm": 97.99875259399414,
      "invalid_tail_samples": 26.8,
      "learning_rate": 3.330186412022876e-05,
      "loss": 0.0258,
      "positive_teacher_pairs": 26158.0,
      "query_h_norm_ratio": 0.42381306886672976,
      "relation_loss": 0.002809474105015397,
      "selected_pairs": 26158.0,
      "step": 18840,
      "student_distance_mean": 23.725415420532226,
      "teacher_distance_mean": 0.8504809737205505,
      "valid_action_samples": 229.2,
      "weighted_auxiliary_loss": 0.002809474105015397,
      "weighted_relation_loss": 0.002809474105015397
    },
    {
      "action_query_batch_std": 0.3584039151668549,
      "action_query_norm": 41.54461402893067,
      "action_relation_loss": 0.0028600260149687527,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.6333021515435,
      "flow_matching_loss": 0.021621453762054443,
      "grad_norm": 0.13334921002388,
      "h_stock_valid_norm": 98.01619720458984,
      "invalid_tail_samples": 24.8,
      "learning_rate": 3.324992300993568e-05,
      "loss": 0.0245,
      "positive_teacher_pairs": 26622.7,
      "query_h_norm_ratio": 0.42385609447956085,
      "relation_loss": 0.0028600260149687527,
      "selected_pairs": 26622.7,
      "step": 18850,
      "student_distance_mean": 23.643452644348145,
      "teacher_distance_mean": 0.8451345086097717,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0028600260149687527,
      "weighted_relation_loss": 0.0028600260149687527
    },
    {
      "action_query_batch_std": 0.35936957597732544,
      "action_query_norm": 41.539282608032224,
      "action_relation_loss": 0.0028713261475786568,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.64265668849392,
      "flow_matching_loss": 0.02323358803987503,
      "grad_norm": 0.1123945489525795,
      "h_stock_valid_norm": 98.43043975830078,
      "invalid_tail_samples": 24.2,
      "learning_rate": 3.319800225256556e-05,
      "loss": 0.0261,
      "positive_teacher_pairs": 26768.6,
      "query_h_norm_ratio": 0.4220190316438675,
      "relation_loss": 0.0028713261475786568,
      "selected_pairs": 26768.6,
      "step": 18860,
      "student_distance_mean": 23.77473373413086,
      "teacher_distance_mean": 0.8524449408054352,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0028713261475786568,
      "weighted_relation_loss": 0.0028713261475786568
    },
    {
      "action_query_batch_std": 0.36041370034217834,
      "action_query_norm": 41.562487030029295,
      "action_relation_loss": 0.0027294097235426307,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.65201122544434,
      "flow_matching_loss": 0.02387097030878067,
      "grad_norm": 0.12577790021896362,
      "h_stock_valid_norm": 98.89133682250977,
      "invalid_tail_samples": 25.4,
      "learning_rate": 3.314610191120702e-05,
      "loss": 0.0266,
      "positive_teacher_pairs": 26481.6,
      "query_h_norm_ratio": 0.42028687298297884,
      "relation_loss": 0.0027294097235426307,
      "selected_pairs": 26481.6,
      "step": 18870,
      "student_distance_mean": 23.925662422180174,
      "teacher_distance_mean": 0.8568965256214142,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0027294097235426307,
      "weighted_relation_loss": 0.0027294097235426307
    },
    {
      "action_query_batch_std": 0.3596271723508835,
      "action_query_norm": 41.539483642578126,
      "action_relation_loss": 0.0027250644750893118,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.661365762394762,
      "flow_matching_loss": 0.02426184844225645,
      "grad_norm": 0.10968275368213654,
      "h_stock_valid_norm": 98.70999221801758,
      "invalid_tail_samples": 23.3,
      "learning_rate": 3.30942220489239e-05,
      "loss": 0.027,
      "positive_teacher_pairs": 26968.2,
      "query_h_norm_ratio": 0.4208300471305847,
      "relation_loss": 0.0027250644750893118,
      "selected_pairs": 26968.2,
      "step": 18880,
      "student_distance_mean": 23.794416427612305,
      "teacher_distance_mean": 0.8545475721359252,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0027250644750893118,
      "weighted_relation_loss": 0.0027250644750893118
    },
    {
      "action_query_batch_std": 0.36023335456848143,
      "action_query_norm": 41.53952522277832,
      "action_relation_loss": 0.002775159990414977,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.67072029934518,
      "flow_matching_loss": 0.024192156083881856,
      "grad_norm": 0.09291891753673553,
      "h_stock_valid_norm": 98.60261764526368,
      "invalid_tail_samples": 26.0,
      "learning_rate": 3.3042362728755086e-05,
      "loss": 0.027,
      "positive_teacher_pairs": 26341.2,
      "query_h_norm_ratio": 0.4212894380092621,
      "relation_loss": 0.002775159990414977,
      "selected_pairs": 26341.2,
      "step": 18890,
      "student_distance_mean": 23.90711917877197,
      "teacher_distance_mean": 0.8570530116558075,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.002775159990414977,
      "weighted_relation_loss": 0.002775159990414977
    },
    {
      "action_query_batch_std": 0.36061089634895327,
      "action_query_norm": 41.55465087890625,
      "action_relation_loss": 0.002718998584896326,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.680074836295603,
      "flow_matching_loss": 0.02388896346092224,
      "grad_norm": 0.08508531749248505,
      "h_stock_valid_norm": 97.99980697631835,
      "invalid_tail_samples": 23.0,
      "learning_rate": 3.299052401371456e-05,
      "loss": 0.0266,
      "positive_teacher_pairs": 27047.0,
      "query_h_norm_ratio": 0.42403298020362856,
      "relation_loss": 0.002718998584896326,
      "selected_pairs": 27047.0,
      "step": 18900,
      "student_distance_mean": 23.904906463623046,
      "teacher_distance_mean": 0.859105521440506,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.002718998584896326,
      "weighted_relation_loss": 0.002718998584896326
    },
    {
      "action_query_batch_std": 0.36239842474460604,
      "action_query_norm": 41.56935615539551,
      "action_relation_loss": 0.002717875875532627,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.689429373246025,
      "flow_matching_loss": 0.022372651472687723,
      "grad_norm": 0.09759984165430069,
      "h_stock_valid_norm": 98.17450408935547,
      "invalid_tail_samples": 21.9,
      "learning_rate": 3.293870596679125e-05,
      "loss": 0.0251,
      "positive_teacher_pairs": 27290.8,
      "query_h_norm_ratio": 0.4234263151884079,
      "relation_loss": 0.002717875875532627,
      "selected_pairs": 27290.8,
      "step": 18910,
      "student_distance_mean": 23.952204513549805,
      "teacher_distance_mean": 0.859463506937027,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.002717875875532627,
      "weighted_relation_loss": 0.002717875875532627
    },
    {
      "action_query_batch_std": 0.3618119090795517,
      "action_query_norm": 41.54955787658692,
      "action_relation_loss": 0.002834203839302063,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.698783910196447,
      "flow_matching_loss": 0.024379417672753333,
      "grad_norm": 0.13816095888614655,
      "h_stock_valid_norm": 98.06003646850586,
      "invalid_tail_samples": 22.5,
      "learning_rate": 3.288690865094895e-05,
      "loss": 0.0272,
      "positive_teacher_pairs": 27160.7,
      "query_h_norm_ratio": 0.4237170577049255,
      "relation_loss": 0.002834203839302063,
      "selected_pairs": 27160.7,
      "step": 18920,
      "student_distance_mean": 23.930883979797365,
      "teacher_distance_mean": 0.8573597371578217,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.002834203839302063,
      "weighted_relation_loss": 0.002834203839302063
    },
    {
      "action_query_batch_std": 0.3573495090007782,
      "action_query_norm": 41.493981170654294,
      "action_relation_loss": 0.0028239669511094688,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.708138447146865,
      "flow_matching_loss": 0.024412672966718674,
      "grad_norm": 0.09400282055139542,
      "h_stock_valid_norm": 97.76975936889649,
      "invalid_tail_samples": 24.9,
      "learning_rate": 3.283513212912632e-05,
      "loss": 0.0272,
      "positive_teacher_pairs": 26593.2,
      "query_h_norm_ratio": 0.42441015839576723,
      "relation_loss": 0.0028239669511094688,
      "selected_pairs": 26593.2,
      "step": 18930,
      "student_distance_mean": 23.624027252197266,
      "teacher_distance_mean": 0.8512759149074555,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0028239669511094688,
      "weighted_relation_loss": 0.0028239669511094688
    },
    {
      "action_query_batch_std": 0.3553832441568375,
      "action_query_norm": 41.50299606323242,
      "action_relation_loss": 0.0028414529981091617,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.717492984097287,
      "flow_matching_loss": 0.02463209703564644,
      "grad_norm": 0.0866384506225586,
      "h_stock_valid_norm": 97.86859741210938,
      "invalid_tail_samples": 24.0,
      "learning_rate": 3.2783376464236684e-05,
      "loss": 0.0275,
      "positive_teacher_pairs": 26803.4,
      "query_h_norm_ratio": 0.4240714073181152,
      "relation_loss": 0.0028414529981091617,
      "selected_pairs": 26803.4,
      "step": 18940,
      "student_distance_mean": 23.601073265075684,
      "teacher_distance_mean": 0.8474095225334167,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0028414529981091617,
      "weighted_relation_loss": 0.0028414529981091617
    },
    {
      "action_query_batch_std": 0.3587613344192505,
      "action_query_norm": 41.50759506225586,
      "action_relation_loss": 0.002850751136429608,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.72684752104771,
      "flow_matching_loss": 0.02250206582248211,
      "grad_norm": 0.08703123033046722,
      "h_stock_valid_norm": 98.00902786254883,
      "invalid_tail_samples": 23.4,
      "learning_rate": 3.273164171916806e-05,
      "loss": 0.0254,
      "positive_teacher_pairs": 26943.3,
      "query_h_norm_ratio": 0.42351458966732025,
      "relation_loss": 0.002850751136429608,
      "selected_pairs": 26943.3,
      "step": 18950,
      "student_distance_mean": 23.734034538269043,
      "teacher_distance_mean": 0.8561357200145722,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002850751136429608,
      "weighted_relation_loss": 0.002850751136429608
    },
    {
      "action_query_batch_std": 0.36078269481658937,
      "action_query_norm": 41.51881866455078,
      "action_relation_loss": 0.0028737506130710245,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.736202057998128,
      "flow_matching_loss": 0.024213880114257336,
      "grad_norm": 0.09268900007009506,
      "h_stock_valid_norm": 97.87969284057617,
      "invalid_tail_samples": 24.8,
      "learning_rate": 3.267992795678306e-05,
      "loss": 0.0271,
      "positive_teacher_pairs": 26634.4,
      "query_h_norm_ratio": 0.424186772108078,
      "relation_loss": 0.0028737506130710245,
      "selected_pairs": 26634.4,
      "step": 18960,
      "student_distance_mean": 23.855875396728514,
      "teacher_distance_mean": 0.8568017423152924,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0028737506130710245,
      "weighted_relation_loss": 0.0028737506130710245
    },
    {
      "action_query_batch_std": 0.3594249993562698,
      "action_query_norm": 41.50054817199707,
      "action_relation_loss": 0.0028646262362599373,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.74555659494855,
      "flow_matching_loss": 0.023777990974485873,
      "grad_norm": 0.08496342599391937,
      "h_stock_valid_norm": 97.66379241943359,
      "invalid_tail_samples": 23.7,
      "learning_rate": 3.2628235239918744e-05,
      "loss": 0.0266,
      "positive_teacher_pairs": 26881.1,
      "query_h_norm_ratio": 0.42493548393249514,
      "relation_loss": 0.0028646262362599373,
      "selected_pairs": 26881.1,
      "step": 18970,
      "student_distance_mean": 23.721554374694826,
      "teacher_distance_mean": 0.8502222955226898,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0028646262362599373,
      "weighted_relation_loss": 0.0028646262362599373
    },
    {
      "action_query_batch_std": 0.3582867443561554,
      "action_query_norm": 41.49713287353516,
      "action_relation_loss": 0.002840516879223287,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.75491113189897,
      "flow_matching_loss": 0.023931561037898064,
      "grad_norm": 0.11863023787736893,
      "h_stock_valid_norm": 97.82100219726563,
      "invalid_tail_samples": 23.4,
      "learning_rate": 3.2576563631386695e-05,
      "loss": 0.0268,
      "positive_teacher_pairs": 26942.5,
      "query_h_norm_ratio": 0.4242159008979797,
      "relation_loss": 0.002840516879223287,
      "selected_pairs": 26942.5,
      "step": 18980,
      "student_distance_mean": 23.629087829589842,
      "teacher_distance_mean": 0.8486142992973328,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002840516879223287,
      "weighted_relation_loss": 0.002840516879223287
    },
    {
      "action_query_batch_std": 0.3589050441980362,
      "action_query_norm": 41.50753402709961,
      "action_relation_loss": 0.002891431818716228,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.764265668849394,
      "flow_matching_loss": 0.022692495584487916,
      "grad_norm": 0.09759300202131271,
      "h_stock_valid_norm": 97.92700576782227,
      "invalid_tail_samples": 26.1,
      "learning_rate": 3.252491319397275e-05,
      "loss": 0.0256,
      "positive_teacher_pairs": 26335.8,
      "query_h_norm_ratio": 0.42386502027511597,
      "relation_loss": 0.002891431818716228,
      "selected_pairs": 26335.8,
      "step": 18990,
      "student_distance_mean": 23.685289192199708,
      "teacher_distance_mean": 0.8515005767345428,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.002891431818716228,
      "weighted_relation_loss": 0.002891431818716228
    },
    {
      "action_query_batch_std": 0.3608406990766525,
      "action_query_norm": 41.51017379760742,
      "action_relation_loss": 0.0028767250245437027,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.773620205799812,
      "flow_matching_loss": 0.02380475401878357,
      "grad_norm": 0.10769370198249817,
      "h_stock_valid_norm": 97.68225631713867,
      "invalid_tail_samples": 24.6,
      "learning_rate": 3.247328399043706e-05,
      "loss": 0.0267,
      "positive_teacher_pairs": 26663.1,
      "query_h_norm_ratio": 0.4249523639678955,
      "relation_loss": 0.0028767250245437027,
      "selected_pairs": 26663.1,
      "step": 19000,
      "student_distance_mean": 23.863079071044922,
      "teacher_distance_mean": 0.8576821208000183,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0028767250245437027,
      "weighted_relation_loss": 0.0028767250245437027
    },
    {
      "action_query_batch_std": 0.3607034295797348,
      "action_query_norm": 41.539558792114256,
      "action_relation_loss": 0.002809323580004275,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.782974742750234,
      "flow_matching_loss": 0.023486892506480218,
      "grad_norm": 0.11660180240869522,
      "h_stock_valid_norm": 97.6992088317871,
      "invalid_tail_samples": 21.6,
      "learning_rate": 3.242167608351399e-05,
      "loss": 0.0263,
      "positive_teacher_pairs": 27358.8,
      "query_h_norm_ratio": 0.42518246471881865,
      "relation_loss": 0.002809323580004275,
      "selected_pairs": 27358.8,
      "step": 19010,
      "student_distance_mean": 23.884040069580077,
      "teacher_distance_mean": 0.8560178697109222,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.002809323580004275,
      "weighted_relation_loss": 0.002809323580004275
    },
    {
      "action_query_batch_std": 0.3620270162820816,
      "action_query_norm": 41.53199577331543,
      "action_relation_loss": 0.0028511233162134887,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.792329279700656,
      "flow_matching_loss": 0.023607007041573525,
      "grad_norm": 0.11745256185531616,
      "h_stock_valid_norm": 98.40322952270508,
      "invalid_tail_samples": 22.4,
      "learning_rate": 3.2370089535911986e-05,
      "loss": 0.0265,
      "positive_teacher_pairs": 27173.5,
      "query_h_norm_ratio": 0.4220657616853714,
      "relation_loss": 0.0028511233162134887,
      "selected_pairs": 27173.5,
      "step": 19020,
      "student_distance_mean": 23.821820640563963,
      "teacher_distance_mean": 0.8555988490581512,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.0028511233162134887,
      "weighted_relation_loss": 0.0028511233162134887
    },
    {
      "action_query_batch_std": 0.3586531072854996,
      "action_query_norm": 41.50255279541015,
      "action_relation_loss": 0.002928019408136606,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.801683816651074,
      "flow_matching_loss": 0.0240304185077548,
      "grad_norm": 0.1024695411324501,
      "h_stock_valid_norm": 98.48077774047852,
      "invalid_tail_samples": 25.0,
      "learning_rate": 3.23185244103136e-05,
      "loss": 0.027,
      "positive_teacher_pairs": 26572.0,
      "query_h_norm_ratio": 0.42143098413944247,
      "relation_loss": 0.002928019408136606,
      "selected_pairs": 26572.0,
      "step": 19030,
      "student_distance_mean": 23.607814598083497,
      "teacher_distance_mean": 0.8479469001293183,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.002928019408136606,
      "weighted_relation_loss": 0.002928019408136606
    },
    {
      "action_query_batch_std": 0.36173448860645296,
      "action_query_norm": 41.491799926757814,
      "action_relation_loss": 0.0027551221661269664,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.811038353601496,
      "flow_matching_loss": 0.024089815840125085,
      "grad_norm": 0.10187318921089172,
      "h_stock_valid_norm": 98.1889144897461,
      "invalid_tail_samples": 25.8,
      "learning_rate": 3.22669807693753e-05,
      "loss": 0.0268,
      "positive_teacher_pairs": 26384.6,
      "query_h_norm_ratio": 0.42257300317287444,
      "relation_loss": 0.0027551221661269664,
      "selected_pairs": 26384.6,
      "step": 19040,
      "student_distance_mean": 23.73546085357666,
      "teacher_distance_mean": 0.8528661787509918,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0027551221661269664,
      "weighted_relation_loss": 0.0027551221661269664
    },
    {
      "action_query_batch_std": 0.36795127093791963,
      "action_query_norm": 41.54629364013672,
      "action_relation_loss": 0.0027972857234999536,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.82039289055192,
      "flow_matching_loss": 0.025444414094090463,
      "grad_norm": 0.10477697849273682,
      "h_stock_valid_norm": 98.32815551757812,
      "invalid_tail_samples": 20.7,
      "learning_rate": 3.2215458675727495e-05,
      "loss": 0.0282,
      "positive_teacher_pairs": 27579.6,
      "query_h_norm_ratio": 0.4225297838449478,
      "relation_loss": 0.0027972857234999536,
      "selected_pairs": 27579.6,
      "step": 19050,
      "student_distance_mean": 24.170174407958985,
      "teacher_distance_mean": 0.8675057232379914,
      "valid_action_samples": 235.3,
      "weighted_auxiliary_loss": 0.0027972857234999536,
      "weighted_relation_loss": 0.0027972857234999536
    },
    {
      "action_query_batch_std": 0.36313726007938385,
      "action_query_norm": 41.513056945800784,
      "action_relation_loss": 0.002839313098229468,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.82974742750234,
      "flow_matching_loss": 0.023823751509189604,
      "grad_norm": 0.10303233563899994,
      "h_stock_valid_norm": 98.0692352294922,
      "invalid_tail_samples": 24.7,
      "learning_rate": 3.216395819197438e-05,
      "loss": 0.0267,
      "positive_teacher_pairs": 26638.7,
      "query_h_norm_ratio": 0.42330682575702666,
      "relation_loss": 0.002839313098229468,
      "selected_pairs": 26638.7,
      "step": 19060,
      "student_distance_mean": 23.988732528686523,
      "teacher_distance_mean": 0.8628901720046998,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.002839313098229468,
      "weighted_relation_loss": 0.002839313098229468
    },
    {
      "action_query_batch_std": 0.3563916623592377,
      "action_query_norm": 41.492240524291994,
      "action_relation_loss": 0.002823011390864849,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.83910196445276,
      "flow_matching_loss": 0.02310802899301052,
      "grad_norm": 0.17021654546260834,
      "h_stock_valid_norm": 98.00934982299805,
      "invalid_tail_samples": 23.1,
      "learning_rate": 3.211247938069387e-05,
      "loss": 0.0259,
      "positive_teacher_pairs": 27025.6,
      "query_h_norm_ratio": 0.42335209548473357,
      "relation_loss": 0.002823011390864849,
      "selected_pairs": 27025.6,
      "step": 19070,
      "student_distance_mean": 23.643420028686524,
      "teacher_distance_mean": 0.8530223488807678,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002823011390864849,
      "weighted_relation_loss": 0.002823011390864849
    },
    {
      "action_query_batch_std": 0.3612700581550598,
      "action_query_norm": 41.50589218139648,
      "action_relation_loss": 0.0026712379651144145,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.84845650140318,
      "flow_matching_loss": 0.024167279340326787,
      "grad_norm": 0.10026083141565323,
      "h_stock_valid_norm": 98.1725357055664,
      "invalid_tail_samples": 25.6,
      "learning_rate": 3.206102230443759e-05,
      "loss": 0.0268,
      "positive_teacher_pairs": 26436.9,
      "query_h_norm_ratio": 0.42278910875320436,
      "relation_loss": 0.0026712379651144145,
      "selected_pairs": 26436.9,
      "step": 19080,
      "student_distance_mean": 23.888658714294433,
      "teacher_distance_mean": 0.8581237733364105,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0026712379651144145,
      "weighted_relation_loss": 0.0026712379651144145
    },
    {
      "action_query_batch_std": 0.35685337483882906,
      "action_query_norm": 41.482547760009766,
      "action_relation_loss": 0.0027846617856994273,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.857811038353603,
      "flow_matching_loss": 0.024244486540555953,
      "grad_norm": 0.11290956288576126,
      "h_stock_valid_norm": 98.34361267089844,
      "invalid_tail_samples": 27.0,
      "learning_rate": 3.2009587025730764e-05,
      "loss": 0.027,
      "positive_teacher_pairs": 26113.4,
      "query_h_norm_ratio": 0.4218147784471512,
      "relation_loss": 0.0027846617856994273,
      "selected_pairs": 26113.4,
      "step": 19090,
      "student_distance_mean": 23.708225059509278,
      "teacher_distance_mean": 0.8547636091709137,
      "valid_action_samples": 229.0,
      "weighted_auxiliary_loss": 0.0027846617856994273,
      "weighted_relation_loss": 0.0027846617856994273
    },
    {
      "action_query_batch_std": 0.36196483075618746,
      "action_query_norm": 41.50472221374512,
      "action_relation_loss": 0.0028033946407958867,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.86716557530402,
      "flow_matching_loss": 0.022957334481179714,
      "grad_norm": 0.10723144561052322,
      "h_stock_valid_norm": 97.95856704711915,
      "invalid_tail_samples": 22.1,
      "learning_rate": 3.195817360707207e-05,
      "loss": 0.0258,
      "positive_teacher_pairs": 27247.5,
      "query_h_norm_ratio": 0.4236998915672302,
      "relation_loss": 0.0028033946407958867,
      "selected_pairs": 27247.5,
      "step": 19100,
      "student_distance_mean": 23.8128454208374,
      "teacher_distance_mean": 0.8567805290222168,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0028033946407958867,
      "weighted_relation_loss": 0.0028033946407958867
    },
    {
      "action_query_batch_std": 0.3620623052120209,
      "action_query_norm": 41.4915771484375,
      "action_relation_loss": 0.002723863604478538,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.876520112254443,
      "flow_matching_loss": 0.021597293205559254,
      "grad_norm": 0.09045445173978806,
      "h_stock_valid_norm": 97.82177963256837,
      "invalid_tail_samples": 25.2,
      "learning_rate": 3.19067821109337e-05,
      "loss": 0.0243,
      "positive_teacher_pairs": 26526.9,
      "query_h_norm_ratio": 0.42415663599967957,
      "relation_loss": 0.002723863604478538,
      "selected_pairs": 26526.9,
      "step": 19110,
      "student_distance_mean": 23.99604721069336,
      "teacher_distance_mean": 0.8633838295936584,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.002723863604478538,
      "weighted_relation_loss": 0.002723863604478538
    },
    {
      "action_query_batch_std": 0.3601249784231186,
      "action_query_norm": 41.482369613647464,
      "action_relation_loss": 0.0027581699658185245,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.885874649204865,
      "flow_matching_loss": 0.023617283441126347,
      "grad_norm": 0.09971446543931961,
      "h_stock_valid_norm": 97.71108932495117,
      "invalid_tail_samples": 23.5,
      "learning_rate": 3.185541259976114e-05,
      "loss": 0.0264,
      "positive_teacher_pairs": 26915.6,
      "query_h_norm_ratio": 0.4245435416698456,
      "relation_loss": 0.0027581699658185245,
      "selected_pairs": 26915.6,
      "step": 19120,
      "student_distance_mean": 23.74877586364746,
      "teacher_distance_mean": 0.8546323716640473,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0027581699658185245,
      "weighted_relation_loss": 0.0027581699658185245
    },
    {
      "action_query_batch_std": 0.3572971343994141,
      "action_query_norm": 41.490908813476565,
      "action_relation_loss": 0.0026908279629424213,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.895229186155284,
      "flow_matching_loss": 0.02277868930250406,
      "grad_norm": 0.10333620011806488,
      "h_stock_valid_norm": 97.89835205078126,
      "invalid_tail_samples": 22.4,
      "learning_rate": 3.180406513597316e-05,
      "loss": 0.0255,
      "positive_teacher_pairs": 27181.6,
      "query_h_norm_ratio": 0.42381811141967773,
      "relation_loss": 0.0026908279629424213,
      "selected_pairs": 27181.6,
      "step": 19130,
      "student_distance_mean": 23.57673625946045,
      "teacher_distance_mean": 0.8495950758457184,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.0026908279629424213,
      "weighted_relation_loss": 0.0026908279629424213
    },
    {
      "action_query_batch_std": 0.3622234106063843,
      "action_query_norm": 41.49251174926758,
      "action_relation_loss": 0.0028177032014355063,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.904583723105706,
      "flow_matching_loss": 0.02486497312784195,
      "grad_norm": 0.11503422260284424,
      "h_stock_valid_norm": 98.08628997802734,
      "invalid_tail_samples": 22.8,
      "learning_rate": 3.1752739781961835e-05,
      "loss": 0.0277,
      "positive_teacher_pairs": 27081.5,
      "query_h_norm_ratio": 0.4230236500501633,
      "relation_loss": 0.0028177032014355063,
      "selected_pairs": 27081.5,
      "step": 19140,
      "student_distance_mean": 23.847881507873534,
      "teacher_distance_mean": 0.8600888133049012,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0028177032014355063,
      "weighted_relation_loss": 0.0028177032014355063
    },
    {
      "action_query_batch_std": 0.35484537184238435,
      "action_query_norm": 41.45974617004394,
      "action_relation_loss": 0.002673183334991336,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.913938260056128,
      "flow_matching_loss": 0.024810289964079857,
      "grad_norm": 0.08469665050506592,
      "h_stock_valid_norm": 98.10970840454101,
      "invalid_tail_samples": 26.1,
      "learning_rate": 3.170143660009228e-05,
      "loss": 0.0275,
      "positive_teacher_pairs": 26319.1,
      "query_h_norm_ratio": 0.42258996367454527,
      "relation_loss": 0.002673183334991336,
      "selected_pairs": 26319.1,
      "step": 19150,
      "student_distance_mean": 23.518774604797365,
      "teacher_distance_mean": 0.8480320274829865,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.002673183334991336,
      "weighted_relation_loss": 0.002673183334991336
    },
    {
      "action_query_batch_std": 0.35855036079883573,
      "action_query_norm": 41.45901679992676,
      "action_relation_loss": 0.0027475317940115927,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.92329279700655,
      "flow_matching_loss": 0.0228135509416461,
      "grad_norm": 0.09406659752130508,
      "h_stock_valid_norm": 98.40460586547852,
      "invalid_tail_samples": 24.8,
      "learning_rate": 3.16501556527027e-05,
      "loss": 0.0256,
      "positive_teacher_pairs": 26629.6,
      "query_h_norm_ratio": 0.42131529450416566,
      "relation_loss": 0.0027475317940115927,
      "selected_pairs": 26629.6,
      "step": 19160,
      "student_distance_mean": 23.493017578125,
      "teacher_distance_mean": 0.8475718796253204,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0027475317940115927,
      "weighted_relation_loss": 0.0027475317940115927
    },
    {
      "action_query_batch_std": 0.3573917210102081,
      "action_query_norm": 41.472578811645505,
      "action_relation_loss": 0.002753725461661816,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.932647333956968,
      "flow_matching_loss": 0.02318431679159403,
      "grad_norm": 0.0802338570356369,
      "h_stock_valid_norm": 98.39999923706054,
      "invalid_tail_samples": 20.9,
      "learning_rate": 3.1598897002104265e-05,
      "loss": 0.0259,
      "positive_teacher_pairs": 27526.3,
      "query_h_norm_ratio": 0.421471044421196,
      "relation_loss": 0.002753725461661816,
      "selected_pairs": 27526.3,
      "step": 19170,
      "student_distance_mean": 23.608107376098634,
      "teacher_distance_mean": 0.8486570298671723,
      "valid_action_samples": 235.1,
      "weighted_auxiliary_loss": 0.002753725461661816,
      "weighted_relation_loss": 0.002753725461661816
    },
    {
      "action_query_batch_std": 0.36230852007865905,
      "action_query_norm": 41.50362358093262,
      "action_relation_loss": 0.0028609688859432937,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.94200187090739,
      "flow_matching_loss": 0.023000023700296877,
      "grad_norm": 0.09073717892169952,
      "h_stock_valid_norm": 98.28783416748047,
      "invalid_tail_samples": 25.4,
      "learning_rate": 3.154766071058108e-05,
      "loss": 0.0259,
      "positive_teacher_pairs": 26483.2,
      "query_h_norm_ratio": 0.42226931154727937,
      "relation_loss": 0.0028609688859432937,
      "selected_pairs": 26483.2,
      "step": 19180,
      "student_distance_mean": 23.93866138458252,
      "teacher_distance_mean": 0.859505432844162,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0028609688859432937,
      "weighted_relation_loss": 0.0028609688859432937
    },
    {
      "action_query_batch_std": 0.35748687386512756,
      "action_query_norm": 41.44481773376465,
      "action_relation_loss": 0.0028607377549633383,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.951356407857812,
      "flow_matching_loss": 0.02304181382060051,
      "grad_norm": 0.09975646436214447,
      "h_stock_valid_norm": 97.8285514831543,
      "invalid_tail_samples": 23.6,
      "learning_rate": 3.149644684039008e-05,
      "loss": 0.0259,
      "positive_teacher_pairs": 26896.1,
      "query_h_norm_ratio": 0.42365066409111024,
      "relation_loss": 0.0028607377549633383,
      "selected_pairs": 26896.1,
      "step": 19190,
      "student_distance_mean": 23.58475532531738,
      "teacher_distance_mean": 0.8509203493595123,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0028607377549633383,
      "weighted_relation_loss": 0.0028607377549633383
    },
    {
      "action_query_batch_std": 0.3584145486354828,
      "action_query_norm": 41.44669761657715,
      "action_relation_loss": 0.0028712031897157432,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.96071094480823,
      "flow_matching_loss": 0.02372115906327963,
      "grad_norm": 0.10137857496738434,
      "h_stock_valid_norm": 97.64738540649414,
      "invalid_tail_samples": 26.5,
      "learning_rate": 3.144525545376095e-05,
      "loss": 0.0266,
      "positive_teacher_pairs": 26225.4,
      "query_h_norm_ratio": 0.4244550108909607,
      "relation_loss": 0.0028712031897157432,
      "selected_pairs": 26225.4,
      "step": 19200,
      "student_distance_mean": 23.69122943878174,
      "teacher_distance_mean": 0.8552833795547485,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.0028712031897157432,
      "weighted_relation_loss": 0.0028712031897157432
    },
    {
      "action_query_batch_std": 0.35771130621433256,
      "action_query_norm": 41.468824768066405,
      "action_relation_loss": 0.002817454095929861,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.970065481758652,
      "flow_matching_loss": 0.022931273840367795,
      "grad_norm": 0.08102462440729141,
      "h_stock_valid_norm": 97.65718536376953,
      "invalid_tail_samples": 25.8,
      "learning_rate": 3.139408661289603e-05,
      "loss": 0.0257,
      "positive_teacher_pairs": 26388.2,
      "query_h_norm_ratio": 0.4246378064155579,
      "relation_loss": 0.002817454095929861,
      "selected_pairs": 26388.2,
      "step": 19210,
      "student_distance_mean": 23.665503311157227,
      "teacher_distance_mean": 0.8518975794315338,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.002817454095929861,
      "weighted_relation_loss": 0.002817454095929861
    },
    {
      "action_query_batch_std": 0.3629310578107834,
      "action_query_norm": 41.475590515136716,
      "action_relation_loss": 0.002833176264539361,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.979420018709074,
      "flow_matching_loss": 0.02265915498137474,
      "grad_norm": 0.10410536825656891,
      "h_stock_valid_norm": 97.77486038208008,
      "invalid_tail_samples": 25.0,
      "learning_rate": 3.134294037997032e-05,
      "loss": 0.0255,
      "positive_teacher_pairs": 26577.9,
      "query_h_norm_ratio": 0.42419821321964263,
      "relation_loss": 0.002833176264539361,
      "selected_pairs": 26577.9,
      "step": 19220,
      "student_distance_mean": 24.00935592651367,
      "teacher_distance_mean": 0.8646668195724487,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.002833176264539361,
      "weighted_relation_loss": 0.002833176264539361
    },
    {
      "action_query_batch_std": 0.35673301815986636,
      "action_query_norm": 41.457810974121095,
      "action_relation_loss": 0.002763610752299428,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.988774555659496,
      "flow_matching_loss": 0.022704635560512543,
      "grad_norm": 0.1001870408654213,
      "h_stock_valid_norm": 97.62434310913086,
      "invalid_tail_samples": 22.9,
      "learning_rate": 3.129181681713127e-05,
      "loss": 0.0255,
      "positive_teacher_pairs": 27057.2,
      "query_h_norm_ratio": 0.4246671885251999,
      "relation_loss": 0.002763610752299428,
      "selected_pairs": 27057.2,
      "step": 19230,
      "student_distance_mean": 23.643404388427733,
      "teacher_distance_mean": 0.8520458459854126,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.002763610752299428,
      "weighted_relation_loss": 0.002763610752299428
    },
    {
      "action_query_batch_std": 0.35639767348766327,
      "action_query_norm": 41.43563957214356,
      "action_relation_loss": 0.002669444400817156,
      "auxiliary_loss_weight": 1.0,
      "epoch": 17.998129092609915,
      "flow_matching_loss": 0.024563679285347462,
      "grad_norm": 0.12982575595378876,
      "h_stock_valid_norm": 98.01669311523438,
      "invalid_tail_samples": 24.3,
      "learning_rate": 3.1240715986498855e-05,
      "loss": 0.0272,
      "positive_teacher_pairs": 26733.5,
      "query_h_norm_ratio": 0.4227430999279022,
      "relation_loss": 0.002669444400817156,
      "selected_pairs": 26733.5,
      "step": 19240,
      "student_distance_mean": 23.4914493560791,
      "teacher_distance_mean": 0.849436491727829,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002669444400817156,
      "weighted_relation_loss": 0.002669444400817156
    },
    {
      "action_query_batch_std": 0.3581672877073288,
      "action_query_norm": 41.4740852355957,
      "action_relation_loss": 0.002736278926022351,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.007483629560337,
      "flow_matching_loss": 0.024217230454087257,
      "grad_norm": 0.11759805679321289,
      "h_stock_valid_norm": 98.04992294311523,
      "invalid_tail_samples": 19.1,
      "learning_rate": 3.1189637950165394e-05,
      "loss": 0.027,
      "positive_teacher_pairs": 24955.9,
      "query_h_norm_ratio": 0.42299116849899293,
      "relation_loss": 0.002736278926022351,
      "selected_pairs": 24955.9,
      "step": 19250,
      "student_distance_mean": 23.714816093444824,
      "teacher_distance_mean": 0.8541015625,
      "valid_action_samples": 217.0,
      "weighted_auxiliary_loss": 0.002736278926022351,
      "weighted_relation_loss": 0.002736278926022351
    },
    {
      "action_query_batch_std": 0.36507436633110046,
      "action_query_norm": 41.49333534240723,
      "action_relation_loss": 0.0026727113872766495,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.01683816651076,
      "flow_matching_loss": 0.023438489250838756,
      "grad_norm": 0.10641654580831528,
      "h_stock_valid_norm": 98.03143310546875,
      "invalid_tail_samples": 21.3,
      "learning_rate": 3.1138582770195544e-05,
      "loss": 0.0261,
      "positive_teacher_pairs": 27433.0,
      "query_h_norm_ratio": 0.4232703894376755,
      "relation_loss": 0.0026727113872766495,
      "selected_pairs": 27433.0,
      "step": 19260,
      "student_distance_mean": 24.09275665283203,
      "teacher_distance_mean": 0.8679484009742737,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.0026727113872766495,
      "weighted_relation_loss": 0.0026727113872766495
    },
    {
      "action_query_batch_std": 0.35786008536815644,
      "action_query_norm": 41.47642860412598,
      "action_relation_loss": 0.0027554942527785896,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.026192703461177,
      "flow_matching_loss": 0.02355671264231205,
      "grad_norm": 0.08615340292453766,
      "h_stock_valid_norm": 97.82714004516602,
      "invalid_tail_samples": 23.6,
      "learning_rate": 3.108755050862615e-05,
      "loss": 0.0263,
      "positive_teacher_pairs": 26895.0,
      "query_h_norm_ratio": 0.423979315161705,
      "relation_loss": 0.0027554942527785896,
      "selected_pairs": 26895.0,
      "step": 19270,
      "student_distance_mean": 23.620745849609374,
      "teacher_distance_mean": 0.8464107871055603,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0027554942527785896,
      "weighted_relation_loss": 0.0027554942527785896
    },
    {
      "action_query_batch_std": 0.3621866166591644,
      "action_query_norm": 41.48133010864258,
      "action_relation_loss": 0.002535320189781487,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.0355472404116,
      "flow_matching_loss": 0.02501037362962961,
      "grad_norm": 0.17279686033725739,
      "h_stock_valid_norm": 97.99017715454102,
      "invalid_tail_samples": 21.5,
      "learning_rate": 3.1036541227466204e-05,
      "loss": 0.0275,
      "positive_teacher_pairs": 27379.5,
      "query_h_norm_ratio": 0.4233247756958008,
      "relation_loss": 0.002535320189781487,
      "selected_pairs": 27379.5,
      "step": 19280,
      "student_distance_mean": 23.90064353942871,
      "teacher_distance_mean": 0.8583579957485199,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.002535320189781487,
      "weighted_relation_loss": 0.002535320189781487
    },
    {
      "action_query_batch_std": 0.3623041152954102,
      "action_query_norm": 41.44520454406738,
      "action_relation_loss": 0.0026081228163093327,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.04490177736202,
      "flow_matching_loss": 0.02327980250120163,
      "grad_norm": 0.11320079118013382,
      "h_stock_valid_norm": 98.22073287963867,
      "invalid_tail_samples": 24.8,
      "learning_rate": 3.0985554988696786e-05,
      "loss": 0.0259,
      "positive_teacher_pairs": 26622.2,
      "query_h_norm_ratio": 0.42196066677570343,
      "relation_loss": 0.0026081228163093327,
      "selected_pairs": 26622.2,
      "step": 19290,
      "student_distance_mean": 24.016537284851076,
      "teacher_distance_mean": 0.8675544619560241,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0026081228163093327,
      "weighted_relation_loss": 0.0026081228163093327
    },
    {
      "action_query_batch_std": 0.36061896979808805,
      "action_query_norm": 41.463731384277345,
      "action_relation_loss": 0.0027009929064661263,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.054256314312443,
      "flow_matching_loss": 0.023587910272181035,
      "grad_norm": 0.1179160624742508,
      "h_stock_valid_norm": 98.21023941040039,
      "invalid_tail_samples": 22.5,
      "learning_rate": 3.093459185427102e-05,
      "loss": 0.0263,
      "positive_teacher_pairs": 27150.4,
      "query_h_norm_ratio": 0.42219552397727966,
      "relation_loss": 0.0027009929064661263,
      "selected_pairs": 27150.4,
      "step": 19300,
      "student_distance_mean": 23.759381675720213,
      "teacher_distance_mean": 0.8553444743156433,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0027009929064661263,
      "weighted_relation_loss": 0.0027009929064661263
    },
    {
      "action_query_batch_std": 0.3589116603136063,
      "action_query_norm": 41.44496307373047,
      "action_relation_loss": 0.0026479646330699325,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.06361085126286,
      "flow_matching_loss": 0.022270402126014232,
      "grad_norm": 0.07799434661865234,
      "h_stock_valid_norm": 98.36068572998047,
      "invalid_tail_samples": 25.0,
      "learning_rate": 3.0883651886113905e-05,
      "loss": 0.0249,
      "positive_teacher_pairs": 26582.5,
      "query_h_norm_ratio": 0.42136096954345703,
      "relation_loss": 0.0026479646330699325,
      "selected_pairs": 26582.5,
      "step": 19310,
      "student_distance_mean": 23.694946098327637,
      "teacher_distance_mean": 0.8530652940273284,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0026479646330699325,
      "weighted_relation_loss": 0.0026479646330699325
    },
    {
      "action_query_batch_std": 0.3570502996444702,
      "action_query_norm": 41.460107421875,
      "action_relation_loss": 0.0028401265852153303,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.072965388213284,
      "flow_matching_loss": 0.023603460192680357,
      "grad_norm": 0.0947563424706459,
      "h_stock_valid_norm": 98.15598602294922,
      "invalid_tail_samples": 24.3,
      "learning_rate": 3.0832735146122294e-05,
      "loss": 0.0264,
      "positive_teacher_pairs": 26733.4,
      "query_h_norm_ratio": 0.4223919838666916,
      "relation_loss": 0.0028401265852153303,
      "selected_pairs": 26733.4,
      "step": 19320,
      "student_distance_mean": 23.62377986907959,
      "teacher_distance_mean": 0.8500696539878845,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0028401265852153303,
      "weighted_relation_loss": 0.0028401265852153303
    },
    {
      "action_query_batch_std": 0.3606134384870529,
      "action_query_norm": 41.481888961791995,
      "action_relation_loss": 0.002553727198392153,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.082319925163706,
      "flow_matching_loss": 0.02342517152428627,
      "grad_norm": 0.10154161602258682,
      "h_stock_valid_norm": 97.96357879638671,
      "invalid_tail_samples": 22.7,
      "learning_rate": 3.078184169616485e-05,
      "loss": 0.026,
      "positive_teacher_pairs": 27105.8,
      "query_h_norm_ratio": 0.42344391644001006,
      "relation_loss": 0.002553727198392153,
      "selected_pairs": 27105.8,
      "step": 19330,
      "student_distance_mean": 23.905222702026368,
      "teacher_distance_mean": 0.8617388665676117,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.002553727198392153,
      "weighted_relation_loss": 0.002553727198392153
    },
    {
      "action_query_batch_std": 0.3574449300765991,
      "action_query_norm": 41.44853248596191,
      "action_relation_loss": 0.0025571966310963035,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.091674462114124,
      "flow_matching_loss": 0.02296879794448614,
      "grad_norm": 0.10318510234355927,
      "h_stock_valid_norm": 97.66276168823242,
      "invalid_tail_samples": 25.6,
      "learning_rate": 3.073097159808187e-05,
      "loss": 0.0255,
      "positive_teacher_pairs": 26442.9,
      "query_h_norm_ratio": 0.42440742552280425,
      "relation_loss": 0.0025571966310963035,
      "selected_pairs": 26442.9,
      "step": 19340,
      "student_distance_mean": 23.626297569274904,
      "teacher_distance_mean": 0.8542176842689514,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0025571966310963035,
      "weighted_relation_loss": 0.0025571966310963035
    },
    {
      "action_query_batch_std": 0.3586744487285614,
      "action_query_norm": 41.42660675048828,
      "action_relation_loss": 0.002694515325129032,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.101028999064546,
      "flow_matching_loss": 0.021434206515550613,
      "grad_norm": 0.0943421795964241,
      "h_stock_valid_norm": 97.23234634399414,
      "invalid_tail_samples": 26.7,
      "learning_rate": 3.068012491368537e-05,
      "loss": 0.0241,
      "positive_teacher_pairs": 26190.4,
      "query_h_norm_ratio": 0.42606072723865507,
      "relation_loss": 0.002694515325129032,
      "selected_pairs": 26190.4,
      "step": 19350,
      "student_distance_mean": 23.730061531066895,
      "teacher_distance_mean": 0.8568482577800751,
      "valid_action_samples": 229.3,
      "weighted_auxiliary_loss": 0.002694515325129032,
      "weighted_relation_loss": 0.002694515325129032
    },
    {
      "action_query_batch_std": 0.3557680696249008,
      "action_query_norm": 41.41257247924805,
      "action_relation_loss": 0.0027668218594044448,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.110383536014968,
      "flow_matching_loss": 0.023550966940820216,
      "grad_norm": 0.08553709089756012,
      "h_stock_valid_norm": 97.66891174316406,
      "invalid_tail_samples": 24.4,
      "learning_rate": 3.062930170475885e-05,
      "loss": 0.0263,
      "positive_teacher_pairs": 26709.4,
      "query_h_norm_ratio": 0.42402098178863523,
      "relation_loss": 0.0027668218594044448,
      "selected_pairs": 26709.4,
      "step": 19360,
      "student_distance_mean": 23.426888275146485,
      "teacher_distance_mean": 0.8473185241222382,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0027668218594044448,
      "weighted_relation_loss": 0.0027668218594044448
    },
    {
      "action_query_batch_std": 0.3637657254934311,
      "action_query_norm": 41.478105545043945,
      "action_relation_loss": 0.002742646145634353,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.119738072965387,
      "flow_matching_loss": 0.02261907756328583,
      "grad_norm": 0.1023261621594429,
      "h_stock_valid_norm": 98.01263046264648,
      "invalid_tail_samples": 22.6,
      "learning_rate": 3.057850203305729e-05,
      "loss": 0.0254,
      "positive_teacher_pairs": 27129.6,
      "query_h_norm_ratio": 0.42319317162036896,
      "relation_loss": 0.002742646145634353,
      "selected_pairs": 27129.6,
      "step": 19370,
      "student_distance_mean": 23.939492416381835,
      "teacher_distance_mean": 0.8642533183097839,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.002742646145634353,
      "weighted_relation_loss": 0.002742646145634353
    },
    {
      "action_query_batch_std": 0.35907046496868134,
      "action_query_norm": 41.429375076293944,
      "action_relation_loss": 0.002613860764540732,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.12909260991581,
      "flow_matching_loss": 0.02274713385850191,
      "grad_norm": 0.0836653858423233,
      "h_stock_valid_norm": 98.18522415161132,
      "invalid_tail_samples": 23.6,
      "learning_rate": 3.052772596030708e-05,
      "loss": 0.0254,
      "positive_teacher_pairs": 26898.7,
      "query_h_norm_ratio": 0.4219519764184952,
      "relation_loss": 0.002613860764540732,
      "selected_pairs": 26898.7,
      "step": 19380,
      "student_distance_mean": 23.756561279296875,
      "teacher_distance_mean": 0.8604044556617737,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.002613860764540732,
      "weighted_relation_loss": 0.002613860764540732
    },
    {
      "action_query_batch_std": 0.3611716955900192,
      "action_query_norm": 41.44697189331055,
      "action_relation_loss": 0.002480732393451035,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.13844714686623,
      "flow_matching_loss": 0.022302616946399213,
      "grad_norm": 0.0933859795331955,
      "h_stock_valid_norm": 97.89070205688476,
      "invalid_tail_samples": 23.3,
      "learning_rate": 3.0476973548205943e-05,
      "loss": 0.0248,
      "positive_teacher_pairs": 26962.1,
      "query_h_norm_ratio": 0.4234019845724106,
      "relation_loss": 0.002480732393451035,
      "selected_pairs": 26962.1,
      "step": 19390,
      "student_distance_mean": 23.702657318115236,
      "teacher_distance_mean": 0.8590641498565674,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.002480732393451035,
      "weighted_relation_loss": 0.002480732393451035
    },
    {
      "action_query_batch_std": 0.3600744217634201,
      "action_query_norm": 41.442613983154295,
      "action_relation_loss": 0.0025790201034396887,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.147801683816652,
      "flow_matching_loss": 0.02449124976992607,
      "grad_norm": 0.10186479985713959,
      "h_stock_valid_norm": 98.13371505737305,
      "invalid_tail_samples": 24.1,
      "learning_rate": 3.0426244858422847e-05,
      "loss": 0.0271,
      "positive_teacher_pairs": 26778.1,
      "query_h_norm_ratio": 0.422308087348938,
      "relation_loss": 0.0025790201034396887,
      "selected_pairs": 26778.1,
      "step": 19400,
      "student_distance_mean": 23.805531120300294,
      "teacher_distance_mean": 0.8601747334003449,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0025790201034396887,
      "weighted_relation_loss": 0.0025790201034396887
    },
    {
      "action_query_batch_std": 0.36036407947540283,
      "action_query_norm": 41.415785217285155,
      "action_relation_loss": 0.002679607574827969,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.15715622076707,
      "flow_matching_loss": 0.0226452449336648,
      "grad_norm": 0.10524068772792816,
      "h_stock_valid_norm": 98.22320785522462,
      "invalid_tail_samples": 24.6,
      "learning_rate": 3.0375539952597943e-05,
      "loss": 0.0253,
      "positive_teacher_pairs": 26666.9,
      "query_h_norm_ratio": 0.42165350914001465,
      "relation_loss": 0.002679607574827969,
      "selected_pairs": 26666.9,
      "step": 19410,
      "student_distance_mean": 23.76598491668701,
      "teacher_distance_mean": 0.8590372681617737,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.002679607574827969,
      "weighted_relation_loss": 0.002679607574827969
    },
    {
      "action_query_batch_std": 0.35667567551136015,
      "action_query_norm": 41.402244567871094,
      "action_relation_loss": 0.0026987661607563497,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.166510757717493,
      "flow_matching_loss": 0.02094700112938881,
      "grad_norm": 0.09207426756620407,
      "h_stock_valid_norm": 97.87904510498046,
      "invalid_tail_samples": 25.8,
      "learning_rate": 3.0324858892342468e-05,
      "loss": 0.0236,
      "positive_teacher_pairs": 26387.5,
      "query_h_norm_ratio": 0.4229989528656006,
      "relation_loss": 0.0026987661607563497,
      "selected_pairs": 26387.5,
      "step": 19420,
      "student_distance_mean": 23.613859176635742,
      "teacher_distance_mean": 0.8549656212329865,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0026987661607563497,
      "weighted_relation_loss": 0.0026987661607563497
    },
    {
      "action_query_batch_std": 0.35340261161327363,
      "action_query_norm": 41.37889175415039,
      "action_relation_loss": 0.002690470195375383,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.175865294667915,
      "flow_matching_loss": 0.022701639123260973,
      "grad_norm": 0.09193913638591766,
      "h_stock_valid_norm": 97.7849952697754,
      "invalid_tail_samples": 25.8,
      "learning_rate": 3.0274201739238672e-05,
      "loss": 0.0254,
      "positive_teacher_pairs": 26388.4,
      "query_h_norm_ratio": 0.4231646955013275,
      "relation_loss": 0.002690470195375383,
      "selected_pairs": 26388.4,
      "step": 19430,
      "student_distance_mean": 23.436806869506835,
      "teacher_distance_mean": 0.8447066426277161,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.002690470195375383,
      "weighted_relation_loss": 0.002690470195375383
    },
    {
      "action_query_batch_std": 0.36092441976070405,
      "action_query_norm": 41.44686775207519,
      "action_relation_loss": 0.002609138307161629,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.185219831618333,
      "flow_matching_loss": 0.02216362152248621,
      "grad_norm": 0.10850401967763901,
      "h_stock_valid_norm": 98.10480270385742,
      "invalid_tail_samples": 22.2,
      "learning_rate": 3.0223568554839786e-05,
      "loss": 0.0248,
      "positive_teacher_pairs": 27230.7,
      "query_h_norm_ratio": 0.42247962653636933,
      "relation_loss": 0.002609138307161629,
      "selected_pairs": 27230.7,
      "step": 19440,
      "student_distance_mean": 23.88570079803467,
      "teacher_distance_mean": 0.8573775887489319,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.002609138307161629,
      "weighted_relation_loss": 0.002609138307161629
    },
    {
      "action_query_batch_std": 0.3555977940559387,
      "action_query_norm": 41.40472831726074,
      "action_relation_loss": 0.0027166108367964624,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.194574368568755,
      "flow_matching_loss": 0.02141523826867342,
      "grad_norm": 0.10226690769195557,
      "h_stock_valid_norm": 98.1840103149414,
      "invalid_tail_samples": 24.0,
      "learning_rate": 3.0172959400669886e-05,
      "loss": 0.0241,
      "positive_teacher_pairs": 26812.1,
      "query_h_norm_ratio": 0.4217079162597656,
      "relation_loss": 0.0027166108367964624,
      "selected_pairs": 26812.1,
      "step": 19450,
      "student_distance_mean": 23.43861026763916,
      "teacher_distance_mean": 0.8464733242988587,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0027166108367964624,
      "weighted_relation_loss": 0.0027166108367964624
    },
    {
      "action_query_batch_std": 0.3547834575176239,
      "action_query_norm": 41.39588890075684,
      "action_relation_loss": 0.0026815916411578656,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.203928905519177,
      "flow_matching_loss": 0.021900357306003572,
      "grad_norm": 0.11034874618053436,
      "h_stock_valid_norm": 98.26103973388672,
      "invalid_tail_samples": 24.3,
      "learning_rate": 3.0122374338223902e-05,
      "loss": 0.0246,
      "positive_teacher_pairs": 26732.2,
      "query_h_norm_ratio": 0.4212875157594681,
      "relation_loss": 0.0026815916411578656,
      "selected_pairs": 26732.2,
      "step": 19460,
      "student_distance_mean": 23.395539474487304,
      "teacher_distance_mean": 0.845202374458313,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0026815916411578656,
      "weighted_relation_loss": 0.0026815916411578656
    },
    {
      "action_query_batch_std": 0.35593743324279786,
      "action_query_norm": 41.423905563354495,
      "action_relation_loss": 0.0026621269062161446,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.2132834424696,
      "flow_matching_loss": 0.023101303912699222,
      "grad_norm": 0.08892893046140671,
      "h_stock_valid_norm": 98.407958984375,
      "invalid_tail_samples": 24.9,
      "learning_rate": 3.0071813428967427e-05,
      "loss": 0.0258,
      "positive_teacher_pairs": 26601.3,
      "query_h_norm_ratio": 0.42094252705574037,
      "relation_loss": 0.0026621269062161446,
      "selected_pairs": 26601.3,
      "step": 19470,
      "student_distance_mean": 23.492894172668457,
      "teacher_distance_mean": 0.8506606340408325,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0026621269062161446,
      "weighted_relation_loss": 0.0026621269062161446
    },
    {
      "action_query_batch_std": 0.35920064747333524,
      "action_query_norm": 41.415354537963864,
      "action_relation_loss": 0.002576515800319612,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.222637979420018,
      "flow_matching_loss": 0.022840193659067153,
      "grad_norm": 0.09743218123912811,
      "h_stock_valid_norm": 98.52072601318359,
      "invalid_tail_samples": 24.3,
      "learning_rate": 3.0021276734336746e-05,
      "loss": 0.0254,
      "positive_teacher_pairs": 26733.7,
      "query_h_norm_ratio": 0.42037380337715147,
      "relation_loss": 0.002576515800319612,
      "selected_pairs": 26733.7,
      "step": 19480,
      "student_distance_mean": 23.791150093078613,
      "teacher_distance_mean": 0.8625891327857971,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002576515800319612,
      "weighted_relation_loss": 0.002576515800319612
    },
    {
      "action_query_batch_std": 0.3574478805065155,
      "action_query_norm": 41.397988891601564,
      "action_relation_loss": 0.002619596687145531,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.23199251637044,
      "flow_matching_loss": 0.020988879911601543,
      "grad_norm": 0.1012367531657219,
      "h_stock_valid_norm": 98.28734741210937,
      "invalid_tail_samples": 24.0,
      "learning_rate": 2.997076431573871e-05,
      "loss": 0.0236,
      "positive_teacher_pairs": 26814.7,
      "query_h_norm_ratio": 0.4212006390094757,
      "relation_loss": 0.002619596687145531,
      "selected_pairs": 26814.7,
      "step": 19490,
      "student_distance_mean": 23.535295677185058,
      "teacher_distance_mean": 0.854188597202301,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.002619596687145531,
      "weighted_relation_loss": 0.002619596687145531
    },
    {
      "action_query_batch_std": 0.3578900694847107,
      "action_query_norm": 41.38380393981934,
      "action_relation_loss": 0.0025151583133265374,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.24134705332086,
      "flow_matching_loss": 0.02238942068070173,
      "grad_norm": 0.09900405257940292,
      "h_stock_valid_norm": 98.28351364135742,
      "invalid_tail_samples": 24.6,
      "learning_rate": 2.9920276234550636e-05,
      "loss": 0.0249,
      "positive_teacher_pairs": 26666.1,
      "query_h_norm_ratio": 0.4210682183504105,
      "relation_loss": 0.0025151583133265374,
      "selected_pairs": 26666.1,
      "step": 19500,
      "student_distance_mean": 23.59519863128662,
      "teacher_distance_mean": 0.858772611618042,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0025151583133265374,
      "weighted_relation_loss": 0.0025151583133265374
    },
    {
      "action_query_batch_std": 0.35594346821308137,
      "action_query_norm": 41.40449409484863,
      "action_relation_loss": 0.0027436514850705863,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.25070159027128,
      "flow_matching_loss": 0.024015955440700055,
      "grad_norm": 0.09272312372922897,
      "h_stock_valid_norm": 98.51466827392578,
      "invalid_tail_samples": 26.1,
      "learning_rate": 2.986981255212035e-05,
      "loss": 0.0268,
      "positive_teacher_pairs": 26322.2,
      "query_h_norm_ratio": 0.4202889323234558,
      "relation_loss": 0.0027436514850705863,
      "selected_pairs": 26322.2,
      "step": 19510,
      "student_distance_mean": 23.577046203613282,
      "teacher_distance_mean": 0.8560577154159545,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.0027436514850705863,
      "weighted_relation_loss": 0.0027436514850705863
    },
    {
      "action_query_batch_std": 0.3569790482521057,
      "action_query_norm": 41.39748573303223,
      "action_relation_loss": 0.002626288472674787,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.260056127221702,
      "flow_matching_loss": 0.024236392602324484,
      "grad_norm": 0.1103837862610817,
      "h_stock_valid_norm": 98.46048049926758,
      "invalid_tail_samples": 23.6,
      "learning_rate": 2.9819373329765977e-05,
      "loss": 0.0269,
      "positive_teacher_pairs": 26897.4,
      "query_h_norm_ratio": 0.4204498648643494,
      "relation_loss": 0.002626288472674787,
      "selected_pairs": 26897.4,
      "step": 19520,
      "student_distance_mean": 23.607198905944824,
      "teacher_distance_mean": 0.8494709014892579,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.002626288472674787,
      "weighted_relation_loss": 0.002626288472674787
    },
    {
      "action_query_batch_std": 0.35919623970985415,
      "action_query_norm": 41.40957489013672,
      "action_relation_loss": 0.002631566091440618,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.269410664172124,
      "flow_matching_loss": 0.020875893719494342,
      "grad_norm": 0.10643650591373444,
      "h_stock_valid_norm": 98.28373184204102,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.9768958628775902e-05,
      "loss": 0.0235,
      "positive_teacher_pairs": 26965.3,
      "query_h_norm_ratio": 0.4213291883468628,
      "relation_loss": 0.002631566091440618,
      "selected_pairs": 26965.3,
      "step": 19530,
      "student_distance_mean": 23.73583564758301,
      "teacher_distance_mean": 0.8549091815948486,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.002631566091440618,
      "weighted_relation_loss": 0.002631566091440618
    },
    {
      "action_query_batch_std": 0.35595890283584597,
      "action_query_norm": 41.38478012084961,
      "action_relation_loss": 0.002726719621568918,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.278765201122546,
      "flow_matching_loss": 0.020870660804212092,
      "grad_norm": 0.08906856179237366,
      "h_stock_valid_norm": 98.29844436645507,
      "invalid_tail_samples": 24.9,
      "learning_rate": 2.9718568510408762e-05,
      "loss": 0.0236,
      "positive_teacher_pairs": 26598.2,
      "query_h_norm_ratio": 0.4210124731063843,
      "relation_loss": 0.002726719621568918,
      "selected_pairs": 26598.2,
      "step": 19540,
      "student_distance_mean": 23.5503719329834,
      "teacher_distance_mean": 0.8511991500854492,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.002726719621568918,
      "weighted_relation_loss": 0.002726719621568918
    },
    {
      "action_query_batch_std": 0.3542787045240402,
      "action_query_norm": 41.368818283081055,
      "action_relation_loss": 0.0026567675173282622,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.288119738072965,
      "flow_matching_loss": 0.023651565238833428,
      "grad_norm": 0.09059194475412369,
      "h_stock_valid_norm": 98.15455474853516,
      "invalid_tail_samples": 22.4,
      "learning_rate": 2.9668203035893272e-05,
      "loss": 0.0263,
      "positive_teacher_pairs": 27171.5,
      "query_h_norm_ratio": 0.42146817445755,
      "relation_loss": 0.0026567675173282622,
      "selected_pairs": 27171.5,
      "step": 19550,
      "student_distance_mean": 23.370348930358887,
      "teacher_distance_mean": 0.8483382046222687,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.0026567675173282622,
      "weighted_relation_loss": 0.0026567675173282622
    },
    {
      "action_query_batch_std": 0.35629654228687285,
      "action_query_norm": 41.3918342590332,
      "action_relation_loss": 0.0026406050892546773,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.297474275023387,
      "flow_matching_loss": 0.023109016567468645,
      "grad_norm": 0.09879928082227707,
      "h_stock_valid_norm": 98.03748168945313,
      "invalid_tail_samples": 24.7,
      "learning_rate": 2.9617862266428287e-05,
      "loss": 0.0257,
      "positive_teacher_pairs": 26642.5,
      "query_h_norm_ratio": 0.42220781147480013,
      "relation_loss": 0.0026406050892546773,
      "selected_pairs": 26642.5,
      "step": 19560,
      "student_distance_mean": 23.562988090515137,
      "teacher_distance_mean": 0.850154447555542,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0026406050892546773,
      "weighted_relation_loss": 0.0026406050892546773
    },
    {
      "action_query_batch_std": 0.35659637451171877,
      "action_query_norm": 41.40700988769531,
      "action_relation_loss": 0.0027947211638092995,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.30682881197381,
      "flow_matching_loss": 0.02316934783011675,
      "grad_norm": 0.10217069089412689,
      "h_stock_valid_norm": 98.01484375,
      "invalid_tail_samples": 19.8,
      "learning_rate": 2.9567546263182556e-05,
      "loss": 0.026,
      "positive_teacher_pairs": 27784.2,
      "query_h_norm_ratio": 0.4224591463804245,
      "relation_loss": 0.0027947211638092995,
      "selected_pairs": 27784.2,
      "step": 19570,
      "student_distance_mean": 23.53284149169922,
      "teacher_distance_mean": 0.8502439260482788,
      "valid_action_samples": 236.2,
      "weighted_auxiliary_loss": 0.0027947211638092995,
      "weighted_relation_loss": 0.0027947211638092995
    },
    {
      "action_query_batch_std": 0.355823016166687,
      "action_query_norm": 41.373412704467775,
      "action_relation_loss": 0.002762938756495714,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.316183348924227,
      "flow_matching_loss": 0.02488679401576519,
      "grad_norm": 0.10518863797187805,
      "h_stock_valid_norm": 98.2110824584961,
      "invalid_tail_samples": 25.5,
      "learning_rate": 2.951725508729476e-05,
      "loss": 0.0276,
      "positive_teacher_pairs": 26463.2,
      "query_h_norm_ratio": 0.4212724775075912,
      "relation_loss": 0.002762938756495714,
      "selected_pairs": 26463.2,
      "step": 19580,
      "student_distance_mean": 23.539920806884766,
      "teacher_distance_mean": 0.8492608606815338,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.002762938756495714,
      "weighted_relation_loss": 0.002762938756495714
    },
    {
      "action_query_batch_std": 0.3514871150255203,
      "action_query_norm": 41.35981063842773,
      "action_relation_loss": 0.0028329209424555303,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.32553788587465,
      "flow_matching_loss": 0.022844151966273783,
      "grad_norm": 0.11011535674333572,
      "h_stock_valid_norm": 98.38299560546875,
      "invalid_tail_samples": 24.2,
      "learning_rate": 2.946698879987344e-05,
      "loss": 0.0257,
      "positive_teacher_pairs": 26761.1,
      "query_h_norm_ratio": 0.4203979939222336,
      "relation_loss": 0.0028329209424555303,
      "selected_pairs": 26761.1,
      "step": 19590,
      "student_distance_mean": 23.181852722167967,
      "teacher_distance_mean": 0.8370276272296906,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0028329209424555303,
      "weighted_relation_loss": 0.0028329209424555303
    },
    {
      "action_query_batch_std": 0.36147980093955995,
      "action_query_norm": 41.426152038574216,
      "action_relation_loss": 0.002663146215490997,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.33489242282507,
      "flow_matching_loss": 0.02329769767820835,
      "grad_norm": 0.10243247449398041,
      "h_stock_valid_norm": 98.03048171997071,
      "invalid_tail_samples": 24.8,
      "learning_rate": 2.9416747461996853e-05,
      "loss": 0.026,
      "positive_teacher_pairs": 26625.5,
      "query_h_norm_ratio": 0.4225892871618271,
      "relation_loss": 0.002663146215490997,
      "selected_pairs": 26625.5,
      "step": 19600,
      "student_distance_mean": 23.86171398162842,
      "teacher_distance_mean": 0.8621775865554809,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.002663146215490997,
      "weighted_relation_loss": 0.002663146215490997
    },
    {
      "action_query_batch_std": 0.35823929905891416,
      "action_query_norm": 41.3947265625,
      "action_relation_loss": 0.002645990275777876,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.34424695977549,
      "flow_matching_loss": 0.022595191746950148,
      "grad_norm": 0.11102621257305145,
      "h_stock_valid_norm": 98.19792175292969,
      "invalid_tail_samples": 25.2,
      "learning_rate": 2.9366531134712973e-05,
      "loss": 0.0252,
      "positive_teacher_pairs": 26521.6,
      "query_h_norm_ratio": 0.4215459585189819,
      "relation_loss": 0.002645990275777876,
      "selected_pairs": 26521.6,
      "step": 19610,
      "student_distance_mean": 23.595273590087892,
      "teacher_distance_mean": 0.8535040199756623,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.002645990275777876,
      "weighted_relation_loss": 0.002645990275777876
    },
    {
      "action_query_batch_std": 0.3597446769475937,
      "action_query_norm": 41.39160652160645,
      "action_relation_loss": 0.0026866156375035645,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.35360149672591,
      "flow_matching_loss": 0.021723247319459914,
      "grad_norm": 0.11830814927816391,
      "h_stock_valid_norm": 98.461962890625,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.9316339879039367e-05,
      "loss": 0.0244,
      "positive_teacher_pairs": 26971.1,
      "query_h_norm_ratio": 0.4203841060400009,
      "relation_loss": 0.0026866156375035645,
      "selected_pairs": 26971.1,
      "step": 19620,
      "student_distance_mean": 23.785003089904784,
      "teacher_distance_mean": 0.8578187227249146,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0026866156375035645,
      "weighted_relation_loss": 0.0026866156375035645
    },
    {
      "action_query_batch_std": 0.3564300358295441,
      "action_query_norm": 41.379225158691405,
      "action_relation_loss": 0.002667406597174704,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.362956033676333,
      "flow_matching_loss": 0.02195804752409458,
      "grad_norm": 0.10853725671768188,
      "h_stock_valid_norm": 98.85460662841797,
      "invalid_tail_samples": 22.5,
      "learning_rate": 2.9266173755963167e-05,
      "loss": 0.0246,
      "positive_teacher_pairs": 27152.3,
      "query_h_norm_ratio": 0.41858886182308197,
      "relation_loss": 0.002667406597174704,
      "selected_pairs": 27152.3,
      "step": 19630,
      "student_distance_mean": 23.492617225646974,
      "teacher_distance_mean": 0.8456463754177094,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.002667406597174704,
      "weighted_relation_loss": 0.002667406597174704
    },
    {
      "action_query_batch_std": 0.3541749805212021,
      "action_query_norm": 41.36149978637695,
      "action_relation_loss": 0.0026765787042677402,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.372310570626755,
      "flow_matching_loss": 0.021484961546957494,
      "grad_norm": 0.07650695741176605,
      "h_stock_valid_norm": 98.86350021362304,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.9216032826440926e-05,
      "loss": 0.0242,
      "positive_teacher_pairs": 26963.2,
      "query_h_norm_ratio": 0.41836984157562257,
      "relation_loss": 0.0026765787042677402,
      "selected_pairs": 26963.2,
      "step": 19640,
      "student_distance_mean": 23.37791805267334,
      "teacher_distance_mean": 0.8437573075294494,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0026765787042677402,
      "weighted_relation_loss": 0.0026765787042677402
    },
    {
      "action_query_batch_std": 0.3618071049451828,
      "action_query_norm": 41.417689895629884,
      "action_relation_loss": 0.002622066275216639,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.381665107577174,
      "flow_matching_loss": 0.022268937900662422,
      "grad_norm": 0.11574459820985794,
      "h_stock_valid_norm": 98.86472396850586,
      "invalid_tail_samples": 24.0,
      "learning_rate": 2.9165917151398592e-05,
      "loss": 0.0249,
      "positive_teacher_pairs": 26805.9,
      "query_h_norm_ratio": 0.41893393099308013,
      "relation_loss": 0.002622066275216639,
      "selected_pairs": 26805.9,
      "step": 19650,
      "student_distance_mean": 23.917077827453614,
      "teacher_distance_mean": 0.8637557208538056,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.002622066275216639,
      "weighted_relation_loss": 0.002622066275216639
    },
    {
      "action_query_batch_std": 0.36481223106384275,
      "action_query_norm": 41.43386535644531,
      "action_relation_loss": 0.0024777580983936787,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.391019644527596,
      "flow_matching_loss": 0.02297003027051687,
      "grad_norm": 0.08197879791259766,
      "h_stock_valid_norm": 98.32574768066407,
      "invalid_tail_samples": 22.8,
      "learning_rate": 2.9115826791731428e-05,
      "loss": 0.0254,
      "positive_teacher_pairs": 27093.6,
      "query_h_norm_ratio": 0.42139922380447387,
      "relation_loss": 0.0024777580983936787,
      "selected_pairs": 27093.6,
      "step": 19660,
      "student_distance_mean": 23.99474678039551,
      "teacher_distance_mean": 0.8645600199699401,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0024777580983936787,
      "weighted_relation_loss": 0.0024777580983936787
    },
    {
      "action_query_batch_std": 0.3595277041196823,
      "action_query_norm": 41.40184898376465,
      "action_relation_loss": 0.002677786792628467,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.400374181478018,
      "flow_matching_loss": 0.022863910906016825,
      "grad_norm": 0.08673696964979172,
      "h_stock_valid_norm": 97.97906188964843,
      "invalid_tail_samples": 23.9,
      "learning_rate": 2.906576180830398e-05,
      "loss": 0.0255,
      "positive_teacher_pairs": 26831.8,
      "query_h_norm_ratio": 0.4225623935461044,
      "relation_loss": 0.002677786792628467,
      "selected_pairs": 26831.8,
      "step": 19670,
      "student_distance_mean": 23.884504127502442,
      "teacher_distance_mean": 0.8607929170131683,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.002677786792628467,
      "weighted_relation_loss": 0.002677786792628467
    },
    {
      "action_query_batch_std": 0.35551570653915404,
      "action_query_norm": 41.384291076660155,
      "action_relation_loss": 0.002553486474789679,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.409728718428436,
      "flow_matching_loss": 0.021178372763097288,
      "grad_norm": 0.12872348725795746,
      "h_stock_valid_norm": 97.8449806213379,
      "invalid_tail_samples": 24.4,
      "learning_rate": 2.9015722261949917e-05,
      "loss": 0.0237,
      "positive_teacher_pairs": 26712.5,
      "query_h_norm_ratio": 0.42295933067798613,
      "relation_loss": 0.002553486474789679,
      "selected_pairs": 26712.5,
      "step": 19680,
      "student_distance_mean": 23.49356460571289,
      "teacher_distance_mean": 0.8486388802528382,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.002553486474789679,
      "weighted_relation_loss": 0.002553486474789679
    },
    {
      "action_query_batch_std": 0.3578273594379425,
      "action_query_norm": 41.3651668548584,
      "action_relation_loss": 0.0026614539325237273,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.419083255378858,
      "flow_matching_loss": 0.022904725559055806,
      "grad_norm": 0.1259746551513672,
      "h_stock_valid_norm": 98.10956420898438,
      "invalid_tail_samples": 23.2,
      "learning_rate": 2.8965708213471986e-05,
      "loss": 0.0256,
      "positive_teacher_pairs": 26984.7,
      "query_h_norm_ratio": 0.42162426710128786,
      "relation_loss": 0.0026614539325237273,
      "selected_pairs": 26984.7,
      "step": 19690,
      "student_distance_mean": 23.59221134185791,
      "teacher_distance_mean": 0.8547618389129639,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0026614539325237273,
      "weighted_relation_loss": 0.0026614539325237273
    },
    {
      "action_query_batch_std": 0.3588246822357178,
      "action_query_norm": 41.41553649902344,
      "action_relation_loss": 0.0026156280422583223,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.42843779232928,
      "flow_matching_loss": 0.022151867486536504,
      "grad_norm": 0.13998223841190338,
      "h_stock_valid_norm": 98.17529602050782,
      "invalid_tail_samples": 24.0,
      "learning_rate": 2.8915719723641975e-05,
      "loss": 0.0248,
      "positive_teacher_pairs": 26802.2,
      "query_h_norm_ratio": 0.42185634970664976,
      "relation_loss": 0.0026156280422583223,
      "selected_pairs": 26802.2,
      "step": 19700,
      "student_distance_mean": 23.575327682495118,
      "teacher_distance_mean": 0.84865021109581,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0026156280422583223,
      "weighted_relation_loss": 0.0026156280422583223
    },
    {
      "action_query_batch_std": 0.36106758415699003,
      "action_query_norm": 41.40872344970703,
      "action_relation_loss": 0.0027723102597519756,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.437792329279702,
      "flow_matching_loss": 0.021963108330965042,
      "grad_norm": 0.11503072082996368,
      "h_stock_valid_norm": 97.74918060302734,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.8865756853200604e-05,
      "loss": 0.0247,
      "positive_teacher_pairs": 26972.9,
      "query_h_norm_ratio": 0.42363132536411285,
      "relation_loss": 0.0027723102597519756,
      "selected_pairs": 26972.9,
      "step": 19710,
      "student_distance_mean": 23.847781944274903,
      "teacher_distance_mean": 0.8594668686389924,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0027723102597519756,
      "weighted_relation_loss": 0.0027723102597519756
    },
    {
      "action_query_batch_std": 0.3578753054141998,
      "action_query_norm": 41.380736923217775,
      "action_relation_loss": 0.0026065065059810875,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.44714686623012,
      "flow_matching_loss": 0.02189441882073879,
      "grad_norm": 0.12581583857536316,
      "h_stock_valid_norm": 97.85414199829101,
      "invalid_tail_samples": 26.0,
      "learning_rate": 2.8815819662857503e-05,
      "loss": 0.0245,
      "positive_teacher_pairs": 26343.1,
      "query_h_norm_ratio": 0.4228838622570038,
      "relation_loss": 0.0026065065059810875,
      "selected_pairs": 26343.1,
      "step": 19720,
      "student_distance_mean": 23.77409973144531,
      "teacher_distance_mean": 0.8570678651332855,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.0026065065059810875,
      "weighted_relation_loss": 0.0026065065059810875
    },
    {
      "action_query_batch_std": 0.36046393811702726,
      "action_query_norm": 41.375905227661136,
      "action_relation_loss": 0.0026387337129563093,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.456501403180543,
      "flow_matching_loss": 0.023972827196121215,
      "grad_norm": 0.1539067029953003,
      "h_stock_valid_norm": 98.18235473632812,
      "invalid_tail_samples": 22.3,
      "learning_rate": 2.876590821329105e-05,
      "loss": 0.0266,
      "positive_teacher_pairs": 27195.8,
      "query_h_norm_ratio": 0.4214207947254181,
      "relation_loss": 0.0026387337129563093,
      "selected_pairs": 27195.8,
      "step": 19730,
      "student_distance_mean": 23.742191886901857,
      "teacher_distance_mean": 0.8560385882854462,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0026387337129563093,
      "weighted_relation_loss": 0.0026387337129563093
    },
    {
      "action_query_batch_std": 0.3610475420951843,
      "action_query_norm": 41.375936126708986,
      "action_relation_loss": 0.0027030311757698655,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.465855940130965,
      "flow_matching_loss": 0.023368004895746707,
      "grad_norm": 0.13853630423545837,
      "h_stock_valid_norm": 98.16306686401367,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.871602256514836e-05,
      "loss": 0.0261,
      "positive_teacher_pairs": 26966.2,
      "query_h_norm_ratio": 0.42150344848632815,
      "relation_loss": 0.0027030311757698655,
      "selected_pairs": 26966.2,
      "step": 19740,
      "student_distance_mean": 23.86467037200928,
      "teacher_distance_mean": 0.863780802488327,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0027030311757698655,
      "weighted_relation_loss": 0.0027030311757698655
    },
    {
      "action_query_batch_std": 0.358934623003006,
      "action_query_norm": 41.3751106262207,
      "action_relation_loss": 0.00272595772985369,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.475210477081383,
      "flow_matching_loss": 0.020744152925908567,
      "grad_norm": 0.12530578672885895,
      "h_stock_valid_norm": 97.93003921508789,
      "invalid_tail_samples": 23.0,
      "learning_rate": 2.8666162779045203e-05,
      "loss": 0.0235,
      "positive_teacher_pairs": 27036.5,
      "query_h_norm_ratio": 0.4225002646446228,
      "relation_loss": 0.00272595772985369,
      "selected_pairs": 27036.5,
      "step": 19750,
      "student_distance_mean": 23.72723846435547,
      "teacher_distance_mean": 0.8592106580734253,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.00272595772985369,
      "weighted_relation_loss": 0.00272595772985369
    },
    {
      "action_query_batch_std": 0.35667158365249635,
      "action_query_norm": 41.324711608886716,
      "action_relation_loss": 0.0025983996456488967,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.484565014031805,
      "flow_matching_loss": 0.02246224693953991,
      "grad_norm": 0.11072702705860138,
      "h_stock_valid_norm": 97.8064697265625,
      "invalid_tail_samples": 22.0,
      "learning_rate": 2.8616328915565904e-05,
      "loss": 0.0251,
      "positive_teacher_pairs": 27267.6,
      "query_h_norm_ratio": 0.4225189179182053,
      "relation_loss": 0.0025983996456488967,
      "selected_pairs": 27267.6,
      "step": 19760,
      "student_distance_mean": 23.543304443359375,
      "teacher_distance_mean": 0.8537887811660767,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.0025983996456488967,
      "weighted_relation_loss": 0.0025983996456488967
    },
    {
      "action_query_batch_std": 0.3619531661272049,
      "action_query_norm": 41.38009910583496,
      "action_relation_loss": 0.002657887525856495,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.493919550982227,
      "flow_matching_loss": 0.02349626161158085,
      "grad_norm": 0.11045220494270325,
      "h_stock_valid_norm": 98.31630935668946,
      "invalid_tail_samples": 24.7,
      "learning_rate": 2.856652103526334e-05,
      "loss": 0.0262,
      "positive_teacher_pairs": 26648.7,
      "query_h_norm_ratio": 0.4208905905485153,
      "relation_loss": 0.002657887525856495,
      "selected_pairs": 26648.7,
      "step": 19770,
      "student_distance_mean": 23.822317695617677,
      "teacher_distance_mean": 0.8601207256317138,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.002657887525856495,
      "weighted_relation_loss": 0.002657887525856495
    },
    {
      "action_query_batch_std": 0.35978823304176333,
      "action_query_norm": 41.37784881591797,
      "action_relation_loss": 0.0025720989098772405,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.50327408793265,
      "flow_matching_loss": 0.021788700856268405,
      "grad_norm": 0.10798805952072144,
      "h_stock_valid_norm": 98.28110961914062,
      "invalid_tail_samples": 23.1,
      "learning_rate": 2.8516739198658755e-05,
      "loss": 0.0244,
      "positive_teacher_pairs": 27010.2,
      "query_h_norm_ratio": 0.42101784348487853,
      "relation_loss": 0.0025720989098772405,
      "selected_pairs": 27010.2,
      "step": 19780,
      "student_distance_mean": 23.78249855041504,
      "teacher_distance_mean": 0.8561775147914886,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0025720989098772405,
      "weighted_relation_loss": 0.0025720989098772405
    },
    {
      "action_query_batch_std": 0.36792449057102206,
      "action_query_norm": 41.42776947021484,
      "action_relation_loss": 0.0026913550915196537,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.512628624883067,
      "flow_matching_loss": 0.021921231225132943,
      "grad_norm": 0.09597475826740265,
      "h_stock_valid_norm": 98.33642883300782,
      "invalid_tail_samples": 22.3,
      "learning_rate": 2.846698346624177e-05,
      "loss": 0.0246,
      "positive_teacher_pairs": 27197.8,
      "query_h_norm_ratio": 0.4212906450033188,
      "relation_loss": 0.0026913550915196537,
      "selected_pairs": 27197.8,
      "step": 19790,
      "student_distance_mean": 24.226238632202147,
      "teacher_distance_mean": 0.8694282591342926,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0026913550915196537,
      "weighted_relation_loss": 0.0026913550915196537
    },
    {
      "action_query_batch_std": 0.35965092182159425,
      "action_query_norm": 41.36722946166992,
      "action_relation_loss": 0.0026205964153632522,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.52198316183349,
      "flow_matching_loss": 0.020195372402668,
      "grad_norm": 0.13719776272773743,
      "h_stock_valid_norm": 98.35628967285156,
      "invalid_tail_samples": 22.0,
      "learning_rate": 2.841725389847032e-05,
      "loss": 0.0228,
      "positive_teacher_pairs": 27264.6,
      "query_h_norm_ratio": 0.4205898970365524,
      "relation_loss": 0.0026205964153632522,
      "selected_pairs": 27264.6,
      "step": 19800,
      "student_distance_mean": 23.68711929321289,
      "teacher_distance_mean": 0.8536351382732391,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.0026205964153632522,
      "weighted_relation_loss": 0.0026205964153632522
    },
    {
      "action_query_batch_std": 0.3593859076499939,
      "action_query_norm": 41.38500480651855,
      "action_relation_loss": 0.002538703242316842,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.53133769878391,
      "flow_matching_loss": 0.022501344606280328,
      "grad_norm": 0.08456255495548248,
      "h_stock_valid_norm": 98.33751220703125,
      "invalid_tail_samples": 24.8,
      "learning_rate": 2.8367550555770506e-05,
      "loss": 0.025,
      "positive_teacher_pairs": 26622.3,
      "query_h_norm_ratio": 0.42084677517414093,
      "relation_loss": 0.002538703242316842,
      "selected_pairs": 26622.3,
      "step": 19810,
      "student_distance_mean": 23.764331436157228,
      "teacher_distance_mean": 0.8602001070976257,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.002538703242316842,
      "weighted_relation_loss": 0.002538703242316842
    },
    {
      "action_query_batch_std": 0.3672528237104416,
      "action_query_norm": 41.404777908325194,
      "action_relation_loss": 0.002587216906249523,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.54069223573433,
      "flow_matching_loss": 0.021948764473199843,
      "grad_norm": 0.13926994800567627,
      "h_stock_valid_norm": 98.3197006225586,
      "invalid_tail_samples": 22.7,
      "learning_rate": 2.831787349853655e-05,
      "loss": 0.0245,
      "positive_teacher_pairs": 27105.9,
      "query_h_norm_ratio": 0.42112602591514586,
      "relation_loss": 0.002587216906249523,
      "selected_pairs": 27105.9,
      "step": 19820,
      "student_distance_mean": 24.11591148376465,
      "teacher_distance_mean": 0.874002856016159,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.002587216906249523,
      "weighted_relation_loss": 0.002587216906249523
    },
    {
      "action_query_batch_std": 0.35580851435661315,
      "action_query_norm": 41.3259391784668,
      "action_relation_loss": 0.002647037268616259,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.55004677268475,
      "flow_matching_loss": 0.02159866727888584,
      "grad_norm": 0.12357795238494873,
      "h_stock_valid_norm": 98.2731948852539,
      "invalid_tail_samples": 22.8,
      "learning_rate": 2.8268222787130806e-05,
      "loss": 0.0242,
      "positive_teacher_pairs": 27079.4,
      "query_h_norm_ratio": 0.4205227971076965,
      "relation_loss": 0.002647037268616259,
      "selected_pairs": 27079.4,
      "step": 19830,
      "student_distance_mean": 23.496504402160646,
      "teacher_distance_mean": 0.8480593204498291,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.002647037268616259,
      "weighted_relation_loss": 0.002647037268616259
    },
    {
      "action_query_batch_std": 0.3620312035083771,
      "action_query_norm": 41.39604797363281,
      "action_relation_loss": 0.002566661289893091,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.559401309635174,
      "flow_matching_loss": 0.02302019000053406,
      "grad_norm": 0.12705574929714203,
      "h_stock_valid_norm": 98.27658004760742,
      "invalid_tail_samples": 23.9,
      "learning_rate": 2.821859848188355e-05,
      "loss": 0.0256,
      "positive_teacher_pairs": 26827.1,
      "query_h_norm_ratio": 0.4212249755859375,
      "relation_loss": 0.002566661289893091,
      "selected_pairs": 26827.1,
      "step": 19840,
      "student_distance_mean": 24.00627613067627,
      "teacher_distance_mean": 0.8670372843742371,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.002566661289893091,
      "weighted_relation_loss": 0.002566661289893091
    },
    {
      "action_query_batch_std": 0.35729049146175385,
      "action_query_norm": 41.3395191192627,
      "action_relation_loss": 0.0026144714327529074,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.568755846585596,
      "flow_matching_loss": 0.022712063416838645,
      "grad_norm": 0.09467917680740356,
      "h_stock_valid_norm": 98.37805404663087,
      "invalid_tail_samples": 25.5,
      "learning_rate": 2.816900064309299e-05,
      "loss": 0.0253,
      "positive_teacher_pairs": 26459.3,
      "query_h_norm_ratio": 0.4202135533094406,
      "relation_loss": 0.0026144714327529074,
      "selected_pairs": 26459.3,
      "step": 19850,
      "student_distance_mean": 23.64160041809082,
      "teacher_distance_mean": 0.8545915007591247,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0026144714327529074,
      "weighted_relation_loss": 0.0026144714327529074
    },
    {
      "action_query_batch_std": 0.35676725804805753,
      "action_query_norm": 41.3462833404541,
      "action_relation_loss": 0.002531116595491767,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.578110383536014,
      "flow_matching_loss": 0.02465194445103407,
      "grad_norm": 0.1456872820854187,
      "h_stock_valid_norm": 98.74438705444337,
      "invalid_tail_samples": 23.7,
      "learning_rate": 2.811942933102517e-05,
      "loss": 0.0272,
      "positive_teacher_pairs": 26871.9,
      "query_h_norm_ratio": 0.41872615814208985,
      "relation_loss": 0.002531116595491767,
      "selected_pairs": 26871.9,
      "step": 19860,
      "student_distance_mean": 23.616798210144044,
      "teacher_distance_mean": 0.8565112113952636,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.002531116595491767,
      "weighted_relation_loss": 0.002531116595491767
    },
    {
      "action_query_batch_std": 0.3567824333906174,
      "action_query_norm": 41.3347785949707,
      "action_relation_loss": 0.0024985318770632146,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.587464920486436,
      "flow_matching_loss": 0.021645222790539265,
      "grad_norm": 0.1867092102766037,
      "h_stock_valid_norm": 98.7174575805664,
      "invalid_tail_samples": 23.5,
      "learning_rate": 2.806988460591391e-05,
      "loss": 0.0241,
      "positive_teacher_pairs": 26916.6,
      "query_h_norm_ratio": 0.4187199592590332,
      "relation_loss": 0.0024985318770632146,
      "selected_pairs": 26916.6,
      "step": 19870,
      "student_distance_mean": 23.569951057434082,
      "teacher_distance_mean": 0.8534242689609528,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0024985318770632146,
      "weighted_relation_loss": 0.0024985318770632146
    },
    {
      "action_query_batch_std": 0.3644925504922867,
      "action_query_norm": 41.38187828063965,
      "action_relation_loss": 0.0026091555831953884,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.596819457436858,
      "flow_matching_loss": 0.023245042935013772,
      "grad_norm": 0.09517563879489899,
      "h_stock_valid_norm": 98.57291793823242,
      "invalid_tail_samples": 21.4,
      "learning_rate": 2.802036652796074e-05,
      "loss": 0.0259,
      "positive_teacher_pairs": 27410.0,
      "query_h_norm_ratio": 0.41981338858604433,
      "relation_loss": 0.0026091555831953884,
      "selected_pairs": 27410.0,
      "step": 19880,
      "student_distance_mean": 23.962178039550782,
      "teacher_distance_mean": 0.8655130565166473,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.0026091555831953884,
      "weighted_relation_loss": 0.0026091555831953884
    },
    {
      "action_query_batch_std": 0.3591341197490692,
      "action_query_norm": 41.35035438537598,
      "action_relation_loss": 0.0027192543260753156,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.606173994387277,
      "flow_matching_loss": 0.02171615567058325,
      "grad_norm": 0.07781301438808441,
      "h_stock_valid_norm": 98.39705123901368,
      "invalid_tail_samples": 26.5,
      "learning_rate": 2.797087515733478e-05,
      "loss": 0.0244,
      "positive_teacher_pairs": 26227.0,
      "query_h_norm_ratio": 0.42024131715297697,
      "relation_loss": 0.0027192543260753156,
      "selected_pairs": 26227.0,
      "step": 19890,
      "student_distance_mean": 23.80873966217041,
      "teacher_distance_mean": 0.8617952525615692,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.0027192543260753156,
      "weighted_relation_loss": 0.0027192543260753156
    },
    {
      "action_query_batch_std": 0.3601096898317337,
      "action_query_norm": 41.358743286132814,
      "action_relation_loss": 0.002565549430437386,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.6155285313377,
      "flow_matching_loss": 0.022372598946094512,
      "grad_norm": 0.09469490498304367,
      "h_stock_valid_norm": 98.19267425537109,
      "invalid_tail_samples": 22.4,
      "learning_rate": 2.7921410554172723e-05,
      "loss": 0.0249,
      "positive_teacher_pairs": 27180.8,
      "query_h_norm_ratio": 0.42120074927806855,
      "relation_loss": 0.002565549430437386,
      "selected_pairs": 27180.8,
      "step": 19900,
      "student_distance_mean": 23.75250949859619,
      "teacher_distance_mean": 0.8594145834445953,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.002565549430437386,
      "weighted_relation_loss": 0.002565549430437386
    },
    {
      "action_query_batch_std": 0.3588172882795334,
      "action_query_norm": 41.337937927246095,
      "action_relation_loss": 0.0026836950331926347,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.62488306828812,
      "flow_matching_loss": 0.02397295366972685,
      "grad_norm": 0.11455629020929337,
      "h_stock_valid_norm": 98.43224411010742,
      "invalid_tail_samples": 24.6,
      "learning_rate": 2.787197277857871e-05,
      "loss": 0.0267,
      "positive_teacher_pairs": 26676.2,
      "query_h_norm_ratio": 0.4199655920267105,
      "relation_loss": 0.0026836950331926347,
      "selected_pairs": 26676.2,
      "step": 19910,
      "student_distance_mean": 23.69338493347168,
      "teacher_distance_mean": 0.8576735079288482,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0026836950331926347,
      "weighted_relation_loss": 0.0026836950331926347
    },
    {
      "action_query_batch_std": 0.3556972563266754,
      "action_query_norm": 41.332268142700194,
      "action_relation_loss": 0.0026428702520206572,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.63423760523854,
      "flow_matching_loss": 0.022447689063847066,
      "grad_norm": 0.11615438014268875,
      "h_stock_valid_norm": 98.51091842651367,
      "invalid_tail_samples": 23.2,
      "learning_rate": 2.782256189062429e-05,
      "loss": 0.0251,
      "positive_teacher_pairs": 26990.8,
      "query_h_norm_ratio": 0.4195722460746765,
      "relation_loss": 0.0026428702520206572,
      "selected_pairs": 26990.8,
      "step": 19920,
      "student_distance_mean": 23.53085117340088,
      "teacher_distance_mean": 0.8463195502758026,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0026428702520206572,
      "weighted_relation_loss": 0.0026428702520206572
    },
    {
      "action_query_batch_std": 0.3558430701494217,
      "action_query_norm": 41.31753768920898,
      "action_relation_loss": 0.0025974272517487406,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.64359214218896,
      "flow_matching_loss": 0.02264396958053112,
      "grad_norm": 0.09928220510482788,
      "h_stock_valid_norm": 98.58549346923829,
      "invalid_tail_samples": 23.1,
      "learning_rate": 2.777317795034839e-05,
      "loss": 0.0252,
      "positive_teacher_pairs": 27018.5,
      "query_h_norm_ratio": 0.4191065490245819,
      "relation_loss": 0.0025974272517487406,
      "selected_pairs": 27018.5,
      "step": 19930,
      "student_distance_mean": 23.573782920837402,
      "teacher_distance_mean": 0.8496235430240631,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0025974272517487406,
      "weighted_relation_loss": 0.0025974272517487406
    },
    {
      "action_query_batch_std": 0.3589083611965179,
      "action_query_norm": 41.333648681640625,
      "action_relation_loss": 0.0024494854966178536,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.652946679139383,
      "flow_matching_loss": 0.022612078487873076,
      "grad_norm": 0.08707337826490402,
      "h_stock_valid_norm": 98.5119873046875,
      "invalid_tail_samples": 21.2,
      "learning_rate": 2.7723821017757112e-05,
      "loss": 0.0251,
      "positive_teacher_pairs": 27455.1,
      "query_h_norm_ratio": 0.41958253681659696,
      "relation_loss": 0.0024494854966178536,
      "selected_pairs": 27455.1,
      "step": 19940,
      "student_distance_mean": 23.622442054748536,
      "teacher_distance_mean": 0.8568404614925385,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.0024494854966178536,
      "weighted_relation_loss": 0.0024494854966178536
    },
    {
      "action_query_batch_std": 0.3624399662017822,
      "action_query_norm": 41.34843101501465,
      "action_relation_loss": 0.0026338213589042424,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.662301216089805,
      "flow_matching_loss": 0.023172469437122346,
      "grad_norm": 0.16246260702610016,
      "h_stock_valid_norm": 98.6130386352539,
      "invalid_tail_samples": 24.2,
      "learning_rate": 2.7674491152823822e-05,
      "loss": 0.0258,
      "positive_teacher_pairs": 26758.1,
      "query_h_norm_ratio": 0.4193021863698959,
      "relation_loss": 0.0026338213589042424,
      "selected_pairs": 26758.1,
      "step": 19950,
      "student_distance_mean": 23.924188232421876,
      "teacher_distance_mean": 0.8664376616477967,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0026338213589042424,
      "weighted_relation_loss": 0.0026338213589042424
    },
    {
      "action_query_batch_std": 0.3548862159252167,
      "action_query_norm": 41.32048797607422,
      "action_relation_loss": 0.0026710324687883257,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.671655753040223,
      "flow_matching_loss": 0.020617880672216416,
      "grad_norm": 0.09338665753602982,
      "h_stock_valid_norm": 98.70869140625,
      "invalid_tail_samples": 23.0,
      "learning_rate": 2.7625188415488944e-05,
      "loss": 0.0233,
      "positive_teacher_pairs": 27031.6,
      "query_h_norm_ratio": 0.41861594319343565,
      "relation_loss": 0.0026710324687883257,
      "selected_pairs": 27031.6,
      "step": 19960,
      "student_distance_mean": 23.41797637939453,
      "teacher_distance_mean": 0.8465493261814118,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0026710324687883257,
      "weighted_relation_loss": 0.0026710324687883257
    },
    {
      "action_query_batch_std": 0.36139225363731386,
      "action_query_norm": 41.331091690063474,
      "action_relation_loss": 0.0026864356826990844,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.681010289990645,
      "flow_matching_loss": 0.0224954454228282,
      "grad_norm": 0.09438344836235046,
      "h_stock_valid_norm": 98.38487243652344,
      "invalid_tail_samples": 25.4,
      "learning_rate": 2.7575912865659924e-05,
      "loss": 0.0252,
      "positive_teacher_pairs": 26490.1,
      "query_h_norm_ratio": 0.4200985640287399,
      "relation_loss": 0.0026864356826990844,
      "selected_pairs": 26490.1,
      "step": 19970,
      "student_distance_mean": 23.93757095336914,
      "teacher_distance_mean": 0.8656992077827453,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0026864356826990844,
      "weighted_relation_loss": 0.0026864356826990844
    },
    {
      "action_query_batch_std": 0.35901489555835725,
      "action_query_norm": 41.326022720336915,
      "action_relation_loss": 0.002703201910480857,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.690364826941067,
      "flow_matching_loss": 0.02355414368212223,
      "grad_norm": 0.09248378872871399,
      "h_stock_valid_norm": 98.212451171875,
      "invalid_tail_samples": 25.0,
      "learning_rate": 2.7526664563211245e-05,
      "loss": 0.0263,
      "positive_teacher_pairs": 26576.1,
      "query_h_norm_ratio": 0.42078384160995486,
      "relation_loss": 0.002703201910480857,
      "selected_pairs": 26576.1,
      "step": 19980,
      "student_distance_mean": 23.78921604156494,
      "teacher_distance_mean": 0.8626927077770233,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.002703201910480857,
      "weighted_relation_loss": 0.002703201910480857
    },
    {
      "action_query_batch_std": 0.3530276447534561,
      "action_query_norm": 41.28097076416016,
      "action_relation_loss": 0.0025645513320341705,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.699719363891486,
      "flow_matching_loss": 0.022856391221284866,
      "grad_norm": 0.10475710034370422,
      "h_stock_valid_norm": 98.2621223449707,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.7477443567984224e-05,
      "loss": 0.0254,
      "positive_teacher_pairs": 26978.5,
      "query_h_norm_ratio": 0.42011313438415526,
      "relation_loss": 0.0025645513320341705,
      "selected_pairs": 26978.5,
      "step": 19990,
      "student_distance_mean": 23.387526321411134,
      "teacher_distance_mean": 0.8480072379112243,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0025645513320341705,
      "weighted_relation_loss": 0.0025645513320341705
    },
    {
      "action_query_batch_std": 0.3538029372692108,
      "action_query_norm": 41.311539840698245,
      "action_relation_loss": 0.002594932587817311,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.709073900841908,
      "flow_matching_loss": 0.022114622034132482,
      "grad_norm": 0.10049362480640411,
      "h_stock_valid_norm": 98.12619476318359,
      "invalid_tail_samples": 21.5,
      "learning_rate": 2.7428249939787e-05,
      "loss": 0.0247,
      "positive_teacher_pairs": 27381.1,
      "query_h_norm_ratio": 0.42100509703159333,
      "relation_loss": 0.002594932587817311,
      "selected_pairs": 27381.1,
      "step": 20000,
      "student_distance_mean": 23.29423427581787,
      "teacher_distance_mean": 0.8440041184425354,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.002594932587817311,
      "weighted_relation_loss": 0.002594932587817311
    },
    {
      "action_query_batch_std": 0.3566773861646652,
      "action_query_norm": 41.31823387145996,
      "action_relation_loss": 0.0027287133969366552,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.71842843779233,
      "flow_matching_loss": 0.021774382703006266,
      "grad_norm": 0.11307524889707565,
      "h_stock_valid_norm": 98.0926124572754,
      "invalid_tail_samples": 21.0,
      "learning_rate": 2.7379083738394484e-05,
      "loss": 0.0245,
      "positive_teacher_pairs": 27512.7,
      "query_h_norm_ratio": 0.42121856808662417,
      "relation_loss": 0.0027287133969366552,
      "selected_pairs": 27512.7,
      "step": 20010,
      "student_distance_mean": 23.45247287750244,
      "teacher_distance_mean": 0.8519261658191681,
      "valid_action_samples": 235.0,
      "weighted_auxiliary_loss": 0.0027287133969366552,
      "weighted_relation_loss": 0.0027287133969366552
    },
    {
      "action_query_batch_std": 0.35241506397724154,
      "action_query_norm": 41.28863792419433,
      "action_relation_loss": 0.00257877588737756,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.72778297474275,
      "flow_matching_loss": 0.023308438993990422,
      "grad_norm": 0.15201658010482788,
      "h_stock_valid_norm": 98.1703987121582,
      "invalid_tail_samples": 24.6,
      "learning_rate": 2.732994502354823e-05,
      "loss": 0.0259,
      "positive_teacher_pairs": 26662.0,
      "query_h_norm_ratio": 0.420582640171051,
      "relation_loss": 0.00257877588737756,
      "selected_pairs": 26662.0,
      "step": 20020,
      "student_distance_mean": 23.294841003417968,
      "teacher_distance_mean": 0.8431577920913697,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.00257877588737756,
      "weighted_relation_loss": 0.00257877588737756
    },
    {
      "action_query_batch_std": 0.3569696694612503,
      "action_query_norm": 41.29378814697266,
      "action_relation_loss": 0.0027113237651064993,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.73713751169317,
      "flow_matching_loss": 0.02312593311071396,
      "grad_norm": 0.13923393189907074,
      "h_stock_valid_norm": 98.18006820678711,
      "invalid_tail_samples": 25.0,
      "learning_rate": 2.72808338549564e-05,
      "loss": 0.0258,
      "positive_teacher_pairs": 26573.9,
      "query_h_norm_ratio": 0.4205964863300323,
      "relation_loss": 0.0027113237651064993,
      "selected_pairs": 26573.9,
      "step": 20030,
      "student_distance_mean": 23.60538806915283,
      "teacher_distance_mean": 0.8563238024711609,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0027113237651064993,
      "weighted_relation_loss": 0.0027113237651064993
    },
    {
      "action_query_batch_std": 0.3544674217700958,
      "action_query_norm": 41.28792266845703,
      "action_relation_loss": 0.0027383025968447327,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.746492048643592,
      "flow_matching_loss": 0.021011458523571492,
      "grad_norm": 0.08662734925746918,
      "h_stock_valid_norm": 98.47890853881836,
      "invalid_tail_samples": 23.8,
      "learning_rate": 2.723175029229374e-05,
      "loss": 0.0237,
      "positive_teacher_pairs": 26844.8,
      "query_h_norm_ratio": 0.4192588746547699,
      "relation_loss": 0.0027383025968447327,
      "selected_pairs": 26844.8,
      "step": 20040,
      "student_distance_mean": 23.37554168701172,
      "teacher_distance_mean": 0.8492612302303314,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0027383025968447327,
      "weighted_relation_loss": 0.0027383025968447327
    },
    {
      "action_query_batch_std": 0.35559056997299193,
      "action_query_norm": 41.30058250427246,
      "action_relation_loss": 0.002588012535125017,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.755846585594014,
      "flow_matching_loss": 0.02147888783365488,
      "grad_norm": 0.12603752315044403,
      "h_stock_valid_norm": 98.51394729614258,
      "invalid_tail_samples": 23.7,
      "learning_rate": 2.718269439520138e-05,
      "loss": 0.0241,
      "positive_teacher_pairs": 26881.7,
      "query_h_norm_ratio": 0.41924052238464354,
      "relation_loss": 0.002588012535125017,
      "selected_pairs": 26881.7,
      "step": 20050,
      "student_distance_mean": 23.47619762420654,
      "teacher_distance_mean": 0.8476019024848938,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.002588012535125017,
      "weighted_relation_loss": 0.002588012535125017
    },
    {
      "action_query_batch_std": 0.3578022688627243,
      "action_query_norm": 41.31504096984863,
      "action_relation_loss": 0.002556099742650986,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.765201122544433,
      "flow_matching_loss": 0.021850184351205826,
      "grad_norm": 0.14527001976966858,
      "h_stock_valid_norm": 98.11121520996093,
      "invalid_tail_samples": 27.5,
      "learning_rate": 2.7133666223286858e-05,
      "loss": 0.0244,
      "positive_teacher_pairs": 26004.6,
      "query_h_norm_ratio": 0.4211046814918518,
      "relation_loss": 0.002556099742650986,
      "selected_pairs": 26004.6,
      "step": 20060,
      "student_distance_mean": 23.689273643493653,
      "teacher_distance_mean": 0.8590044736862182,
      "valid_action_samples": 228.5,
      "weighted_auxiliary_loss": 0.002556099742650986,
      "weighted_relation_loss": 0.002556099742650986
    },
    {
      "action_query_batch_std": 0.352472311258316,
      "action_query_norm": 41.26922950744629,
      "action_relation_loss": 0.002574447588995099,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.774555659494855,
      "flow_matching_loss": 0.02150556556880474,
      "grad_norm": 0.1501600742340088,
      "h_stock_valid_norm": 97.56019897460938,
      "invalid_tail_samples": 22.3,
      "learning_rate": 2.7084665836124006e-05,
      "loss": 0.0241,
      "positive_teacher_pairs": 27195.3,
      "query_h_norm_ratio": 0.42301576435565946,
      "relation_loss": 0.002574447588995099,
      "selected_pairs": 27195.3,
      "step": 20070,
      "student_distance_mean": 23.26881332397461,
      "teacher_distance_mean": 0.8443916440010071,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.002574447588995099,
      "weighted_relation_loss": 0.002574447588995099
    },
    {
      "action_query_batch_std": 0.35791846811771394,
      "action_query_norm": 41.305016708374026,
      "action_relation_loss": 0.002529076603241265,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.783910196445277,
      "flow_matching_loss": 0.020812257006764413,
      "grad_norm": 0.0885283574461937,
      "h_stock_valid_norm": 97.23640518188476,
      "invalid_tail_samples": 21.8,
      "learning_rate": 2.703569329325296e-05,
      "loss": 0.0233,
      "positive_teacher_pairs": 27321.4,
      "query_h_norm_ratio": 0.42479305863380434,
      "relation_loss": 0.002529076603241265,
      "selected_pairs": 27321.4,
      "step": 20080,
      "student_distance_mean": 23.650253295898438,
      "teacher_distance_mean": 0.8572917997837066,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.002529076603241265,
      "weighted_relation_loss": 0.002529076603241265
    },
    {
      "action_query_batch_std": 0.3587322264909744,
      "action_query_norm": 41.299289321899415,
      "action_relation_loss": 0.0025678796926513316,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.7932647333957,
      "flow_matching_loss": 0.02116076145321131,
      "grad_norm": 0.1210450604557991,
      "h_stock_valid_norm": 97.24338455200196,
      "invalid_tail_samples": 23.1,
      "learning_rate": 2.698674865417994e-05,
      "loss": 0.0237,
      "positive_teacher_pairs": 27007.5,
      "query_h_norm_ratio": 0.42470415234565734,
      "relation_loss": 0.0025678796926513316,
      "selected_pairs": 27007.5,
      "step": 20090,
      "student_distance_mean": 23.566673851013185,
      "teacher_distance_mean": 0.8556505441665649,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0025678796926513316,
      "weighted_relation_loss": 0.0025678796926513316
    },
    {
      "action_query_batch_std": 0.35932793021202086,
      "action_query_norm": 41.312451171875,
      "action_relation_loss": 0.0025998976780101657,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.802619270346117,
      "flow_matching_loss": 0.022234075888991355,
      "grad_norm": 0.10218685865402222,
      "h_stock_valid_norm": 98.02817611694336,
      "invalid_tail_samples": 25.1,
      "learning_rate": 2.6937831978377325e-05,
      "loss": 0.0248,
      "positive_teacher_pairs": 26550.1,
      "query_h_norm_ratio": 0.42143902778625486,
      "relation_loss": 0.0025998976780101657,
      "selected_pairs": 26550.1,
      "step": 20100,
      "student_distance_mean": 23.73218822479248,
      "teacher_distance_mean": 0.8589517951011658,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0025998976780101657,
      "weighted_relation_loss": 0.0025998976780101657
    },
    {
      "action_query_batch_std": 0.35191107392311094,
      "action_query_norm": 41.25356330871582,
      "action_relation_loss": 0.002699529938399792,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.81197380729654,
      "flow_matching_loss": 0.022117330692708492,
      "grad_norm": 0.09517297148704529,
      "h_stock_valid_norm": 98.18150939941407,
      "invalid_tail_samples": 23.8,
      "learning_rate": 2.6888943325283485e-05,
      "loss": 0.0248,
      "positive_teacher_pairs": 26858.8,
      "query_h_norm_ratio": 0.4201792001724243,
      "relation_loss": 0.002699529938399792,
      "selected_pairs": 26858.8,
      "step": 20110,
      "student_distance_mean": 23.285000801086426,
      "teacher_distance_mean": 0.8447826206684113,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.002699529938399792,
      "weighted_relation_loss": 0.002699529938399792
    },
    {
      "action_query_batch_std": 0.3544351369142532,
      "action_query_norm": 41.26200180053711,
      "action_relation_loss": 0.002662289282307029,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.82132834424696,
      "flow_matching_loss": 0.023394573107361795,
      "grad_norm": 0.10924913734197617,
      "h_stock_valid_norm": 97.86902236938477,
      "invalid_tail_samples": 23.7,
      "learning_rate": 2.6840082754302733e-05,
      "loss": 0.0261,
      "positive_teacher_pairs": 26878.1,
      "query_h_norm_ratio": 0.4216076284646988,
      "relation_loss": 0.002662289282307029,
      "selected_pairs": 26878.1,
      "step": 20120,
      "student_distance_mean": 23.381339073181152,
      "teacher_distance_mean": 0.8463830411434173,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.002662289282307029,
      "weighted_relation_loss": 0.002662289282307029
    },
    {
      "action_query_batch_std": 0.35599756240844727,
      "action_query_norm": 41.27452507019043,
      "action_relation_loss": 0.0027099279453977943,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.83068288119738,
      "flow_matching_loss": 0.022113875858485697,
      "grad_norm": 0.10944023728370667,
      "h_stock_valid_norm": 97.5851821899414,
      "invalid_tail_samples": 24.4,
      "learning_rate": 2.679125032480525e-05,
      "loss": 0.0248,
      "positive_teacher_pairs": 26710.1,
      "query_h_norm_ratio": 0.42296143174171447,
      "relation_loss": 0.0027099279453977943,
      "selected_pairs": 26710.1,
      "step": 20130,
      "student_distance_mean": 23.53398208618164,
      "teacher_distance_mean": 0.8528085052967072,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0027099279453977943,
      "weighted_relation_loss": 0.0027099279453977943
    },
    {
      "action_query_batch_std": 0.36017599105834963,
      "action_query_norm": 41.32020111083985,
      "action_relation_loss": 0.002532225660979748,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.8400374181478,
      "flow_matching_loss": 0.021820996701717377,
      "grad_norm": 0.1324475258588791,
      "h_stock_valid_norm": 97.56577911376954,
      "invalid_tail_samples": 22.5,
      "learning_rate": 2.674244609612708e-05,
      "loss": 0.0244,
      "positive_teacher_pairs": 27152.8,
      "query_h_norm_ratio": 0.42351568043231963,
      "relation_loss": 0.002532225660979748,
      "selected_pairs": 27152.8,
      "step": 20140,
      "student_distance_mean": 23.748859214782716,
      "teacher_distance_mean": 0.8606077253818512,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.002532225660979748,
      "weighted_relation_loss": 0.002532225660979748
    },
    {
      "action_query_batch_std": 0.35994523763656616,
      "action_query_norm": 41.309869384765626,
      "action_relation_loss": 0.0025120690697804093,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.849391955098223,
      "flow_matching_loss": 0.020228963159024716,
      "grad_norm": 0.10119267553091049,
      "h_stock_valid_norm": 98.07919692993164,
      "invalid_tail_samples": 23.1,
      "learning_rate": 2.6693670127569958e-05,
      "loss": 0.0227,
      "positive_teacher_pairs": 27010.9,
      "query_h_norm_ratio": 0.4211921840906143,
      "relation_loss": 0.0025120690697804093,
      "selected_pairs": 27010.9,
      "step": 20150,
      "student_distance_mean": 23.676639747619628,
      "teacher_distance_mean": 0.8550749301910401,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0025120690697804093,
      "weighted_relation_loss": 0.0025120690697804093
    },
    {
      "action_query_batch_std": 0.3554902315139771,
      "action_query_norm": 41.27087936401367,
      "action_relation_loss": 0.0028360530268400908,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.858746492048645,
      "flow_matching_loss": 0.021017466485500336,
      "grad_norm": 0.10230797529220581,
      "h_stock_valid_norm": 98.51671981811523,
      "invalid_tail_samples": 25.7,
      "learning_rate": 2.6644922478401268e-05,
      "loss": 0.0239,
      "positive_teacher_pairs": 26413.8,
      "query_h_norm_ratio": 0.41892354786396024,
      "relation_loss": 0.0028360530268400908,
      "selected_pairs": 26413.8,
      "step": 20160,
      "student_distance_mean": 23.46788501739502,
      "teacher_distance_mean": 0.847030633687973,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0028360530268400908,
      "weighted_relation_loss": 0.0028360530268400908
    },
    {
      "action_query_batch_std": 0.36090241372585297,
      "action_query_norm": 41.321572875976564,
      "action_relation_loss": 0.002509630867280066,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.868101028999064,
      "flow_matching_loss": 0.022499901801347734,
      "grad_norm": 0.09739919006824493,
      "h_stock_valid_norm": 98.6750259399414,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.6596203207854004e-05,
      "loss": 0.025,
      "positive_teacher_pairs": 26964.3,
      "query_h_norm_ratio": 0.41876574158668517,
      "relation_loss": 0.002509630867280066,
      "selected_pairs": 26964.3,
      "step": 20170,
      "student_distance_mean": 23.733274841308592,
      "teacher_distance_mean": 0.858119684457779,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.002509630867280066,
      "weighted_relation_loss": 0.002509630867280066
    },
    {
      "action_query_batch_std": 0.3578417718410492,
      "action_query_norm": 41.26668243408203,
      "action_relation_loss": 0.0026297467993572354,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.877455565949486,
      "flow_matching_loss": 0.0219601234421134,
      "grad_norm": 0.09465968608856201,
      "h_stock_valid_norm": 98.24654998779297,
      "invalid_tail_samples": 24.2,
      "learning_rate": 2.654751237512666e-05,
      "loss": 0.0246,
      "positive_teacher_pairs": 26754.7,
      "query_h_norm_ratio": 0.42003474533557894,
      "relation_loss": 0.0026297467993572354,
      "selected_pairs": 26754.7,
      "step": 20180,
      "student_distance_mean": 23.605106353759766,
      "teacher_distance_mean": 0.8584266543388367,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0026297467993572354,
      "weighted_relation_loss": 0.0026297467993572354
    },
    {
      "action_query_batch_std": 0.3563786864280701,
      "action_query_norm": 41.2621452331543,
      "action_relation_loss": 0.0028084066230803727,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.886810102899908,
      "flow_matching_loss": 0.02339555248618126,
      "grad_norm": 0.12599702179431915,
      "h_stock_valid_norm": 97.82634201049805,
      "invalid_tail_samples": 23.1,
      "learning_rate": 2.649885003938323e-05,
      "loss": 0.0262,
      "positive_teacher_pairs": 27019.8,
      "query_h_norm_ratio": 0.4217956095933914,
      "relation_loss": 0.0028084066230803727,
      "selected_pairs": 27019.8,
      "step": 20190,
      "student_distance_mean": 23.54237804412842,
      "teacher_distance_mean": 0.857443916797638,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0028084066230803727,
      "weighted_relation_loss": 0.0028084066230803727
    },
    {
      "action_query_batch_std": 0.3610017538070679,
      "action_query_norm": 41.322607803344724,
      "action_relation_loss": 0.002540275268256664,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.896164639850326,
      "flow_matching_loss": 0.02133347503840923,
      "grad_norm": 0.10769540071487427,
      "h_stock_valid_norm": 97.97584686279296,
      "invalid_tail_samples": 25.6,
      "learning_rate": 2.6450216259753e-05,
      "loss": 0.0239,
      "positive_teacher_pairs": 26439.1,
      "query_h_norm_ratio": 0.42176465690135956,
      "relation_loss": 0.002540275268256664,
      "selected_pairs": 26439.1,
      "step": 20200,
      "student_distance_mean": 23.808857917785645,
      "teacher_distance_mean": 0.8612808465957642,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.002540275268256664,
      "weighted_relation_loss": 0.002540275268256664
    },
    {
      "action_query_batch_std": 0.3575089037418365,
      "action_query_norm": 41.29107856750488,
      "action_relation_loss": 0.0026153221493586897,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.90551917680075,
      "flow_matching_loss": 0.020344369672238828,
      "grad_norm": 0.08517381548881531,
      "h_stock_valid_norm": 98.04021759033203,
      "invalid_tail_samples": 22.3,
      "learning_rate": 2.640161109533063e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 27197.3,
      "query_h_norm_ratio": 0.42117137610912325,
      "relation_loss": 0.0026153221493586897,
      "selected_pairs": 27197.3,
      "step": 20210,
      "student_distance_mean": 23.477422523498536,
      "teacher_distance_mean": 0.851829570531845,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0026153221493586897,
      "weighted_relation_loss": 0.0026153221493586897
    },
    {
      "action_query_batch_std": 0.3552309960126877,
      "action_query_norm": 41.26386070251465,
      "action_relation_loss": 0.002639248175546527,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.91487371375117,
      "flow_matching_loss": 0.022640109807252885,
      "grad_norm": 0.09779194742441177,
      "h_stock_valid_norm": 98.12931289672852,
      "invalid_tail_samples": 23.5,
      "learning_rate": 2.6353034605175936e-05,
      "loss": 0.0253,
      "positive_teacher_pairs": 26914.6,
      "query_h_norm_ratio": 0.42051029205322266,
      "relation_loss": 0.002639248175546527,
      "selected_pairs": 26914.6,
      "step": 20220,
      "student_distance_mean": 23.521957778930663,
      "teacher_distance_mean": 0.8522580564022064,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.002639248175546527,
      "weighted_relation_loss": 0.002639248175546527
    },
    {
      "action_query_batch_std": 0.35662115514278414,
      "action_query_norm": 41.27950820922852,
      "action_relation_loss": 0.0025921664666384457,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.92422825070159,
      "flow_matching_loss": 0.025185782089829446,
      "grad_norm": 0.09681390225887299,
      "h_stock_valid_norm": 98.31827239990234,
      "invalid_tail_samples": 24.4,
      "learning_rate": 2.630448684831398e-05,
      "loss": 0.0278,
      "positive_teacher_pairs": 26709.7,
      "query_h_norm_ratio": 0.4198582112789154,
      "relation_loss": 0.0025921664666384457,
      "selected_pairs": 26709.7,
      "step": 20230,
      "student_distance_mean": 23.600069427490233,
      "teacher_distance_mean": 0.8548310935497284,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0025921664666384457,
      "weighted_relation_loss": 0.0025921664666384457
    },
    {
      "action_query_batch_std": 0.35298886299133303,
      "action_query_norm": 41.2479923248291,
      "action_relation_loss": 0.0025998949771746994,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.93358278765201,
      "flow_matching_loss": 0.02395041212439537,
      "grad_norm": 0.1484222412109375,
      "h_stock_valid_norm": 98.09399108886718,
      "invalid_tail_samples": 22.9,
      "learning_rate": 2.625596788373482e-05,
      "loss": 0.0266,
      "positive_teacher_pairs": 27075.3,
      "query_h_norm_ratio": 0.4204967349767685,
      "relation_loss": 0.0025998949771746994,
      "selected_pairs": 27075.3,
      "step": 20240,
      "student_distance_mean": 23.260311698913576,
      "teacher_distance_mean": 0.8414458096027374,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0025998949771746994,
      "weighted_relation_loss": 0.0025998949771746994
    },
    {
      "action_query_batch_std": 0.35982379913330076,
      "action_query_norm": 41.3127197265625,
      "action_relation_loss": 0.0025930524338036776,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.942937324602433,
      "flow_matching_loss": 0.023354624211788178,
      "grad_norm": 0.12439883500337601,
      "h_stock_valid_norm": 97.92951583862305,
      "invalid_tail_samples": 23.2,
      "learning_rate": 2.620747777039363e-05,
      "loss": 0.0259,
      "positive_teacher_pairs": 26990.8,
      "query_h_norm_ratio": 0.4218627393245697,
      "relation_loss": 0.0025930524338036776,
      "selected_pairs": 26990.8,
      "step": 20250,
      "student_distance_mean": 23.690692901611328,
      "teacher_distance_mean": 0.8590226113796234,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0025930524338036776,
      "weighted_relation_loss": 0.0025930524338036776
    },
    {
      "action_query_batch_std": 0.3518933027982712,
      "action_query_norm": 41.21431999206543,
      "action_relation_loss": 0.002697371761314571,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.952291861552855,
      "flow_matching_loss": 0.02100256588310003,
      "grad_norm": 0.11496560275554657,
      "h_stock_valid_norm": 97.72486877441406,
      "invalid_tail_samples": 25.7,
      "learning_rate": 2.6159016567210424e-05,
      "loss": 0.0237,
      "positive_teacher_pairs": 26409.9,
      "query_h_norm_ratio": 0.42174471318721773,
      "relation_loss": 0.002697371761314571,
      "selected_pairs": 26409.9,
      "step": 20260,
      "student_distance_mean": 23.170933341979982,
      "teacher_distance_mean": 0.8397339403629303,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.002697371761314571,
      "weighted_relation_loss": 0.002697371761314571
    },
    {
      "action_query_batch_std": 0.3528324574232101,
      "action_query_norm": 41.220451736450194,
      "action_relation_loss": 0.0026683727279305457,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.961646398503273,
      "flow_matching_loss": 0.020706414710730314,
      "grad_norm": 0.08698040246963501,
      "h_stock_valid_norm": 97.86321792602538,
      "invalid_tail_samples": 26.3,
      "learning_rate": 2.6110584333070153e-05,
      "loss": 0.0234,
      "positive_teacher_pairs": 26270.4,
      "query_h_norm_ratio": 0.42120776176452634,
      "relation_loss": 0.0026683727279305457,
      "selected_pairs": 26270.4,
      "step": 20270,
      "student_distance_mean": 23.35675983428955,
      "teacher_distance_mean": 0.8463076233863831,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.0026683727279305457,
      "weighted_relation_loss": 0.0026683727279305457
    },
    {
      "action_query_batch_std": 0.359128350019455,
      "action_query_norm": 41.2543140411377,
      "action_relation_loss": 0.0026538181118667126,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.971000935453695,
      "flow_matching_loss": 0.021685274504125118,
      "grad_norm": 0.10271217674016953,
      "h_stock_valid_norm": 97.87563095092773,
      "invalid_tail_samples": 24.9,
      "learning_rate": 2.6062181126822537e-05,
      "loss": 0.0243,
      "positive_teacher_pairs": 26590.9,
      "query_h_norm_ratio": 0.4215039849281311,
      "relation_loss": 0.0026538181118667126,
      "selected_pairs": 26590.9,
      "step": 20280,
      "student_distance_mean": 23.604777145385743,
      "teacher_distance_mean": 0.8561300277709961,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0026538181118667126,
      "weighted_relation_loss": 0.0026538181118667126
    },
    {
      "action_query_batch_std": 0.3579549461603165,
      "action_query_norm": 41.2629581451416,
      "action_relation_loss": 0.00266910707578063,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.980355472404117,
      "flow_matching_loss": 0.023302202112972736,
      "grad_norm": 0.11394598335027695,
      "h_stock_valid_norm": 97.81619873046876,
      "invalid_tail_samples": 24.8,
      "learning_rate": 2.601380700728203e-05,
      "loss": 0.026,
      "positive_teacher_pairs": 26617.1,
      "query_h_norm_ratio": 0.4218463569879532,
      "relation_loss": 0.00266910707578063,
      "selected_pairs": 26617.1,
      "step": 20290,
      "student_distance_mean": 23.596531486511232,
      "teacher_distance_mean": 0.8568740427494049,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.00266910707578063,
      "weighted_relation_loss": 0.00266910707578063
    },
    {
      "action_query_batch_std": 0.3535026639699936,
      "action_query_norm": 41.24837417602539,
      "action_relation_loss": 0.0027196403359994294,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.989710009354535,
      "flow_matching_loss": 0.022274040244519712,
      "grad_norm": 0.09739185124635696,
      "h_stock_valid_norm": 97.6262596130371,
      "invalid_tail_samples": 21.9,
      "learning_rate": 2.596546203322777e-05,
      "loss": 0.025,
      "positive_teacher_pairs": 27290.2,
      "query_h_norm_ratio": 0.42251557409763335,
      "relation_loss": 0.0027196403359994294,
      "selected_pairs": 27290.2,
      "step": 20300,
      "student_distance_mean": 23.38591785430908,
      "teacher_distance_mean": 0.8454155027866364,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.0027196403359994294,
      "weighted_relation_loss": 0.0027196403359994294
    },
    {
      "action_query_batch_std": 0.35724384486675265,
      "action_query_norm": 41.279862594604495,
      "action_relation_loss": 0.00253694923594594,
      "auxiliary_loss_weight": 1.0,
      "epoch": 18.999064546304957,
      "flow_matching_loss": 0.020498932152986527,
      "grad_norm": 0.11789274960756302,
      "h_stock_valid_norm": 97.79358367919922,
      "invalid_tail_samples": 24.7,
      "learning_rate": 2.5917146263403457e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 26641.6,
      "query_h_norm_ratio": 0.42211476266384124,
      "relation_loss": 0.00253694923594594,
      "selected_pairs": 26641.6,
      "step": 20310,
      "student_distance_mean": 23.57979965209961,
      "teacher_distance_mean": 0.8523175418376923,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.00253694923594594,
      "weighted_relation_loss": 0.00253694923594594
    },
    {
      "action_query_batch_std": 0.3548440128564835,
      "action_query_norm": 41.252234649658206,
      "action_relation_loss": 0.0024218685692176225,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.00841908325538,
      "flow_matching_loss": 0.019953125901520253,
      "grad_norm": 0.12982134521007538,
      "h_stock_valid_norm": 97.28045806884765,
      "invalid_tail_samples": 20.1,
      "learning_rate": 2.5868859756517296e-05,
      "loss": 0.0224,
      "positive_teacher_pairs": 24783.1,
      "query_h_norm_ratio": 0.4240587830543518,
      "relation_loss": 0.0024218685692176225,
      "selected_pairs": 24783.1,
      "step": 20320,
      "student_distance_mean": 23.473668479919432,
      "teacher_distance_mean": 0.8545320928096771,
      "valid_action_samples": 216.0,
      "weighted_auxiliary_loss": 0.0024218685692176225,
      "weighted_relation_loss": 0.0024218685692176225
    },
    {
      "action_query_batch_std": 0.3538114219903946,
      "action_query_norm": 41.24662284851074,
      "action_relation_loss": 0.002596100652590394,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.0177736202058,
      "flow_matching_loss": 0.02224995233118534,
      "grad_norm": 0.1258336752653122,
      "h_stock_valid_norm": 97.60379638671876,
      "invalid_tail_samples": 23.6,
      "learning_rate": 2.5820602571241947e-05,
      "loss": 0.0248,
      "positive_teacher_pairs": 26899.7,
      "query_h_norm_ratio": 0.4225948989391327,
      "relation_loss": 0.002596100652590394,
      "selected_pairs": 26899.7,
      "step": 20330,
      "student_distance_mean": 23.38761329650879,
      "teacher_distance_mean": 0.8486927092075348,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.002596100652590394,
      "weighted_relation_loss": 0.002596100652590394
    },
    {
      "action_query_batch_std": 0.3551014721393585,
      "action_query_norm": 41.255902481079104,
      "action_relation_loss": 0.002543793083168566,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.02712815715622,
      "flow_matching_loss": 0.02162636537104845,
      "grad_norm": 0.11726074665784836,
      "h_stock_valid_norm": 98.04588470458984,
      "invalid_tail_samples": 22.5,
      "learning_rate": 2.577237476621442e-05,
      "loss": 0.0242,
      "positive_teacher_pairs": 27153.6,
      "query_h_norm_ratio": 0.4207884192466736,
      "relation_loss": 0.002543793083168566,
      "selected_pairs": 27153.6,
      "step": 20340,
      "student_distance_mean": 23.38265495300293,
      "teacher_distance_mean": 0.846642130613327,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.002543793083168566,
      "weighted_relation_loss": 0.002543793083168566
    },
    {
      "action_query_batch_std": 0.3614625036716461,
      "action_query_norm": 41.28997993469238,
      "action_relation_loss": 0.0024984056130051614,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.036482694106642,
      "flow_matching_loss": 0.021629124879837036,
      "grad_norm": 0.13205784559249878,
      "h_stock_valid_norm": 98.34067153930664,
      "invalid_tail_samples": 23.0,
      "learning_rate": 2.572417640003609e-05,
      "loss": 0.0241,
      "positive_teacher_pairs": 27034.5,
      "query_h_norm_ratio": 0.41987154483795164,
      "relation_loss": 0.0024984056130051614,
      "selected_pairs": 27034.5,
      "step": 20350,
      "student_distance_mean": 23.84988498687744,
      "teacher_distance_mean": 0.8649242460727692,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0024984056130051614,
      "weighted_relation_loss": 0.0024984056130051614
    },
    {
      "action_query_batch_std": 0.35311519503593447,
      "action_query_norm": 41.22582473754883,
      "action_relation_loss": 0.002444312535226345,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.045837231057064,
      "flow_matching_loss": 0.019631014205515385,
      "grad_norm": 0.09087485074996948,
      "h_stock_valid_norm": 98.58200454711914,
      "invalid_tail_samples": 25.8,
      "learning_rate": 2.5676007531272473e-05,
      "loss": 0.0221,
      "positive_teacher_pairs": 26386.6,
      "query_h_norm_ratio": 0.41818901896476746,
      "relation_loss": 0.002444312535226345,
      "selected_pairs": 26386.6,
      "step": 20360,
      "student_distance_mean": 23.32636661529541,
      "teacher_distance_mean": 0.8467443466186524,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.002444312535226345,
      "weighted_relation_loss": 0.002444312535226345
    },
    {
      "action_query_batch_std": 0.3533233106136322,
      "action_query_norm": 41.2310791015625,
      "action_relation_loss": 0.002494306582957506,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.055191768007482,
      "flow_matching_loss": 0.022003446333110334,
      "grad_norm": 0.09396729618310928,
      "h_stock_valid_norm": 98.45219192504882,
      "invalid_tail_samples": 25.3,
      "learning_rate": 2.5627868218453332e-05,
      "loss": 0.0245,
      "positive_teacher_pairs": 26512.9,
      "query_h_norm_ratio": 0.4187955021858215,
      "relation_loss": 0.002494306582957506,
      "selected_pairs": 26512.9,
      "step": 20370,
      "student_distance_mean": 23.342057991027833,
      "teacher_distance_mean": 0.8488598346710206,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.002494306582957506,
      "weighted_relation_loss": 0.002494306582957506
    },
    {
      "action_query_batch_std": 0.357212033867836,
      "action_query_norm": 41.24055976867676,
      "action_relation_loss": 0.002344426093623042,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.064546304957904,
      "flow_matching_loss": 0.020843381620943548,
      "grad_norm": 0.10336697101593018,
      "h_stock_valid_norm": 98.11948089599609,
      "invalid_tail_samples": 24.3,
      "learning_rate": 2.5579758520072443e-05,
      "loss": 0.0232,
      "positive_teacher_pairs": 26735.2,
      "query_h_norm_ratio": 0.42031204998493193,
      "relation_loss": 0.002344426093623042,
      "selected_pairs": 26735.2,
      "step": 20380,
      "student_distance_mean": 23.585442733764648,
      "teacher_distance_mean": 0.8581943571567535,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002344426093623042,
      "weighted_relation_loss": 0.002344426093623042
    },
    {
      "action_query_batch_std": 0.3598633587360382,
      "action_query_norm": 41.26972923278809,
      "action_relation_loss": 0.002513816230930388,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.073900841908326,
      "flow_matching_loss": 0.021763997711241247,
      "grad_norm": 0.1033959612250328,
      "h_stock_valid_norm": 97.52647476196289,
      "invalid_tail_samples": 24.3,
      "learning_rate": 2.5531678494587615e-05,
      "loss": 0.0243,
      "positive_teacher_pairs": 26735.0,
      "query_h_norm_ratio": 0.42316837310791017,
      "relation_loss": 0.002513816230930388,
      "selected_pairs": 26735.0,
      "step": 20390,
      "student_distance_mean": 23.77376194000244,
      "teacher_distance_mean": 0.8620216310024261,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002513816230930388,
      "weighted_relation_loss": 0.002513816230930388
    },
    {
      "action_query_batch_std": 0.35921033620834353,
      "action_query_norm": 41.268072509765624,
      "action_relation_loss": 0.002527760015800595,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.08325537885875,
      "flow_matching_loss": 0.020196139626204968,
      "grad_norm": 0.10059574991464615,
      "h_stock_valid_norm": 97.47602462768555,
      "invalid_tail_samples": 24.4,
      "learning_rate": 2.5483628200420644e-05,
      "loss": 0.0227,
      "positive_teacher_pairs": 26723.0,
      "query_h_norm_ratio": 0.4233698308467865,
      "relation_loss": 0.002527760015800595,
      "selected_pairs": 26723.0,
      "step": 20400,
      "student_distance_mean": 23.67998161315918,
      "teacher_distance_mean": 0.8568553626537323,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.002527760015800595,
      "weighted_relation_loss": 0.002527760015800595
    },
    {
      "action_query_batch_std": 0.3607629776000977,
      "action_query_norm": 41.27919235229492,
      "action_relation_loss": 0.0024028885876759888,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.092609915809167,
      "flow_matching_loss": 0.021199660375714302,
      "grad_norm": 0.10273286700248718,
      "h_stock_valid_norm": 97.81751098632813,
      "invalid_tail_samples": 23.6,
      "learning_rate": 2.5435607695957154e-05,
      "loss": 0.0236,
      "positive_teacher_pairs": 26894.4,
      "query_h_norm_ratio": 0.42200509309768675,
      "relation_loss": 0.0024028885876759888,
      "selected_pairs": 26894.4,
      "step": 20410,
      "student_distance_mean": 23.800944328308105,
      "teacher_distance_mean": 0.8607842147350311,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0024028885876759888,
      "weighted_relation_loss": 0.0024028885876759888
    },
    {
      "action_query_batch_std": 0.3581500858068466,
      "action_query_norm": 41.26313667297363,
      "action_relation_loss": 0.00244584025349468,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.10196445275959,
      "flow_matching_loss": 0.021410389617085458,
      "grad_norm": 0.10579556226730347,
      "h_stock_valid_norm": 98.27010803222656,
      "invalid_tail_samples": 21.8,
      "learning_rate": 2.5387617039546586e-05,
      "loss": 0.0239,
      "positive_teacher_pairs": 27322.6,
      "query_h_norm_ratio": 0.41989842653274534,
      "relation_loss": 0.00244584025349468,
      "selected_pairs": 27322.6,
      "step": 20420,
      "student_distance_mean": 23.666666412353514,
      "teacher_distance_mean": 0.8582909345626831,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.00244584025349468,
      "weighted_relation_loss": 0.00244584025349468
    },
    {
      "action_query_batch_std": 0.3583013266324997,
      "action_query_norm": 41.26915283203125,
      "action_relation_loss": 0.0024870823370292785,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.11131898971001,
      "flow_matching_loss": 0.02067771162837744,
      "grad_norm": 0.09564168751239777,
      "h_stock_valid_norm": 98.51641464233398,
      "invalid_tail_samples": 24.1,
      "learning_rate": 2.5339656289502106e-05,
      "loss": 0.0232,
      "positive_teacher_pairs": 26778.8,
      "query_h_norm_ratio": 0.4189101576805115,
      "relation_loss": 0.0024870823370292785,
      "selected_pairs": 26778.8,
      "step": 20430,
      "student_distance_mean": 23.69442653656006,
      "teacher_distance_mean": 0.8599696815013885,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0024870823370292785,
      "weighted_relation_loss": 0.0024870823370292785
    },
    {
      "action_query_batch_std": 0.3575382620096207,
      "action_query_norm": 41.239539337158206,
      "action_relation_loss": 0.002455548895522952,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.12067352666043,
      "flow_matching_loss": 0.02058957740664482,
      "grad_norm": 0.10344081372022629,
      "h_stock_valid_norm": 98.51489334106445,
      "invalid_tail_samples": 24.1,
      "learning_rate": 2.529172550410056e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 26789.0,
      "query_h_norm_ratio": 0.41861593127250674,
      "relation_loss": 0.002455548895522952,
      "selected_pairs": 26789.0,
      "step": 20440,
      "student_distance_mean": 23.510577392578124,
      "teacher_distance_mean": 0.8517818331718445,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.002455548895522952,
      "weighted_relation_loss": 0.002455548895522952
    },
    {
      "action_query_batch_std": 0.35590306520462034,
      "action_query_norm": 41.24114151000977,
      "action_relation_loss": 0.0025360036408528684,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.13002806361085,
      "flow_matching_loss": 0.021402212604880334,
      "grad_norm": 0.12161195278167725,
      "h_stock_valid_norm": 98.3881332397461,
      "invalid_tail_samples": 22.8,
      "learning_rate": 2.524382474158234e-05,
      "loss": 0.0239,
      "positive_teacher_pairs": 27079.8,
      "query_h_norm_ratio": 0.4191719502210617,
      "relation_loss": 0.0025360036408528684,
      "selected_pairs": 27079.8,
      "step": 20450,
      "student_distance_mean": 23.564591217041016,
      "teacher_distance_mean": 0.8519168078899384,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0025360036408528684,
      "weighted_relation_loss": 0.0025360036408528684
    },
    {
      "action_query_batch_std": 0.3570566654205322,
      "action_query_norm": 41.24278335571289,
      "action_relation_loss": 0.002513445424847305,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.139382600561273,
      "flow_matching_loss": 0.022675498761236666,
      "grad_norm": 0.1336217075586319,
      "h_stock_valid_norm": 98.50116958618165,
      "invalid_tail_samples": 23.6,
      "learning_rate": 2.5195954060151433e-05,
      "loss": 0.0252,
      "positive_teacher_pairs": 26898.7,
      "query_h_norm_ratio": 0.41870543658733367,
      "relation_loss": 0.002513445424847305,
      "selected_pairs": 26898.7,
      "step": 20460,
      "student_distance_mean": 23.54936866760254,
      "teacher_distance_mean": 0.8554010391235352,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.002513445424847305,
      "weighted_relation_loss": 0.002513445424847305
    },
    {
      "action_query_batch_std": 0.35748036205768585,
      "action_query_norm": 41.22652244567871,
      "action_relation_loss": 0.0026201332453638314,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.14873713751169,
      "flow_matching_loss": 0.021927267871797086,
      "grad_norm": 0.12321754544973373,
      "h_stock_valid_norm": 98.44405746459961,
      "invalid_tail_samples": 25.7,
      "learning_rate": 2.5148113517975212e-05,
      "loss": 0.0245,
      "positive_teacher_pairs": 26411.9,
      "query_h_norm_ratio": 0.4187853425741196,
      "relation_loss": 0.0026201332453638314,
      "selected_pairs": 26411.9,
      "step": 20470,
      "student_distance_mean": 23.6272834777832,
      "teacher_distance_mean": 0.8617434084415436,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0026201332453638314,
      "weighted_relation_loss": 0.0026201332453638314
    },
    {
      "action_query_batch_std": 0.35307380855083464,
      "action_query_norm": 41.20241928100586,
      "action_relation_loss": 0.0025300498818978667,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.158091674462113,
      "flow_matching_loss": 0.02214444875717163,
      "grad_norm": 0.09375987201929092,
      "h_stock_valid_norm": 98.20038223266602,
      "invalid_tail_samples": 22.9,
      "learning_rate": 2.5100303173184447e-05,
      "loss": 0.0247,
      "positive_teacher_pairs": 27056.8,
      "query_h_norm_ratio": 0.4195770412683487,
      "relation_loss": 0.0025300498818978667,
      "selected_pairs": 27056.8,
      "step": 20480,
      "student_distance_mean": 23.286230850219727,
      "teacher_distance_mean": 0.8485614359378815,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0025300498818978667,
      "weighted_relation_loss": 0.0025300498818978667
    },
    {
      "action_query_batch_std": 0.3560927927494049,
      "action_query_norm": 41.21014175415039,
      "action_relation_loss": 0.002583629940636456,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.167446211412535,
      "flow_matching_loss": 0.020370223466306923,
      "grad_norm": 0.09538672119379044,
      "h_stock_valid_norm": 98.09262466430664,
      "invalid_tail_samples": 26.4,
      "learning_rate": 2.5052523083873196e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 26247.3,
      "query_h_norm_ratio": 0.4201184004545212,
      "relation_loss": 0.002583629940636456,
      "selected_pairs": 26247.3,
      "step": 20490,
      "student_distance_mean": 23.489109230041503,
      "teacher_distance_mean": 0.8538908660411835,
      "valid_action_samples": 229.6,
      "weighted_auxiliary_loss": 0.002583629940636456,
      "weighted_relation_loss": 0.002583629940636456
    },
    {
      "action_query_batch_std": 0.35676389634609224,
      "action_query_norm": 41.23764305114746,
      "action_relation_loss": 0.0025232362328097223,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.176800748362957,
      "flow_matching_loss": 0.020841182582080363,
      "grad_norm": 0.09000909328460693,
      "h_stock_valid_norm": 98.12751998901368,
      "invalid_tail_samples": 24.9,
      "learning_rate": 2.500477330809881e-05,
      "loss": 0.0234,
      "positive_teacher_pairs": 26597.2,
      "query_h_norm_ratio": 0.4202464997768402,
      "relation_loss": 0.0025232362328097223,
      "selected_pairs": 26597.2,
      "step": 20500,
      "student_distance_mean": 23.618011665344238,
      "teacher_distance_mean": 0.8566208600997924,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0025232362328097223,
      "weighted_relation_loss": 0.0025232362328097223
    },
    {
      "action_query_batch_std": 0.35619641840457916,
      "action_query_norm": 41.24004783630371,
      "action_relation_loss": 0.0024292933056131007,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.186155285313376,
      "flow_matching_loss": 0.021556035801768302,
      "grad_norm": 0.1039961576461792,
      "h_stock_valid_norm": 98.20929718017578,
      "invalid_tail_samples": 24.7,
      "learning_rate": 2.4957053903881733e-05,
      "loss": 0.024,
      "positive_teacher_pairs": 26641.8,
      "query_h_norm_ratio": 0.4199216902256012,
      "relation_loss": 0.0024292933056131007,
      "selected_pairs": 26641.8,
      "step": 20510,
      "student_distance_mean": 23.584021186828615,
      "teacher_distance_mean": 0.8536025285720825,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0024292933056131007,
      "weighted_relation_loss": 0.0024292933056131007
    },
    {
      "action_query_batch_std": 0.36008617877960203,
      "action_query_norm": 41.21305885314941,
      "action_relation_loss": 0.0024797685444355013,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.195509822263798,
      "flow_matching_loss": 0.023159079626202582,
      "grad_norm": 0.09190710633993149,
      "h_stock_valid_norm": 98.46674880981445,
      "invalid_tail_samples": 25.5,
      "learning_rate": 2.4909364929205576e-05,
      "loss": 0.0256,
      "positive_teacher_pairs": 26461.8,
      "query_h_norm_ratio": 0.4185508966445923,
      "relation_loss": 0.0024797685444355013,
      "selected_pairs": 26461.8,
      "step": 20520,
      "student_distance_mean": 23.750569343566895,
      "teacher_distance_mean": 0.8640811979770661,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0024797685444355013,
      "weighted_relation_loss": 0.0024797685444355013
    },
    {
      "action_query_batch_std": 0.35693089067935946,
      "action_query_norm": 41.229858779907225,
      "action_relation_loss": 0.0024673404404893517,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.20486435921422,
      "flow_matching_loss": 0.020769105385988952,
      "grad_norm": 0.10478352010250092,
      "h_stock_valid_norm": 98.14714050292969,
      "invalid_tail_samples": 27.3,
      "learning_rate": 2.4861706442016923e-05,
      "loss": 0.0232,
      "positive_teacher_pairs": 26049.9,
      "query_h_norm_ratio": 0.42008598148822784,
      "relation_loss": 0.0024673404404893517,
      "selected_pairs": 26049.9,
      "step": 20530,
      "student_distance_mean": 23.578345489501952,
      "teacher_distance_mean": 0.8568997383117676,
      "valid_action_samples": 228.7,
      "weighted_auxiliary_loss": 0.0024673404404893517,
      "weighted_relation_loss": 0.0024673404404893517
    },
    {
      "action_query_batch_std": 0.3580689549446106,
      "action_query_norm": 41.21538543701172,
      "action_relation_loss": 0.002425667527131736,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.21421889616464,
      "flow_matching_loss": 0.021052449755370618,
      "grad_norm": 0.10474555939435959,
      "h_stock_valid_norm": 98.06524887084962,
      "invalid_tail_samples": 23.9,
      "learning_rate": 2.481407850022533e-05,
      "loss": 0.0235,
      "positive_teacher_pairs": 26840.1,
      "query_h_norm_ratio": 0.4202880471944809,
      "relation_loss": 0.002425667527131736,
      "selected_pairs": 26840.1,
      "step": 20540,
      "student_distance_mean": 23.478864860534667,
      "teacher_distance_mean": 0.8540429949760437,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.002425667527131736,
      "weighted_relation_loss": 0.002425667527131736
    },
    {
      "action_query_batch_std": 0.35693740248680117,
      "action_query_norm": 41.20273399353027,
      "action_relation_loss": 0.0025492121931165457,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.22357343311506,
      "flow_matching_loss": 0.022642349638044835,
      "grad_norm": 0.09852689504623413,
      "h_stock_valid_norm": 97.77341995239257,
      "invalid_tail_samples": 25.2,
      "learning_rate": 2.4766481161703213e-05,
      "loss": 0.0252,
      "positive_teacher_pairs": 26528.2,
      "query_h_norm_ratio": 0.4214148551225662,
      "relation_loss": 0.0025492121931165457,
      "selected_pairs": 26528.2,
      "step": 20550,
      "student_distance_mean": 23.532014656066895,
      "teacher_distance_mean": 0.8546782195568084,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0025492121931165457,
      "weighted_relation_loss": 0.0025492121931165457
    },
    {
      "action_query_batch_std": 0.3530094027519226,
      "action_query_norm": 41.19824447631836,
      "action_relation_loss": 0.002473701606504619,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.232927970065482,
      "flow_matching_loss": 0.020496219769120215,
      "grad_norm": 0.09458249062299728,
      "h_stock_valid_norm": 97.64965438842773,
      "invalid_tail_samples": 24.0,
      "learning_rate": 2.4718914484285876e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 26800.8,
      "query_h_norm_ratio": 0.42190084159374236,
      "relation_loss": 0.002473701606504619,
      "selected_pairs": 26800.8,
      "step": 20560,
      "student_distance_mean": 23.302256393432618,
      "teacher_distance_mean": 0.8462191581726074,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.002473701606504619,
      "weighted_relation_loss": 0.002473701606504619
    },
    {
      "action_query_batch_std": 0.35513125360012054,
      "action_query_norm": 41.20552406311035,
      "action_relation_loss": 0.0023907575057819487,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.242282507015904,
      "flow_matching_loss": 0.019792041555047036,
      "grad_norm": 0.12376553565263748,
      "h_stock_valid_norm": 97.33621292114258,
      "invalid_tail_samples": 24.1,
      "learning_rate": 2.467137852577129e-05,
      "loss": 0.0222,
      "positive_teacher_pairs": 26780.2,
      "query_h_norm_ratio": 0.4233327955007553,
      "relation_loss": 0.0023907575057819487,
      "selected_pairs": 26780.2,
      "step": 20570,
      "student_distance_mean": 23.386537551879883,
      "teacher_distance_mean": 0.8526998579502105,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0023907575057819487,
      "weighted_relation_loss": 0.0023907575057819487
    },
    {
      "action_query_batch_std": 0.35570707619190217,
      "action_query_norm": 41.198782730102536,
      "action_relation_loss": 0.002622132725082338,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.251637043966323,
      "flow_matching_loss": 0.021654252894222736,
      "grad_norm": 0.10534064471721649,
      "h_stock_valid_norm": 97.46932907104492,
      "invalid_tail_samples": 24.8,
      "learning_rate": 2.4623873343920123e-05,
      "loss": 0.0243,
      "positive_teacher_pairs": 26623.3,
      "query_h_norm_ratio": 0.42268656492233275,
      "relation_loss": 0.002622132725082338,
      "selected_pairs": 26623.3,
      "step": 20580,
      "student_distance_mean": 23.47716655731201,
      "teacher_distance_mean": 0.8544344186782837,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.002622132725082338,
      "weighted_relation_loss": 0.002622132725082338
    },
    {
      "action_query_batch_std": 0.3582139045000076,
      "action_query_norm": 41.2289379119873,
      "action_relation_loss": 0.002413375838659704,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.260991580916745,
      "flow_matching_loss": 0.021381580270826817,
      "grad_norm": 0.11530537903308868,
      "h_stock_valid_norm": 97.6081314086914,
      "invalid_tail_samples": 21.7,
      "learning_rate": 2.4576398996455656e-05,
      "loss": 0.0238,
      "positive_teacher_pairs": 27343.1,
      "query_h_norm_ratio": 0.42239580750465394,
      "relation_loss": 0.002413375838659704,
      "selected_pairs": 27343.1,
      "step": 20590,
      "student_distance_mean": 23.614050483703615,
      "teacher_distance_mean": 0.8579456984996796,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.002413375838659704,
      "weighted_relation_loss": 0.002413375838659704
    },
    {
      "action_query_batch_std": 0.3581290602684021,
      "action_query_norm": 41.20871086120606,
      "action_relation_loss": 0.0024913713103160264,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.270346117867167,
      "flow_matching_loss": 0.02253756280988455,
      "grad_norm": 0.10455089062452316,
      "h_stock_valid_norm": 97.82547760009766,
      "invalid_tail_samples": 20.8,
      "learning_rate": 2.452895554106368e-05,
      "loss": 0.025,
      "positive_teacher_pairs": 27548.9,
      "query_h_norm_ratio": 0.42125022411346436,
      "relation_loss": 0.0024913713103160264,
      "selected_pairs": 27548.9,
      "step": 20600,
      "student_distance_mean": 23.619594192504884,
      "teacher_distance_mean": 0.8588152766227722,
      "valid_action_samples": 235.2,
      "weighted_auxiliary_loss": 0.0024913713103160264,
      "weighted_relation_loss": 0.0024913713103160264
    },
    {
      "action_query_batch_std": 0.35505245327949525,
      "action_query_norm": 41.190606689453126,
      "action_relation_loss": 0.002605677116662264,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.279700654817585,
      "flow_matching_loss": 0.02185645271092653,
      "grad_norm": 0.10322257876396179,
      "h_stock_valid_norm": 97.7494140625,
      "invalid_tail_samples": 22.3,
      "learning_rate": 2.4481543035392506e-05,
      "loss": 0.0245,
      "positive_teacher_pairs": 27197.5,
      "query_h_norm_ratio": 0.42139383852481843,
      "relation_loss": 0.002605677116662264,
      "selected_pairs": 27197.5,
      "step": 20610,
      "student_distance_mean": 23.34589042663574,
      "teacher_distance_mean": 0.8495869576931,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.002605677116662264,
      "weighted_relation_loss": 0.002605677116662264
    },
    {
      "action_query_batch_std": 0.3501790136098862,
      "action_query_norm": 41.1531795501709,
      "action_relation_loss": 0.002590023004449904,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.289055191768007,
      "flow_matching_loss": 0.02107368828728795,
      "grad_norm": 0.1269165277481079,
      "h_stock_valid_norm": 97.45953903198242,
      "invalid_tail_samples": 25.0,
      "learning_rate": 2.4434161537052774e-05,
      "loss": 0.0237,
      "positive_teacher_pairs": 26575.2,
      "query_h_norm_ratio": 0.4222627431154251,
      "relation_loss": 0.002590023004449904,
      "selected_pairs": 26575.2,
      "step": 20620,
      "student_distance_mean": 23.138168144226075,
      "teacher_distance_mean": 0.8437453627586364,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.002590023004449904,
      "weighted_relation_loss": 0.002590023004449904
    },
    {
      "action_query_batch_std": 0.3559260666370392,
      "action_query_norm": 41.19982681274414,
      "action_relation_loss": 0.0025673114694654943,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.29840972871843,
      "flow_matching_loss": 0.02130030132830143,
      "grad_norm": 0.09840232878923416,
      "h_stock_valid_norm": 97.4428482055664,
      "invalid_tail_samples": 24.0,
      "learning_rate": 2.4386811103617474e-05,
      "loss": 0.0239,
      "positive_teacher_pairs": 26802.2,
      "query_h_norm_ratio": 0.4228130131959915,
      "relation_loss": 0.0025673114694654943,
      "selected_pairs": 26802.2,
      "step": 20630,
      "student_distance_mean": 23.314450454711913,
      "teacher_distance_mean": 0.8488773167133331,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0025673114694654943,
      "weighted_relation_loss": 0.0025673114694654943
    },
    {
      "action_query_batch_std": 0.35686067342758176,
      "action_query_norm": 41.20243606567383,
      "action_relation_loss": 0.0025915249017998574,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.30776426566885,
      "flow_matching_loss": 0.021452457644045353,
      "grad_norm": 0.10731368511915207,
      "h_stock_valid_norm": 97.58427581787109,
      "invalid_tail_samples": 22.8,
      "learning_rate": 2.4339491792621833e-05,
      "loss": 0.024,
      "positive_teacher_pairs": 27085.8,
      "query_h_norm_ratio": 0.42222629487514496,
      "relation_loss": 0.0025915249017998574,
      "selected_pairs": 27085.8,
      "step": 20640,
      "student_distance_mean": 23.491821479797363,
      "teacher_distance_mean": 0.8541159510612488,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0025915249017998574,
      "weighted_relation_loss": 0.0025915249017998574
    },
    {
      "action_query_batch_std": 0.3573232084512711,
      "action_query_norm": 41.19375877380371,
      "action_relation_loss": 0.0025732306530699134,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.31711880261927,
      "flow_matching_loss": 0.02043631635606289,
      "grad_norm": 0.12230178713798523,
      "h_stock_valid_norm": 97.99563598632812,
      "invalid_tail_samples": 23.5,
      "learning_rate": 2.4292203661563312e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 26920.3,
      "query_h_norm_ratio": 0.4203655689954758,
      "relation_loss": 0.0025732306530699134,
      "selected_pairs": 26920.3,
      "step": 20650,
      "student_distance_mean": 23.478157806396485,
      "teacher_distance_mean": 0.856582760810852,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0025732306530699134,
      "weighted_relation_loss": 0.0025732306530699134
    },
    {
      "action_query_batch_std": 0.35632666647434236,
      "action_query_norm": 41.181695556640626,
      "action_relation_loss": 0.002496825018897653,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.32647333956969,
      "flow_matching_loss": 0.020978260040283202,
      "grad_norm": 0.12179645150899887,
      "h_stock_valid_norm": 97.92079467773438,
      "invalid_tail_samples": 21.9,
      "learning_rate": 2.424494676790141e-05,
      "loss": 0.0235,
      "positive_teacher_pairs": 27289.7,
      "query_h_norm_ratio": 0.4205628871917725,
      "relation_loss": 0.002496825018897653,
      "selected_pairs": 27289.7,
      "step": 20660,
      "student_distance_mean": 23.416480827331544,
      "teacher_distance_mean": 0.8541265368461609,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.002496825018897653,
      "weighted_relation_loss": 0.002496825018897653
    },
    {
      "action_query_batch_std": 0.3586688458919525,
      "action_query_norm": 41.212730407714844,
      "action_relation_loss": 0.0025503049371764066,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.335827876520113,
      "flow_matching_loss": 0.022199755348265172,
      "grad_norm": 0.10055994242429733,
      "h_stock_valid_norm": 97.90097503662109,
      "invalid_tail_samples": 25.2,
      "learning_rate": 2.4197721169057745e-05,
      "loss": 0.0248,
      "positive_teacher_pairs": 26524.0,
      "query_h_norm_ratio": 0.420970955491066,
      "relation_loss": 0.0025503049371764066,
      "selected_pairs": 26524.0,
      "step": 20670,
      "student_distance_mean": 23.754311943054198,
      "teacher_distance_mean": 0.864881831407547,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0025503049371764066,
      "weighted_relation_loss": 0.0025503049371764066
    },
    {
      "action_query_batch_std": 0.35987584888935087,
      "action_query_norm": 41.21487731933594,
      "action_relation_loss": 0.002403718512505293,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.345182413470532,
      "flow_matching_loss": 0.02129723522812128,
      "grad_norm": 0.10339018702507019,
      "h_stock_valid_norm": 97.78286056518554,
      "invalid_tail_samples": 22.7,
      "learning_rate": 2.4150526922415857e-05,
      "loss": 0.0237,
      "positive_teacher_pairs": 27110.4,
      "query_h_norm_ratio": 0.42149660289287566,
      "relation_loss": 0.002403718512505293,
      "selected_pairs": 27110.4,
      "step": 20680,
      "student_distance_mean": 23.63936576843262,
      "teacher_distance_mean": 0.8588218510150909,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.002403718512505293,
      "weighted_relation_loss": 0.002403718512505293
    },
    {
      "action_query_batch_std": 0.35168724358081815,
      "action_query_norm": 41.15486602783203,
      "action_relation_loss": 0.002529632905498147,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.354536950420954,
      "flow_matching_loss": 0.02187868542969227,
      "grad_norm": 0.09270469844341278,
      "h_stock_valid_norm": 97.6706787109375,
      "invalid_tail_samples": 23.4,
      "learning_rate": 2.41033640853212e-05,
      "loss": 0.0244,
      "positive_teacher_pairs": 26951.1,
      "query_h_norm_ratio": 0.4213665068149567,
      "relation_loss": 0.002529632905498147,
      "selected_pairs": 26951.1,
      "step": 20690,
      "student_distance_mean": 23.23093376159668,
      "teacher_distance_mean": 0.8439264595508575,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002529632905498147,
      "weighted_relation_loss": 0.002529632905498147
    },
    {
      "action_query_batch_std": 0.3538693219423294,
      "action_query_norm": 41.16582984924317,
      "action_relation_loss": 0.002376763056963682,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.363891487371376,
      "flow_matching_loss": 0.019854104053229092,
      "grad_norm": 0.09344989061355591,
      "h_stock_valid_norm": 97.915087890625,
      "invalid_tail_samples": 23.7,
      "learning_rate": 2.405623271508108e-05,
      "loss": 0.0222,
      "positive_teacher_pairs": 26876.6,
      "query_h_norm_ratio": 0.42042577266693115,
      "relation_loss": 0.002376763056963682,
      "selected_pairs": 26876.6,
      "step": 20700,
      "student_distance_mean": 23.29548587799072,
      "teacher_distance_mean": 0.8498884081840515,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.002376763056963682,
      "weighted_relation_loss": 0.002376763056963682
    },
    {
      "action_query_batch_std": 0.35231936872005465,
      "action_query_norm": 41.16770629882812,
      "action_relation_loss": 0.0024523611646145582,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.373246024321794,
      "flow_matching_loss": 0.020135718584060668,
      "grad_norm": 0.10223744064569473,
      "h_stock_valid_norm": 97.97978134155274,
      "invalid_tail_samples": 24.2,
      "learning_rate": 2.4009132868964522e-05,
      "loss": 0.0226,
      "positive_teacher_pairs": 26755.6,
      "query_h_norm_ratio": 0.4201699733734131,
      "relation_loss": 0.0024523611646145582,
      "selected_pairs": 26755.6,
      "step": 20710,
      "student_distance_mean": 23.253325271606446,
      "teacher_distance_mean": 0.847421008348465,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0024523611646145582,
      "weighted_relation_loss": 0.0024523611646145582
    },
    {
      "action_query_batch_std": 0.3512034177780151,
      "action_query_norm": 41.16729507446289,
      "action_relation_loss": 0.0024504652246832848,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.382600561272216,
      "flow_matching_loss": 0.022342630475759507,
      "grad_norm": 0.11379864811897278,
      "h_stock_valid_norm": 97.92314529418945,
      "invalid_tail_samples": 25.7,
      "learning_rate": 2.3962064604202327e-05,
      "loss": 0.0248,
      "positive_teacher_pairs": 26416.6,
      "query_h_norm_ratio": 0.4204060584306717,
      "relation_loss": 0.0024504652246832848,
      "selected_pairs": 26416.6,
      "step": 20720,
      "student_distance_mean": 23.179970741271973,
      "teacher_distance_mean": 0.8431564331054687,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0024504652246832848,
      "weighted_relation_loss": 0.0024504652246832848
    },
    {
      "action_query_batch_std": 0.35803778767585753,
      "action_query_norm": 41.19730644226074,
      "action_relation_loss": 0.0024036138085648416,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.39195509822264,
      "flow_matching_loss": 0.020692568458616733,
      "grad_norm": 0.09875648468732834,
      "h_stock_valid_norm": 97.93683700561523,
      "invalid_tail_samples": 24.5,
      "learning_rate": 2.3915027977986858e-05,
      "loss": 0.0231,
      "positive_teacher_pairs": 26696.2,
      "query_h_norm_ratio": 0.42065576612949374,
      "relation_loss": 0.0024036138085648416,
      "selected_pairs": 26696.2,
      "step": 20730,
      "student_distance_mean": 23.58182716369629,
      "teacher_distance_mean": 0.8611415743827819,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0024036138085648416,
      "weighted_relation_loss": 0.0024036138085648416
    },
    {
      "action_query_batch_std": 0.3532515287399292,
      "action_query_norm": 41.16096420288086,
      "action_relation_loss": 0.002487456030212343,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.40130963517306,
      "flow_matching_loss": 0.022954692132771015,
      "grad_norm": 0.09810654073953629,
      "h_stock_valid_norm": 97.83175506591797,
      "invalid_tail_samples": 22.4,
      "learning_rate": 2.386802304747205e-05,
      "loss": 0.0254,
      "positive_teacher_pairs": 27178.0,
      "query_h_norm_ratio": 0.42073517143726347,
      "relation_loss": 0.002487456030212343,
      "selected_pairs": 27178.0,
      "step": 20740,
      "student_distance_mean": 23.288357734680176,
      "teacher_distance_mean": 0.8479250907897949,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.002487456030212343,
      "weighted_relation_loss": 0.002487456030212343
    },
    {
      "action_query_batch_std": 0.35308838486671446,
      "action_query_norm": 41.14351463317871,
      "action_relation_loss": 0.002388557605445385,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.41066417212348,
      "flow_matching_loss": 0.0213964831084013,
      "grad_norm": 0.0857366994023323,
      "h_stock_valid_norm": 97.90363082885742,
      "invalid_tail_samples": 23.4,
      "learning_rate": 2.382104986977332e-05,
      "loss": 0.0238,
      "positive_teacher_pairs": 26940.8,
      "query_h_norm_ratio": 0.4202471375465393,
      "relation_loss": 0.002388557605445385,
      "selected_pairs": 26940.8,
      "step": 20750,
      "student_distance_mean": 23.264055252075195,
      "teacher_distance_mean": 0.8499489605426789,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002388557605445385,
      "weighted_relation_loss": 0.002388557605445385
    },
    {
      "action_query_batch_std": 0.35827969908714297,
      "action_query_norm": 41.19026870727539,
      "action_relation_loss": 0.002513884985819459,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.4200187090739,
      "flow_matching_loss": 0.021760342735797168,
      "grad_norm": 0.08466657996177673,
      "h_stock_valid_norm": 98.02865676879883,
      "invalid_tail_samples": 20.9,
      "learning_rate": 2.377410850196749e-05,
      "loss": 0.0243,
      "positive_teacher_pairs": 27525.3,
      "query_h_norm_ratio": 0.4201875180006027,
      "relation_loss": 0.002513884985819459,
      "selected_pairs": 27525.3,
      "step": 20760,
      "student_distance_mean": 23.567083168029786,
      "teacher_distance_mean": 0.8604134678840637,
      "valid_action_samples": 235.1,
      "weighted_auxiliary_loss": 0.002513884985819459,
      "weighted_relation_loss": 0.002513884985819459
    },
    {
      "action_query_batch_std": 0.3599929243326187,
      "action_query_norm": 41.18854026794433,
      "action_relation_loss": 0.0023898114915937184,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.429373246024323,
      "flow_matching_loss": 0.022340320516377687,
      "grad_norm": 0.10427963733673096,
      "h_stock_valid_norm": 97.84160766601562,
      "invalid_tail_samples": 23.5,
      "learning_rate": 2.3727199001092797e-05,
      "loss": 0.0247,
      "positive_teacher_pairs": 26920.2,
      "query_h_norm_ratio": 0.4209736555814743,
      "relation_loss": 0.0023898114915937184,
      "selected_pairs": 26920.2,
      "step": 20770,
      "student_distance_mean": 23.645342445373537,
      "teacher_distance_mean": 0.8621614813804627,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0023898114915937184,
      "weighted_relation_loss": 0.0023898114915937184
    },
    {
      "action_query_batch_std": 0.3548332154750824,
      "action_query_norm": 41.18098831176758,
      "action_relation_loss": 0.0024194483878090976,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.43872778297474,
      "flow_matching_loss": 0.01991759315133095,
      "grad_norm": 0.10585086047649384,
      "h_stock_valid_norm": 97.69696731567383,
      "invalid_tail_samples": 22.2,
      "learning_rate": 2.3680321424148677e-05,
      "loss": 0.0223,
      "positive_teacher_pairs": 27217.7,
      "query_h_norm_ratio": 0.421524578332901,
      "relation_loss": 0.0024194483878090976,
      "selected_pairs": 27217.7,
      "step": 20780,
      "student_distance_mean": 23.417063903808593,
      "teacher_distance_mean": 0.8535641729831696,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0024194483878090976,
      "weighted_relation_loss": 0.0024194483878090976
    },
    {
      "action_query_batch_std": 0.3529189735651016,
      "action_query_norm": 41.15956153869629,
      "action_relation_loss": 0.0025960490806028248,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.448082319925163,
      "flow_matching_loss": 0.0214657349511981,
      "grad_norm": 0.0981394574046135,
      "h_stock_valid_norm": 97.53450012207031,
      "invalid_tail_samples": 21.2,
      "learning_rate": 2.3633475828095793e-05,
      "loss": 0.0241,
      "positive_teacher_pairs": 27456.6,
      "query_h_norm_ratio": 0.42200144231319425,
      "relation_loss": 0.0025960490806028248,
      "selected_pairs": 27456.6,
      "step": 20790,
      "student_distance_mean": 23.276371574401857,
      "teacher_distance_mean": 0.8485074043273926,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.0025960490806028248,
      "weighted_relation_loss": 0.0025960490806028248
    },
    {
      "action_query_batch_std": 0.3548136979341507,
      "action_query_norm": 41.16788177490234,
      "action_relation_loss": 0.0024616159964352844,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.457436856875585,
      "flow_matching_loss": 0.02137096170336008,
      "grad_norm": 0.10398489981889725,
      "h_stock_valid_norm": 97.58359069824219,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.358666226985599e-05,
      "loss": 0.0238,
      "positive_teacher_pairs": 26966.4,
      "query_h_norm_ratio": 0.42187574207782746,
      "relation_loss": 0.0024616159964352844,
      "selected_pairs": 26966.4,
      "step": 20800,
      "student_distance_mean": 23.432235145568846,
      "teacher_distance_mean": 0.8529497385025024,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0024616159964352844,
      "weighted_relation_loss": 0.0024616159964352844
    },
    {
      "action_query_batch_std": 0.3572328627109528,
      "action_query_norm": 41.15083656311035,
      "action_relation_loss": 0.002371598710305989,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.466791393826007,
      "flow_matching_loss": 0.020715409703552724,
      "grad_norm": 0.08668249845504761,
      "h_stock_valid_norm": 97.79848937988281,
      "invalid_tail_samples": 26.7,
      "learning_rate": 2.353988080631213e-05,
      "loss": 0.0231,
      "positive_teacher_pairs": 26176.7,
      "query_h_norm_ratio": 0.42077392935752866,
      "relation_loss": 0.002371598710305989,
      "selected_pairs": 26176.7,
      "step": 20810,
      "student_distance_mean": 23.513108253479004,
      "teacher_distance_mean": 0.8578766822814942,
      "valid_action_samples": 229.3,
      "weighted_auxiliary_loss": 0.002371598710305989,
      "weighted_relation_loss": 0.002371598710305989
    },
    {
      "action_query_batch_std": 0.35692674219608306,
      "action_query_norm": 41.17460250854492,
      "action_relation_loss": 0.00235864263959229,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.476145930776426,
      "flow_matching_loss": 0.02003376539796591,
      "grad_norm": 0.09578695148229599,
      "h_stock_valid_norm": 97.71363143920898,
      "invalid_tail_samples": 23.0,
      "learning_rate": 2.3493131494308142e-05,
      "loss": 0.0224,
      "positive_teacher_pairs": 27035.6,
      "query_h_norm_ratio": 0.42138386368751524,
      "relation_loss": 0.00235864263959229,
      "selected_pairs": 27035.6,
      "step": 20820,
      "student_distance_mean": 23.536137962341307,
      "teacher_distance_mean": 0.8579287111759186,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.00235864263959229,
      "weighted_relation_loss": 0.00235864263959229
    },
    {
      "action_query_batch_std": 0.3598463922739029,
      "action_query_norm": 41.18679695129394,
      "action_relation_loss": 0.002384451194666326,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.485500467726848,
      "flow_matching_loss": 0.020422383211553097,
      "grad_norm": 0.09006652235984802,
      "h_stock_valid_norm": 97.68451919555665,
      "invalid_tail_samples": 23.9,
      "learning_rate": 2.344641439064881e-05,
      "loss": 0.0228,
      "positive_teacher_pairs": 26828.2,
      "query_h_norm_ratio": 0.4216345876455307,
      "relation_loss": 0.002384451194666326,
      "selected_pairs": 26828.2,
      "step": 20830,
      "student_distance_mean": 23.610967445373536,
      "teacher_distance_mean": 0.8623341858386994,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.002384451194666326,
      "weighted_relation_loss": 0.002384451194666326
    },
    {
      "action_query_batch_std": 0.3557833284139633,
      "action_query_norm": 41.160922622680665,
      "action_relation_loss": 0.0025017713895067574,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.49485500467727,
      "flow_matching_loss": 0.020085790194571018,
      "grad_norm": 0.11111767590045929,
      "h_stock_valid_norm": 98.24641189575195,
      "invalid_tail_samples": 24.0,
      "learning_rate": 2.339972955209984e-05,
      "loss": 0.0226,
      "positive_teacher_pairs": 26813.1,
      "query_h_norm_ratio": 0.41896032989025117,
      "relation_loss": 0.0025017713895067574,
      "selected_pairs": 26813.1,
      "step": 20840,
      "student_distance_mean": 23.424378395080566,
      "teacher_distance_mean": 0.853520268201828,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0025017713895067574,
      "weighted_relation_loss": 0.0025017713895067574
    },
    {
      "action_query_batch_std": 0.3547031283378601,
      "action_query_norm": 41.151484680175784,
      "action_relation_loss": 0.002411357592791319,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.504209541627688,
      "flow_matching_loss": 0.02284929919987917,
      "grad_norm": 0.08833751827478409,
      "h_stock_valid_norm": 98.35219650268554,
      "invalid_tail_samples": 21.8,
      "learning_rate": 2.335307703538771e-05,
      "loss": 0.0253,
      "positive_teacher_pairs": 27317.5,
      "query_h_norm_ratio": 0.4184124290943146,
      "relation_loss": 0.002411357592791319,
      "selected_pairs": 27317.5,
      "step": 20850,
      "student_distance_mean": 23.351209259033205,
      "teacher_distance_mean": 0.8532798409461975,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.002411357592791319,
      "weighted_relation_loss": 0.002411357592791319
    },
    {
      "action_query_batch_std": 0.36082902550697327,
      "action_query_norm": 41.19889183044434,
      "action_relation_loss": 0.0023876762948930263,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.51356407857811,
      "flow_matching_loss": 0.022136747464537622,
      "grad_norm": 0.10014259815216064,
      "h_stock_valid_norm": 98.25969161987305,
      "invalid_tail_samples": 23.2,
      "learning_rate": 2.330645689719962e-05,
      "loss": 0.0245,
      "positive_teacher_pairs": 26987.8,
      "query_h_norm_ratio": 0.419290217757225,
      "relation_loss": 0.0023876762948930263,
      "selected_pairs": 26987.8,
      "step": 20860,
      "student_distance_mean": 23.696081733703615,
      "teacher_distance_mean": 0.8673037886619568,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0023876762948930263,
      "weighted_relation_loss": 0.0023876762948930263
    },
    {
      "action_query_batch_std": 0.35606837570667266,
      "action_query_norm": 41.175558471679686,
      "action_relation_loss": 0.0025041377637535335,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.522918615528532,
      "flow_matching_loss": 0.020560615323483945,
      "grad_norm": 0.12278562784194946,
      "h_stock_valid_norm": 98.21347274780274,
      "invalid_tail_samples": 22.5,
      "learning_rate": 2.3259869194183414e-05,
      "loss": 0.0231,
      "positive_teacher_pairs": 27150.1,
      "query_h_norm_ratio": 0.41925043165683745,
      "relation_loss": 0.0025041377637535335,
      "selected_pairs": 27150.1,
      "step": 20870,
      "student_distance_mean": 23.57216148376465,
      "teacher_distance_mean": 0.8588767766952514,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0025041377637535335,
      "weighted_relation_loss": 0.0025041377637535335
    },
    {
      "action_query_batch_std": 0.3556240350008011,
      "action_query_norm": 41.1852481842041,
      "action_relation_loss": 0.0024185116635635495,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.532273152478954,
      "flow_matching_loss": 0.019230378791689873,
      "grad_norm": 0.08424112200737,
      "h_stock_valid_norm": 98.62224349975585,
      "invalid_tail_samples": 25.4,
      "learning_rate": 2.321331398294759e-05,
      "loss": 0.0216,
      "positive_teacher_pairs": 26482.7,
      "query_h_norm_ratio": 0.41761431396007537,
      "relation_loss": 0.0024185116635635495,
      "selected_pairs": 26482.7,
      "step": 20880,
      "student_distance_mean": 23.50809803009033,
      "teacher_distance_mean": 0.8566109657287597,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0024185116635635495,
      "weighted_relation_loss": 0.0024185116635635495
    },
    {
      "action_query_batch_std": 0.3540663480758667,
      "action_query_norm": 41.17834243774414,
      "action_relation_loss": 0.002455872902646661,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.541627689429372,
      "flow_matching_loss": 0.02020274605602026,
      "grad_norm": 0.16836413741111755,
      "h_stock_valid_norm": 98.65409774780274,
      "invalid_tail_samples": 22.9,
      "learning_rate": 2.3166791320061094e-05,
      "loss": 0.0227,
      "positive_teacher_pairs": 27058.6,
      "query_h_norm_ratio": 0.4174023002386093,
      "relation_loss": 0.002455872902646661,
      "selected_pairs": 27058.6,
      "step": 20890,
      "student_distance_mean": 23.22104549407959,
      "teacher_distance_mean": 0.8449335157871246,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.002455872902646661,
      "weighted_relation_loss": 0.002455872902646661
    },
    {
      "action_query_batch_std": 0.36096824109554293,
      "action_query_norm": 41.20226287841797,
      "action_relation_loss": 0.002380034828092903,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.550982226379794,
      "flow_matching_loss": 0.02018573824316263,
      "grad_norm": 0.09330224245786667,
      "h_stock_valid_norm": 98.50757751464843,
      "invalid_tail_samples": 23.0,
      "learning_rate": 2.312030126205335e-05,
      "loss": 0.0226,
      "positive_teacher_pairs": 27041.8,
      "query_h_norm_ratio": 0.4182680666446686,
      "relation_loss": 0.002380034828092903,
      "selected_pairs": 27041.8,
      "step": 20900,
      "student_distance_mean": 23.820124626159668,
      "teacher_distance_mean": 0.8720930397510529,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.002380034828092903,
      "weighted_relation_loss": 0.002380034828092903
    },
    {
      "action_query_batch_std": 0.35113403797149656,
      "action_query_norm": 41.111198043823244,
      "action_relation_loss": 0.002612124942243099,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.560336763330216,
      "flow_matching_loss": 0.01972144525498152,
      "grad_norm": 0.10112150758504868,
      "h_stock_valid_norm": 98.27763366699219,
      "invalid_tail_samples": 24.6,
      "learning_rate": 2.3073843865414162e-05,
      "loss": 0.0223,
      "positive_teacher_pairs": 26666.1,
      "query_h_norm_ratio": 0.41831904351711274,
      "relation_loss": 0.002612124942243099,
      "selected_pairs": 26666.1,
      "step": 20910,
      "student_distance_mean": 23.209861755371094,
      "teacher_distance_mean": 0.8464496970176697,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.002612124942243099,
      "weighted_relation_loss": 0.002612124942243099
    },
    {
      "action_query_batch_std": 0.35793322026729585,
      "action_query_norm": 41.15518455505371,
      "action_relation_loss": 0.002355671371333301,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.569691300280635,
      "flow_matching_loss": 0.02114080060273409,
      "grad_norm": 0.10429860651493073,
      "h_stock_valid_norm": 97.81792297363282,
      "invalid_tail_samples": 24.8,
      "learning_rate": 2.302741918659363e-05,
      "loss": 0.0235,
      "positive_teacher_pairs": 26618.1,
      "query_h_norm_ratio": 0.4207365959882736,
      "relation_loss": 0.002355671371333301,
      "selected_pairs": 26618.1,
      "step": 20920,
      "student_distance_mean": 23.5119348526001,
      "teacher_distance_mean": 0.8589855194091797,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.002355671371333301,
      "weighted_relation_loss": 0.002355671371333301
    },
    {
      "action_query_batch_std": 0.35643433332443236,
      "action_query_norm": 41.13987007141113,
      "action_relation_loss": 0.0024666909594088793,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.579045837231057,
      "flow_matching_loss": 0.019602044858038424,
      "grad_norm": 0.10215643048286438,
      "h_stock_valid_norm": 97.936328125,
      "invalid_tail_samples": 24.0,
      "learning_rate": 2.2981027282002154e-05,
      "loss": 0.0221,
      "positive_teacher_pairs": 26803.9,
      "query_h_norm_ratio": 0.42007095515728,
      "relation_loss": 0.0024666909594088793,
      "selected_pairs": 26803.9,
      "step": 20930,
      "student_distance_mean": 23.537831687927245,
      "teacher_distance_mean": 0.8595274269580842,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0024666909594088793,
      "weighted_relation_loss": 0.0024666909594088793
    },
    {
      "action_query_batch_std": 0.3538141161203384,
      "action_query_norm": 41.12592544555664,
      "action_relation_loss": 0.002464111289009452,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.58840037418148,
      "flow_matching_loss": 0.020072199031710626,
      "grad_norm": 0.10687386989593506,
      "h_stock_valid_norm": 97.89617462158203,
      "invalid_tail_samples": 25.1,
      "learning_rate": 2.2934668208010236e-05,
      "loss": 0.0225,
      "positive_teacher_pairs": 26559.1,
      "query_h_norm_ratio": 0.4200985699892044,
      "relation_loss": 0.002464111289009452,
      "selected_pairs": 26559.1,
      "step": 20940,
      "student_distance_mean": 23.4090295791626,
      "teacher_distance_mean": 0.8525089502334595,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.002464111289009452,
      "weighted_relation_loss": 0.002464111289009452
    },
    {
      "action_query_batch_std": 0.3534138947725296,
      "action_query_norm": 41.13463363647461,
      "action_relation_loss": 0.002468543965369463,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.5977549111319,
      "flow_matching_loss": 0.02061512004584074,
      "grad_norm": 0.0947057455778122,
      "h_stock_valid_norm": 98.10944290161133,
      "invalid_tail_samples": 23.5,
      "learning_rate": 2.2888342020948555e-05,
      "loss": 0.0231,
      "positive_teacher_pairs": 26922.7,
      "query_h_norm_ratio": 0.4192743211984634,
      "relation_loss": 0.002468543965369463,
      "selected_pairs": 26922.7,
      "step": 20950,
      "student_distance_mean": 23.330421257019044,
      "teacher_distance_mean": 0.8513484418392181,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.002468543965369463,
      "weighted_relation_loss": 0.002468543965369463
    },
    {
      "action_query_batch_std": 0.3536087661981583,
      "action_query_norm": 41.1328987121582,
      "action_relation_loss": 0.0024056462803855538,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.60710944808232,
      "flow_matching_loss": 0.020617248862981795,
      "grad_norm": 0.14979666471481323,
      "h_stock_valid_norm": 98.14612274169922,
      "invalid_tail_samples": 22.4,
      "learning_rate": 2.2842048777107783e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 27180.3,
      "query_h_norm_ratio": 0.41909952759742736,
      "relation_loss": 0.0024056462803855538,
      "selected_pairs": 27180.3,
      "step": 20960,
      "student_distance_mean": 23.363059043884277,
      "teacher_distance_mean": 0.8531110465526581,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.0024056462803855538,
      "weighted_relation_loss": 0.0024056462803855538
    },
    {
      "action_query_batch_std": 0.35614014565944674,
      "action_query_norm": 41.15681838989258,
      "action_relation_loss": 0.002392108947969973,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.61646398503274,
      "flow_matching_loss": 0.019645036850124597,
      "grad_norm": 0.10410427302122116,
      "h_stock_valid_norm": 98.09650802612305,
      "invalid_tail_samples": 23.1,
      "learning_rate": 2.2795788532738553e-05,
      "loss": 0.022,
      "positive_teacher_pairs": 27013.6,
      "query_h_norm_ratio": 0.41955626308918,
      "relation_loss": 0.002392108947969973,
      "selected_pairs": 27013.6,
      "step": 20970,
      "student_distance_mean": 23.353488731384278,
      "teacher_distance_mean": 0.8526619911193848,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002392108947969973,
      "weighted_relation_loss": 0.002392108947969973
    },
    {
      "action_query_batch_std": 0.35288701951503754,
      "action_query_norm": 41.129695129394534,
      "action_relation_loss": 0.0025317773688584566,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.625818521983163,
      "flow_matching_loss": 0.021660999953746797,
      "grad_norm": 0.11006204783916473,
      "h_stock_valid_norm": 98.06522216796876,
      "invalid_tail_samples": 23.9,
      "learning_rate": 2.2749561344051468e-05,
      "loss": 0.0242,
      "positive_teacher_pairs": 26831.8,
      "query_h_norm_ratio": 0.4194129705429077,
      "relation_loss": 0.0025317773688584566,
      "selected_pairs": 26831.8,
      "step": 20980,
      "student_distance_mean": 23.25205307006836,
      "teacher_distance_mean": 0.8498416423797608,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0025317773688584566,
      "weighted_relation_loss": 0.0025317773688584566
    },
    {
      "action_query_batch_std": 0.3564710199832916,
      "action_query_norm": 41.17767868041992,
      "action_relation_loss": 0.0024227271787822247,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.63517305893358,
      "flow_matching_loss": 0.020468478463590144,
      "grad_norm": 0.1635923683643341,
      "h_stock_valid_norm": 97.81162796020507,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.2703367267216896e-05,
      "loss": 0.0229,
      "positive_teacher_pairs": 26964.2,
      "query_h_norm_ratio": 0.420992323756218,
      "relation_loss": 0.0024227271787822247,
      "selected_pairs": 26964.2,
      "step": 20990,
      "student_distance_mean": 23.342495727539063,
      "teacher_distance_mean": 0.8534628987312317,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0024227271787822247,
      "weighted_relation_loss": 0.0024227271787822247
    },
    {
      "action_query_batch_std": 0.3566118270158768,
      "action_query_norm": 41.16198310852051,
      "action_relation_loss": 0.0024541130987927318,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.644527595884004,
      "flow_matching_loss": 0.01955606210976839,
      "grad_norm": 0.10300537943840027,
      "h_stock_valid_norm": 97.74997329711914,
      "invalid_tail_samples": 22.4,
      "learning_rate": 2.2657206358364997e-05,
      "loss": 0.022,
      "positive_teacher_pairs": 27180.4,
      "query_h_norm_ratio": 0.421096995472908,
      "relation_loss": 0.0024541130987927318,
      "selected_pairs": 27180.4,
      "step": 21000,
      "student_distance_mean": 23.507527160644532,
      "teacher_distance_mean": 0.8563460767269134,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.0024541130987927318,
      "weighted_relation_loss": 0.0024541130987927318
    },
    {
      "action_query_batch_std": 0.3523163586854935,
      "action_query_norm": 41.14704475402832,
      "action_relation_loss": 0.0024378196336328982,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.653882132834426,
      "flow_matching_loss": 0.021441625244915486,
      "grad_norm": 0.09479736536741257,
      "h_stock_valid_norm": 97.84136428833008,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.261107867358563e-05,
      "loss": 0.0239,
      "positive_teacher_pairs": 26969.9,
      "query_h_norm_ratio": 0.420551934838295,
      "relation_loss": 0.0024378196336328982,
      "selected_pairs": 26969.9,
      "step": 21010,
      "student_distance_mean": 23.257819747924806,
      "teacher_distance_mean": 0.8459058821201324,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0024378196336328982,
      "weighted_relation_loss": 0.0024378196336328982
    },
    {
      "action_query_batch_std": 0.3575033754110336,
      "action_query_norm": 41.169673156738284,
      "action_relation_loss": 0.00241561378352344,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.663236669784844,
      "flow_matching_loss": 0.020134420692920686,
      "grad_norm": 0.10061004012823105,
      "h_stock_valid_norm": 97.97097091674804,
      "invalid_tail_samples": 23.4,
      "learning_rate": 2.2564984268928263e-05,
      "loss": 0.0226,
      "positive_teacher_pairs": 26945.5,
      "query_h_norm_ratio": 0.4202245384454727,
      "relation_loss": 0.00241561378352344,
      "selected_pairs": 26945.5,
      "step": 21020,
      "student_distance_mean": 23.56921844482422,
      "teacher_distance_mean": 0.8603592813014984,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.00241561378352344,
      "weighted_relation_loss": 0.00241561378352344
    },
    {
      "action_query_batch_std": 0.352164426445961,
      "action_query_norm": 41.11510467529297,
      "action_relation_loss": 0.002512814477086067,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.672591206735266,
      "flow_matching_loss": 0.02223347593098879,
      "grad_norm": 0.10028384625911713,
      "h_stock_valid_norm": 98.06679000854493,
      "invalid_tail_samples": 25.6,
      "learning_rate": 2.251892320040198e-05,
      "loss": 0.0247,
      "positive_teacher_pairs": 26432.8,
      "query_h_norm_ratio": 0.419258314371109,
      "relation_loss": 0.002512814477086067,
      "selected_pairs": 26432.8,
      "step": 21030,
      "student_distance_mean": 23.185825920104982,
      "teacher_distance_mean": 0.8482008814811707,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.002512814477086067,
      "weighted_relation_loss": 0.002512814477086067
    },
    {
      "action_query_batch_std": 0.3561264991760254,
      "action_query_norm": 41.13571319580078,
      "action_relation_loss": 0.002389914938248694,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.681945743685688,
      "flow_matching_loss": 0.020153612829744816,
      "grad_norm": 0.12284523993730545,
      "h_stock_valid_norm": 98.07933197021484,
      "invalid_tail_samples": 22.5,
      "learning_rate": 2.2472895523975312e-05,
      "loss": 0.0225,
      "positive_teacher_pairs": 27149.0,
      "query_h_norm_ratio": 0.41941521465778353,
      "relation_loss": 0.002389914938248694,
      "selected_pairs": 27149.0,
      "step": 21040,
      "student_distance_mean": 23.384742546081544,
      "teacher_distance_mean": 0.8537565886974334,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.002389914938248694,
      "weighted_relation_loss": 0.002389914938248694
    },
    {
      "action_query_batch_std": 0.3557699978351593,
      "action_query_norm": 41.14555130004883,
      "action_relation_loss": 0.002548105036839843,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.69130028063611,
      "flow_matching_loss": 0.02047793259844184,
      "grad_norm": 0.07605370879173279,
      "h_stock_valid_norm": 98.04266510009765,
      "invalid_tail_samples": 23.4,
      "learning_rate": 2.2426901295576212e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 26943.2,
      "query_h_norm_ratio": 0.4196724593639374,
      "relation_loss": 0.002548105036839843,
      "selected_pairs": 26943.2,
      "step": 21050,
      "student_distance_mean": 23.426186752319335,
      "teacher_distance_mean": 0.8577932715415955,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002548105036839843,
      "weighted_relation_loss": 0.002548105036839843
    },
    {
      "action_query_batch_std": 0.3520794302225113,
      "action_query_norm": 41.11707878112793,
      "action_relation_loss": 0.0025676370831206443,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.70065481758653,
      "flow_matching_loss": 0.020379840582609176,
      "grad_norm": 0.10098891705274582,
      "h_stock_valid_norm": 98.01846618652344,
      "invalid_tail_samples": 21.6,
      "learning_rate": 2.2380940571092013e-05,
      "loss": 0.0229,
      "positive_teacher_pairs": 27365.3,
      "query_h_norm_ratio": 0.41948592066764834,
      "relation_loss": 0.0025676370831206443,
      "selected_pairs": 27365.3,
      "step": 21060,
      "student_distance_mean": 23.100732231140135,
      "teacher_distance_mean": 0.8447649538516998,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.0025676370831206443,
      "weighted_relation_loss": 0.0025676370831206443
    },
    {
      "action_query_batch_std": 0.35485748052597044,
      "action_query_norm": 41.14862060546875,
      "action_relation_loss": 0.0023738923715427517,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.71000935453695,
      "flow_matching_loss": 0.022390094958245753,
      "grad_norm": 0.10463591665029526,
      "h_stock_valid_norm": 98.16730194091797,
      "invalid_tail_samples": 21.7,
      "learning_rate": 2.2335013406369302e-05,
      "loss": 0.0248,
      "positive_teacher_pairs": 27344.0,
      "query_h_norm_ratio": 0.4191693991422653,
      "relation_loss": 0.0023738923715427517,
      "selected_pairs": 27344.0,
      "step": 21070,
      "student_distance_mean": 23.470610046386717,
      "teacher_distance_mean": 0.8575551033020019,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0023738923715427517,
      "weighted_relation_loss": 0.0023738923715427517
    },
    {
      "action_query_batch_std": 0.3546704024076462,
      "action_query_norm": 41.14657783508301,
      "action_relation_loss": 0.00241143899038434,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.719363891487372,
      "flow_matching_loss": 0.020735901407897474,
      "grad_norm": 0.11300720274448395,
      "h_stock_valid_norm": 98.00754776000977,
      "invalid_tail_samples": 25.7,
      "learning_rate": 2.228911985721397e-05,
      "loss": 0.0231,
      "positive_teacher_pairs": 26414.5,
      "query_h_norm_ratio": 0.41983396708965304,
      "relation_loss": 0.00241143899038434,
      "selected_pairs": 26414.5,
      "step": 21080,
      "student_distance_mean": 23.43655433654785,
      "teacher_distance_mean": 0.8538690447807312,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.00241143899038434,
      "weighted_relation_loss": 0.00241143899038434
    },
    {
      "action_query_batch_std": 0.3553370475769043,
      "action_query_norm": 41.12236557006836,
      "action_relation_loss": 0.0024784676032140853,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.72871842843779,
      "flow_matching_loss": 0.020089869480580093,
      "grad_norm": 0.11859192699193954,
      "h_stock_valid_norm": 98.20736694335938,
      "invalid_tail_samples": 22.8,
      "learning_rate": 2.224325997939095e-05,
      "loss": 0.0226,
      "positive_teacher_pairs": 27083.4,
      "query_h_norm_ratio": 0.4187304198741913,
      "relation_loss": 0.0024784676032140853,
      "selected_pairs": 27083.4,
      "step": 21090,
      "student_distance_mean": 23.396327400207518,
      "teacher_distance_mean": 0.8558204114437103,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0024784676032140853,
      "weighted_relation_loss": 0.0024784676032140853
    },
    {
      "action_query_batch_std": 0.35355812311172485,
      "action_query_norm": 41.09989204406738,
      "action_relation_loss": 0.002512845839373767,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.738072965388213,
      "flow_matching_loss": 0.021340103819966316,
      "grad_norm": 0.11407491564750671,
      "h_stock_valid_norm": 98.06134872436523,
      "invalid_tail_samples": 24.2,
      "learning_rate": 2.219743382862437e-05,
      "loss": 0.0239,
      "positive_teacher_pairs": 26762.6,
      "query_h_norm_ratio": 0.4191273868083954,
      "relation_loss": 0.002512845839373767,
      "selected_pairs": 26762.6,
      "step": 21100,
      "student_distance_mean": 23.392310905456544,
      "teacher_distance_mean": 0.855107432603836,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.002512845839373767,
      "weighted_relation_loss": 0.002512845839373767
    },
    {
      "action_query_batch_std": 0.35389175117015836,
      "action_query_norm": 41.13523902893066,
      "action_relation_loss": 0.0024701459798961876,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.747427502338635,
      "flow_matching_loss": 0.020759089011698963,
      "grad_norm": 0.09893452376127243,
      "h_stock_valid_norm": 97.58051986694336,
      "invalid_tail_samples": 25.9,
      "learning_rate": 2.2151641460597294e-05,
      "loss": 0.0232,
      "positive_teacher_pairs": 26363.2,
      "query_h_norm_ratio": 0.42155244052410124,
      "relation_loss": 0.0024701459798961876,
      "selected_pairs": 26363.2,
      "step": 21110,
      "student_distance_mean": 23.31480827331543,
      "teacher_distance_mean": 0.8523089528083801,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.0024701459798961876,
      "weighted_relation_loss": 0.0024701459798961876
    },
    {
      "action_query_batch_std": 0.3527189761400223,
      "action_query_norm": 41.113932037353514,
      "action_relation_loss": 0.0024309220490977166,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.756782039289057,
      "flow_matching_loss": 0.017628098651766778,
      "grad_norm": 0.10943096876144409,
      "h_stock_valid_norm": 97.65790252685547,
      "invalid_tail_samples": 21.0,
      "learning_rate": 2.210588293095177e-05,
      "loss": 0.0201,
      "positive_teacher_pairs": 27502.1,
      "query_h_norm_ratio": 0.42100362479686737,
      "relation_loss": 0.0024309220490977166,
      "selected_pairs": 27502.1,
      "step": 21120,
      "student_distance_mean": 23.264935874938963,
      "teacher_distance_mean": 0.8468704044818878,
      "valid_action_samples": 235.0,
      "weighted_auxiliary_loss": 0.0024309220490977166,
      "weighted_relation_loss": 0.0024309220490977166
    },
    {
      "action_query_batch_std": 0.35383606851100924,
      "action_query_norm": 41.11367416381836,
      "action_relation_loss": 0.0024921112228184937,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.766136576239475,
      "flow_matching_loss": 0.02175320703536272,
      "grad_norm": 0.0924428328871727,
      "h_stock_valid_norm": 98.08992614746094,
      "invalid_tail_samples": 23.1,
      "learning_rate": 2.2060158295288714e-05,
      "loss": 0.0242,
      "positive_teacher_pairs": 27009.7,
      "query_h_norm_ratio": 0.4191478431224823,
      "relation_loss": 0.0024921112228184937,
      "selected_pairs": 27009.7,
      "step": 21130,
      "student_distance_mean": 23.426318168640137,
      "teacher_distance_mean": 0.8532328605651855,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0024921112228184937,
      "weighted_relation_loss": 0.0024921112228184937
    },
    {
      "action_query_batch_std": 0.36057782769203184,
      "action_query_norm": 41.16186180114746,
      "action_relation_loss": 0.002440260723233223,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.775491113189897,
      "flow_matching_loss": 0.020575304701924323,
      "grad_norm": 0.11374568194150925,
      "h_stock_valid_norm": 98.18341217041015,
      "invalid_tail_samples": 22.1,
      "learning_rate": 2.2014467609167905e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 27247.0,
      "query_h_norm_ratio": 0.41923647522926333,
      "relation_loss": 0.002440260723233223,
      "selected_pairs": 27247.0,
      "step": 21140,
      "student_distance_mean": 23.658187484741212,
      "teacher_distance_mean": 0.8654014825820923,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.002440260723233223,
      "weighted_relation_loss": 0.002440260723233223
    },
    {
      "action_query_batch_std": 0.3516050487756729,
      "action_query_norm": 41.09518394470215,
      "action_relation_loss": 0.002476942352950573,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.78484565014032,
      "flow_matching_loss": 0.02028384804725647,
      "grad_norm": 0.11189033836126328,
      "h_stock_valid_norm": 98.06021728515626,
      "invalid_tail_samples": 24.6,
      "learning_rate": 2.1968810928107806e-05,
      "loss": 0.0228,
      "positive_teacher_pairs": 26671.1,
      "query_h_norm_ratio": 0.419083771109581,
      "relation_loss": 0.002476942352950573,
      "selected_pairs": 26671.1,
      "step": 21150,
      "student_distance_mean": 23.204462242126464,
      "teacher_distance_mean": 0.8492263615131378,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.002476942352950573,
      "weighted_relation_loss": 0.002476942352950573
    },
    {
      "action_query_batch_std": 0.3548377454280853,
      "action_query_norm": 41.12210540771484,
      "action_relation_loss": 0.002505274722352624,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.794200187090738,
      "flow_matching_loss": 0.020561904273927213,
      "grad_norm": 0.15165425837039948,
      "h_stock_valid_norm": 97.64869689941406,
      "invalid_tail_samples": 26.3,
      "learning_rate": 2.1923188307585606e-05,
      "loss": 0.0231,
      "positive_teacher_pairs": 26278.6,
      "query_h_norm_ratio": 0.4211268961429596,
      "relation_loss": 0.002505274722352624,
      "selected_pairs": 26278.6,
      "step": 21160,
      "student_distance_mean": 23.281835556030273,
      "teacher_distance_mean": 0.8507069826126099,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.002505274722352624,
      "weighted_relation_loss": 0.002505274722352624
    },
    {
      "action_query_batch_std": 0.35753807723522185,
      "action_query_norm": 41.15183334350586,
      "action_relation_loss": 0.0023759847972542048,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.80355472404116,
      "flow_matching_loss": 0.020471185259521008,
      "grad_norm": 0.14762534201145172,
      "h_stock_valid_norm": 97.67130584716797,
      "invalid_tail_samples": 22.3,
      "learning_rate": 2.187759980303708e-05,
      "loss": 0.0228,
      "positive_teacher_pairs": 27202.0,
      "query_h_norm_ratio": 0.4213324189186096,
      "relation_loss": 0.0023759847972542048,
      "selected_pairs": 27202.0,
      "step": 21170,
      "student_distance_mean": 23.566792488098145,
      "teacher_distance_mean": 0.8606051564216614,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0023759847972542048,
      "weighted_relation_loss": 0.0023759847972542048
    },
    {
      "action_query_batch_std": 0.3550322026014328,
      "action_query_norm": 41.10702285766602,
      "action_relation_loss": 0.002489486080594361,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.81290926099158,
      "flow_matching_loss": 0.022850301302969455,
      "grad_norm": 0.13083185255527496,
      "h_stock_valid_norm": 97.4837745666504,
      "invalid_tail_samples": 24.9,
      "learning_rate": 2.1832045469856545e-05,
      "loss": 0.0253,
      "positive_teacher_pairs": 26600.1,
      "query_h_norm_ratio": 0.42168561816215516,
      "relation_loss": 0.002489486080594361,
      "selected_pairs": 26600.1,
      "step": 21180,
      "student_distance_mean": 23.407133102416992,
      "teacher_distance_mean": 0.8545000076293945,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.002489486080594361,
      "weighted_relation_loss": 0.002489486080594361
    },
    {
      "action_query_batch_std": 0.3545167833566666,
      "action_query_norm": 41.10519790649414,
      "action_relation_loss": 0.002330090862233192,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.822263797942004,
      "flow_matching_loss": 0.019958303961902856,
      "grad_norm": 0.08525583893060684,
      "h_stock_valid_norm": 97.81184234619141,
      "invalid_tail_samples": 23.2,
      "learning_rate": 2.1786525363396844e-05,
      "loss": 0.0223,
      "positive_teacher_pairs": 26992.0,
      "query_h_norm_ratio": 0.42025491297245027,
      "relation_loss": 0.002330090862233192,
      "selected_pairs": 26992.0,
      "step": 21190,
      "student_distance_mean": 23.388391494750977,
      "teacher_distance_mean": 0.8534283995628357,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.002330090862233192,
      "weighted_relation_loss": 0.002330090862233192
    },
    {
      "action_query_batch_std": 0.3579480141401291,
      "action_query_norm": 41.12312316894531,
      "action_relation_loss": 0.00243040774948895,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.831618334892422,
      "flow_matching_loss": 0.021151081286370756,
      "grad_norm": 0.12399405986070633,
      "h_stock_valid_norm": 97.78617553710937,
      "invalid_tail_samples": 25.5,
      "learning_rate": 2.1741039538969183e-05,
      "loss": 0.0236,
      "positive_teacher_pairs": 26465.1,
      "query_h_norm_ratio": 0.42054429948329924,
      "relation_loss": 0.00243040774948895,
      "selected_pairs": 26465.1,
      "step": 21200,
      "student_distance_mean": 23.550033378601075,
      "teacher_distance_mean": 0.8579222977161407,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.00243040774948895,
      "weighted_relation_loss": 0.00243040774948895
    },
    {
      "action_query_batch_std": 0.35147140026092527,
      "action_query_norm": 41.0938663482666,
      "action_relation_loss": 0.002546926005743444,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.840972871842844,
      "flow_matching_loss": 0.021179012209177017,
      "grad_norm": 0.10044851899147034,
      "h_stock_valid_norm": 97.70638427734374,
      "invalid_tail_samples": 24.5,
      "learning_rate": 2.169558805184313e-05,
      "loss": 0.0237,
      "positive_teacher_pairs": 26685.3,
      "query_h_norm_ratio": 0.42058914601802827,
      "relation_loss": 0.002546926005743444,
      "selected_pairs": 26685.3,
      "step": 21210,
      "student_distance_mean": 23.13525276184082,
      "teacher_distance_mean": 0.8431875348091126,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.002546926005743444,
      "weighted_relation_loss": 0.002546926005743444
    },
    {
      "action_query_batch_std": 0.3568002790212631,
      "action_query_norm": 41.11601219177246,
      "action_relation_loss": 0.0024757725186645985,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.850327408793266,
      "flow_matching_loss": 0.02195242643356323,
      "grad_norm": 0.0922171026468277,
      "h_stock_valid_norm": 97.98769912719726,
      "invalid_tail_samples": 24.7,
      "learning_rate": 2.165017095724651e-05,
      "loss": 0.0244,
      "positive_teacher_pairs": 26649.2,
      "query_h_norm_ratio": 0.4196085661649704,
      "relation_loss": 0.0024757725186645985,
      "selected_pairs": 26649.2,
      "step": 21220,
      "student_distance_mean": 23.580020332336424,
      "teacher_distance_mean": 0.8606465458869934,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0024757725186645985,
      "weighted_relation_loss": 0.0024757725186645985
    },
    {
      "action_query_batch_std": 0.3524134337902069,
      "action_query_norm": 41.085345077514646,
      "action_relation_loss": 0.002433960884809494,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.859681945743684,
      "flow_matching_loss": 0.020807249844074248,
      "grad_norm": 0.09788979589939117,
      "h_stock_valid_norm": 98.040087890625,
      "invalid_tail_samples": 24.9,
      "learning_rate": 2.1604788310365405e-05,
      "loss": 0.0232,
      "positive_teacher_pairs": 26599.7,
      "query_h_norm_ratio": 0.41907021403312683,
      "relation_loss": 0.002433960884809494,
      "selected_pairs": 26599.7,
      "step": 21230,
      "student_distance_mean": 23.293887329101562,
      "teacher_distance_mean": 0.8519236147403717,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.002433960884809494,
      "weighted_relation_loss": 0.002433960884809494
    },
    {
      "action_query_batch_std": 0.3496001183986664,
      "action_query_norm": 41.068386459350585,
      "action_relation_loss": 0.002516385493800044,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.869036482694106,
      "flow_matching_loss": 0.018967788107693194,
      "grad_norm": 0.08661983162164688,
      "h_stock_valid_norm": 97.8896697998047,
      "invalid_tail_samples": 25.7,
      "learning_rate": 2.155944016634398e-05,
      "loss": 0.0215,
      "positive_teacher_pairs": 26420.0,
      "query_h_norm_ratio": 0.41953891813755034,
      "relation_loss": 0.002516385493800044,
      "selected_pairs": 26420.0,
      "step": 21240,
      "student_distance_mean": 23.040003967285156,
      "teacher_distance_mean": 0.8422824919223786,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.002516385493800044,
      "weighted_relation_loss": 0.002516385493800044
    },
    {
      "action_query_batch_std": 0.3528449207544327,
      "action_query_norm": 41.10701026916504,
      "action_relation_loss": 0.002334035257808864,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.87839101964453,
      "flow_matching_loss": 0.019213375635445118,
      "grad_norm": 0.07817153632640839,
      "h_stock_valid_norm": 97.70742263793946,
      "invalid_tail_samples": 23.9,
      "learning_rate": 2.1514126580284538e-05,
      "loss": 0.0215,
      "positive_teacher_pairs": 26823.3,
      "query_h_norm_ratio": 0.4207163602113724,
      "relation_loss": 0.002334035257808864,
      "selected_pairs": 26823.3,
      "step": 21250,
      "student_distance_mean": 23.26838073730469,
      "teacher_distance_mean": 0.8513802528381348,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.002334035257808864,
      "weighted_relation_loss": 0.002334035257808864
    },
    {
      "action_query_batch_std": 0.35508295297622683,
      "action_query_norm": 41.101922607421876,
      "action_relation_loss": 0.0024533417541533708,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.88774555659495,
      "flow_matching_loss": 0.0194211196154356,
      "grad_norm": 0.11496840417385101,
      "h_stock_valid_norm": 97.44944229125977,
      "invalid_tail_samples": 21.4,
      "learning_rate": 2.1468847607247344e-05,
      "loss": 0.0219,
      "positive_teacher_pairs": 27406.0,
      "query_h_norm_ratio": 0.42178007364273074,
      "relation_loss": 0.0024533417541533708,
      "selected_pairs": 27406.0,
      "step": 21260,
      "student_distance_mean": 23.38440990447998,
      "teacher_distance_mean": 0.8561869978904724,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.0024533417541533708,
      "weighted_relation_loss": 0.0024533417541533708
    },
    {
      "action_query_batch_std": 0.35871967673301697,
      "action_query_norm": 41.1244514465332,
      "action_relation_loss": 0.002373825688846409,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.89710009354537,
      "flow_matching_loss": 0.020226328633725643,
      "grad_norm": 0.11558353900909424,
      "h_stock_valid_norm": 97.51172790527343,
      "invalid_tail_samples": 22.4,
      "learning_rate": 2.1423603302250623e-05,
      "loss": 0.0226,
      "positive_teacher_pairs": 27173.8,
      "query_h_norm_ratio": 0.4217421382665634,
      "relation_loss": 0.002373825688846409,
      "selected_pairs": 27173.8,
      "step": 21270,
      "student_distance_mean": 23.630062675476076,
      "teacher_distance_mean": 0.8644936978816986,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.002373825688846409,
      "weighted_relation_loss": 0.002373825688846409
    },
    {
      "action_query_batch_std": 0.3580430865287781,
      "action_query_norm": 41.11700248718262,
      "action_relation_loss": 0.0024156418396160005,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.90645463049579,
      "flow_matching_loss": 0.02113049030303955,
      "grad_norm": 0.30305489897727966,
      "h_stock_valid_norm": 97.67404479980469,
      "invalid_tail_samples": 23.6,
      "learning_rate": 2.1378393720270468e-05,
      "loss": 0.0235,
      "positive_teacher_pairs": 26893.0,
      "query_h_norm_ratio": 0.42096500396728515,
      "relation_loss": 0.0024156418396160005,
      "selected_pairs": 26893.0,
      "step": 21280,
      "student_distance_mean": 23.603069305419922,
      "teacher_distance_mean": 0.860652619600296,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0024156418396160005,
      "weighted_relation_loss": 0.0024156418396160005
    },
    {
      "action_query_batch_std": 0.35497220754623415,
      "action_query_norm": 41.11979866027832,
      "action_relation_loss": 0.002329081273637712,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.915809167446213,
      "flow_matching_loss": 0.020986630395054818,
      "grad_norm": 0.12086959183216095,
      "h_stock_valid_norm": 97.96223907470703,
      "invalid_tail_samples": 23.4,
      "learning_rate": 2.133321891624076e-05,
      "loss": 0.0233,
      "positive_teacher_pairs": 26942.4,
      "query_h_norm_ratio": 0.4197563260793686,
      "relation_loss": 0.002329081273637712,
      "selected_pairs": 26942.4,
      "step": 21290,
      "student_distance_mean": 23.320067596435546,
      "teacher_distance_mean": 0.8502300918102265,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002329081273637712,
      "weighted_relation_loss": 0.002329081273637712
    },
    {
      "action_query_batch_std": 0.3572104126214981,
      "action_query_norm": 41.08684463500977,
      "action_relation_loss": 0.002406806778162718,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.92516370439663,
      "flow_matching_loss": 0.020581818372011184,
      "grad_norm": 0.09749861061573029,
      "h_stock_valid_norm": 98.1005142211914,
      "invalid_tail_samples": 25.3,
      "learning_rate": 2.1288078945053193e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 26506.7,
      "query_h_norm_ratio": 0.4188264936208725,
      "relation_loss": 0.002406806778162718,
      "selected_pairs": 26506.7,
      "step": 21300,
      "student_distance_mean": 23.60292568206787,
      "teacher_distance_mean": 0.8636962234973907,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.002406806778162718,
      "weighted_relation_loss": 0.002406806778162718
    },
    {
      "action_query_batch_std": 0.35043210685253146,
      "action_query_norm": 41.05605239868164,
      "action_relation_loss": 0.0024716909043490886,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.934518241347053,
      "flow_matching_loss": 0.018970589712262155,
      "grad_norm": 0.09167185425758362,
      "h_stock_valid_norm": 98.5309341430664,
      "invalid_tail_samples": 25.2,
      "learning_rate": 2.1242973861557064e-05,
      "loss": 0.0214,
      "positive_teacher_pairs": 26523.8,
      "query_h_norm_ratio": 0.41668767631053927,
      "relation_loss": 0.0024716909043490886,
      "selected_pairs": 26523.8,
      "step": 21310,
      "student_distance_mean": 23.11498603820801,
      "teacher_distance_mean": 0.8453199684619903,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0024716909043490886,
      "weighted_relation_loss": 0.0024716909043490886
    },
    {
      "action_query_batch_std": 0.3577509313821793,
      "action_query_norm": 41.1181941986084,
      "action_relation_loss": 0.0024141987320035695,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.943872778297475,
      "flow_matching_loss": 0.01894173864275217,
      "grad_norm": 0.09991934895515442,
      "h_stock_valid_norm": 98.50388946533204,
      "invalid_tail_samples": 22.5,
      "learning_rate": 2.1197903720559304e-05,
      "loss": 0.0214,
      "positive_teacher_pairs": 27152.9,
      "query_h_norm_ratio": 0.41743132174015046,
      "relation_loss": 0.0024141987320035695,
      "selected_pairs": 27152.9,
      "step": 21320,
      "student_distance_mean": 23.650269508361816,
      "teacher_distance_mean": 0.8659355103969574,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0024141987320035695,
      "weighted_relation_loss": 0.0024141987320035695
    },
    {
      "action_query_batch_std": 0.35772926807403566,
      "action_query_norm": 41.09094924926758,
      "action_relation_loss": 0.002443296927958727,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.953227315247894,
      "flow_matching_loss": 0.02008818741887808,
      "grad_norm": 0.11249673366546631,
      "h_stock_valid_norm": 98.20913314819336,
      "invalid_tail_samples": 26.4,
      "learning_rate": 2.115286857682438e-05,
      "loss": 0.0225,
      "positive_teacher_pairs": 26255.2,
      "query_h_norm_ratio": 0.4184045553207397,
      "relation_loss": 0.002443296927958727,
      "selected_pairs": 26255.2,
      "step": 21330,
      "student_distance_mean": 23.603252029418947,
      "teacher_distance_mean": 0.8625416815280914,
      "valid_action_samples": 229.6,
      "weighted_auxiliary_loss": 0.002443296927958727,
      "weighted_relation_loss": 0.002443296927958727
    },
    {
      "action_query_batch_std": 0.3553559184074402,
      "action_query_norm": 41.097220993041994,
      "action_relation_loss": 0.0023639410035684707,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.962581852198316,
      "flow_matching_loss": 0.022018645703792573,
      "grad_norm": 0.11080475151538849,
      "h_stock_valid_norm": 98.25602416992187,
      "invalid_tail_samples": 24.3,
      "learning_rate": 2.1107868485074228e-05,
      "loss": 0.0244,
      "positive_teacher_pairs": 26736.3,
      "query_h_norm_ratio": 0.4182675749063492,
      "relation_loss": 0.0023639410035684707,
      "selected_pairs": 26736.3,
      "step": 21340,
      "student_distance_mean": 23.454302024841308,
      "teacher_distance_mean": 0.8567478835582734,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0023639410035684707,
      "weighted_relation_loss": 0.0023639410035684707
    },
    {
      "action_query_batch_std": 0.35553973615169526,
      "action_query_norm": 41.07160186767578,
      "action_relation_loss": 0.0024963817792013287,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.971936389148738,
      "flow_matching_loss": 0.0232964594848454,
      "grad_norm": 0.11943913251161575,
      "h_stock_valid_norm": 97.86240615844727,
      "invalid_tail_samples": 22.0,
      "learning_rate": 2.1062903499988234e-05,
      "loss": 0.0258,
      "positive_teacher_pairs": 27263.0,
      "query_h_norm_ratio": 0.41969078183174136,
      "relation_loss": 0.0024963817792013287,
      "selected_pairs": 27263.0,
      "step": 21350,
      "student_distance_mean": 23.39482002258301,
      "teacher_distance_mean": 0.856287705898285,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.0024963817792013287,
      "weighted_relation_loss": 0.0024963817792013287
    },
    {
      "action_query_batch_std": 0.35680682957172394,
      "action_query_norm": 41.094612503051756,
      "action_relation_loss": 0.0023271954152733086,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.98129092609916,
      "flow_matching_loss": 0.01937800841405988,
      "grad_norm": 0.09530945122241974,
      "h_stock_valid_norm": 97.64230728149414,
      "invalid_tail_samples": 26.9,
      "learning_rate": 2.101797367620308e-05,
      "loss": 0.0217,
      "positive_teacher_pairs": 26132.5,
      "query_h_norm_ratio": 0.4208699643611908,
      "relation_loss": 0.0023271954152733086,
      "selected_pairs": 26132.5,
      "step": 21360,
      "student_distance_mean": 23.58305072784424,
      "teacher_distance_mean": 0.8639300227165222,
      "valid_action_samples": 229.1,
      "weighted_auxiliary_loss": 0.0023271954152733086,
      "weighted_relation_loss": 0.0023271954152733086
    },
    {
      "action_query_batch_std": 0.35427753925323485,
      "action_query_norm": 41.07653312683105,
      "action_relation_loss": 0.0025251065148040653,
      "auxiliary_loss_weight": 1.0,
      "epoch": 19.990645463049578,
      "flow_matching_loss": 0.020457311533391476,
      "grad_norm": 0.09968598932027817,
      "h_stock_valid_norm": 97.68145294189453,
      "invalid_tail_samples": 23.4,
      "learning_rate": 2.0973079068312713e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 26946.2,
      "query_h_norm_ratio": 0.42051844894886015,
      "relation_loss": 0.0025251065148040653,
      "selected_pairs": 26946.2,
      "step": 21370,
      "student_distance_mean": 23.375032424926758,
      "teacher_distance_mean": 0.857511454820633,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0025251065148040653,
      "weighted_relation_loss": 0.0025251065148040653
    },
    {
      "action_query_batch_std": 0.35771481692790985,
      "action_query_norm": 41.117269134521486,
      "action_relation_loss": 0.0024040362797677515,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.0,
      "flow_matching_loss": 0.02208117563277483,
      "grad_norm": 0.199631929397583,
      "h_stock_valid_norm": 97.66683349609374,
      "invalid_tail_samples": 22.2,
      "learning_rate": 2.0928219730868358e-05,
      "loss": 0.0245,
      "positive_teacher_pairs": 24176.4,
      "query_h_norm_ratio": 0.42100304961204527,
      "relation_loss": 0.0024040362797677515,
      "selected_pairs": 24176.4,
      "step": 21380,
      "student_distance_mean": 23.673295783996583,
      "teacher_distance_mean": 0.8634997725486755,
      "valid_action_samples": 213.9,
      "weighted_auxiliary_loss": 0.0024040362797677515,
      "weighted_relation_loss": 0.0024040362797677515
    },
    {
      "action_query_batch_std": 0.3508912682533264,
      "action_query_norm": 41.053609466552736,
      "action_relation_loss": 0.002433061436749995,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.009354536950422,
      "flow_matching_loss": 0.021155987121164798,
      "grad_norm": 0.15969818830490112,
      "h_stock_valid_norm": 97.78216857910157,
      "invalid_tail_samples": 24.3,
      "learning_rate": 2.0883395718378302e-05,
      "loss": 0.0236,
      "positive_teacher_pairs": 26738.5,
      "query_h_norm_ratio": 0.4198497414588928,
      "relation_loss": 0.002433061436749995,
      "selected_pairs": 26738.5,
      "step": 21390,
      "student_distance_mean": 23.181756591796876,
      "teacher_distance_mean": 0.8456033170223236,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002433061436749995,
      "weighted_relation_loss": 0.002433061436749995
    },
    {
      "action_query_batch_std": 0.35489111244678495,
      "action_query_norm": 41.068415832519534,
      "action_relation_loss": 0.002211524057202041,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.01870907390084,
      "flow_matching_loss": 0.019307220727205275,
      "grad_norm": 0.11012116819620132,
      "h_stock_valid_norm": 98.23711700439453,
      "invalid_tail_samples": 21.6,
      "learning_rate": 2.083860708530798e-05,
      "loss": 0.0215,
      "positive_teacher_pairs": 27358.8,
      "query_h_norm_ratio": 0.41805807054042815,
      "relation_loss": 0.002211524057202041,
      "selected_pairs": 27358.8,
      "step": 21400,
      "student_distance_mean": 23.31810188293457,
      "teacher_distance_mean": 0.8532844603061676,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.002211524057202041,
      "weighted_relation_loss": 0.002211524057202041
    },
    {
      "action_query_batch_std": 0.3536251187324524,
      "action_query_norm": 41.07903060913086,
      "action_relation_loss": 0.0023255666717886926,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.028063610851262,
      "flow_matching_loss": 0.021358259953558445,
      "grad_norm": 0.09548547863960266,
      "h_stock_valid_norm": 98.33829040527344,
      "invalid_tail_samples": 21.4,
      "learning_rate": 2.0793853886079796e-05,
      "loss": 0.0237,
      "positive_teacher_pairs": 27408.7,
      "query_h_norm_ratio": 0.4177356094121933,
      "relation_loss": 0.0023255666717886926,
      "selected_pairs": 27408.7,
      "step": 21410,
      "student_distance_mean": 23.303352546691894,
      "teacher_distance_mean": 0.8520319998264313,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.0023255666717886926,
      "weighted_relation_loss": 0.0023255666717886926
    },
    {
      "action_query_batch_std": 0.3562025725841522,
      "action_query_norm": 41.09241561889648,
      "action_relation_loss": 0.002184440474957228,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.037418147801684,
      "flow_matching_loss": 0.018909294065088032,
      "grad_norm": 0.11678414791822433,
      "h_stock_valid_norm": 98.41162414550782,
      "invalid_tail_samples": 26.0,
      "learning_rate": 2.074913617507309e-05,
      "loss": 0.0211,
      "positive_teacher_pairs": 26354.4,
      "query_h_norm_ratio": 0.41755781769752504,
      "relation_loss": 0.002184440474957228,
      "selected_pairs": 26354.4,
      "step": 21420,
      "student_distance_mean": 23.506536293029786,
      "teacher_distance_mean": 0.8614849925041199,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.002184440474957228,
      "weighted_relation_loss": 0.002184440474957228
    },
    {
      "action_query_batch_std": 0.3507415443658829,
      "action_query_norm": 41.06829681396484,
      "action_relation_loss": 0.002299749571830034,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.046772684752106,
      "flow_matching_loss": 0.020598297752439976,
      "grad_norm": 0.10253921151161194,
      "h_stock_valid_norm": 98.61096649169922,
      "invalid_tail_samples": 21.8,
      "learning_rate": 2.070445400662411e-05,
      "loss": 0.0229,
      "positive_teacher_pairs": 27312.9,
      "query_h_norm_ratio": 0.41647201478481294,
      "relation_loss": 0.002299749571830034,
      "selected_pairs": 27312.9,
      "step": 21430,
      "student_distance_mean": 23.151806640625,
      "teacher_distance_mean": 0.848334151506424,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.002299749571830034,
      "weighted_relation_loss": 0.002299749571830034
    },
    {
      "action_query_batch_std": 0.35290609002113343,
      "action_query_norm": 41.068791198730466,
      "action_relation_loss": 0.002343347785063088,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.056127221702525,
      "flow_matching_loss": 0.018454311322420834,
      "grad_norm": 0.1114642322063446,
      "h_stock_valid_norm": 98.50767669677734,
      "invalid_tail_samples": 21.2,
      "learning_rate": 2.0659807435025907e-05,
      "loss": 0.0208,
      "positive_teacher_pairs": 27452.7,
      "query_h_norm_ratio": 0.4169120997190475,
      "relation_loss": 0.002343347785063088,
      "selected_pairs": 27452.7,
      "step": 21440,
      "student_distance_mean": 23.271577072143554,
      "teacher_distance_mean": 0.8506998181343078,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.002343347785063088,
      "weighted_relation_loss": 0.002343347785063088
    },
    {
      "action_query_batch_std": 0.35828558802604676,
      "action_query_norm": 41.10804824829101,
      "action_relation_loss": 0.0022442884743213653,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.065481758652947,
      "flow_matching_loss": 0.01930547272786498,
      "grad_norm": 0.09811097383499146,
      "h_stock_valid_norm": 98.45193252563476,
      "invalid_tail_samples": 25.0,
      "learning_rate": 2.061519651452825e-05,
      "loss": 0.0215,
      "positive_teacher_pairs": 26577.0,
      "query_h_norm_ratio": 0.41754749715328215,
      "relation_loss": 0.0022442884743213653,
      "selected_pairs": 26577.0,
      "step": 21450,
      "student_distance_mean": 23.646924209594726,
      "teacher_distance_mean": 0.8634220242500306,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0022442884743213653,
      "weighted_relation_loss": 0.0022442884743213653
    },
    {
      "action_query_batch_std": 0.3562535524368286,
      "action_query_norm": 41.11012077331543,
      "action_relation_loss": 0.002258172375150025,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.07483629560337,
      "flow_matching_loss": 0.019719632528722286,
      "grad_norm": 0.09237167239189148,
      "h_stock_valid_norm": 98.29825286865234,
      "invalid_tail_samples": 21.5,
      "learning_rate": 2.0570621299337656e-05,
      "loss": 0.022,
      "positive_teacher_pairs": 27388.4,
      "query_h_norm_ratio": 0.41821915209293364,
      "relation_loss": 0.002258172375150025,
      "selected_pairs": 27388.4,
      "step": 21460,
      "student_distance_mean": 23.467362213134766,
      "teacher_distance_mean": 0.8582565307617187,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.002258172375150025,
      "weighted_relation_loss": 0.002258172375150025
    },
    {
      "action_query_batch_std": 0.3547758966684341,
      "action_query_norm": 41.074210739135744,
      "action_relation_loss": 0.0024181450949981808,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.084190832553787,
      "flow_matching_loss": 0.02084645889699459,
      "grad_norm": 0.10750094801187515,
      "h_stock_valid_norm": 98.43445053100587,
      "invalid_tail_samples": 25.1,
      "learning_rate": 2.0526081843617183e-05,
      "loss": 0.0233,
      "positive_teacher_pairs": 26558.4,
      "query_h_norm_ratio": 0.4172757208347321,
      "relation_loss": 0.0024181450949981808,
      "selected_pairs": 26558.4,
      "step": 21470,
      "student_distance_mean": 23.514167976379394,
      "teacher_distance_mean": 0.8597387671470642,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0024181450949981808,
      "weighted_relation_loss": 0.0024181450949981808
    },
    {
      "action_query_batch_std": 0.35219936072826385,
      "action_query_norm": 41.05126190185547,
      "action_relation_loss": 0.0023883525282144547,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.09354536950421,
      "flow_matching_loss": 0.021026354655623435,
      "grad_norm": 0.09123481065034866,
      "h_stock_valid_norm": 98.23166046142578,
      "invalid_tail_samples": 23.2,
      "learning_rate": 2.0481578201486485e-05,
      "loss": 0.0234,
      "positive_teacher_pairs": 26984.8,
      "query_h_norm_ratio": 0.4179043143987656,
      "relation_loss": 0.0023883525282144547,
      "selected_pairs": 26984.8,
      "step": 21480,
      "student_distance_mean": 23.194408226013184,
      "teacher_distance_mean": 0.8502972424030304,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0023883525282144547,
      "weighted_relation_loss": 0.0023883525282144547
    },
    {
      "action_query_batch_std": 0.3515554517507553,
      "action_query_norm": 41.039730072021484,
      "action_relation_loss": 0.002298139384947717,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.10289990645463,
      "flow_matching_loss": 0.017921657115221024,
      "grad_norm": 0.10005924105644226,
      "h_stock_valid_norm": 98.11813583374024,
      "invalid_tail_samples": 25.9,
      "learning_rate": 2.0437110427021677e-05,
      "loss": 0.0202,
      "positive_teacher_pairs": 26360.7,
      "query_h_norm_ratio": 0.4182703346014023,
      "relation_loss": 0.002298139384947717,
      "selected_pairs": 26360.7,
      "step": 21490,
      "student_distance_mean": 23.09849319458008,
      "teacher_distance_mean": 0.8449884593486786,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.002298139384947717,
      "weighted_relation_loss": 0.002298139384947717
    },
    {
      "action_query_batch_std": 0.354197558760643,
      "action_query_norm": 41.05708045959473,
      "action_relation_loss": 0.0023969741305336356,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.112254443405053,
      "flow_matching_loss": 0.019343852065503597,
      "grad_norm": 0.15996143221855164,
      "h_stock_valid_norm": 98.00337448120118,
      "invalid_tail_samples": 24.7,
      "learning_rate": 2.0392678574255282e-05,
      "loss": 0.0217,
      "positive_teacher_pairs": 26643.8,
      "query_h_norm_ratio": 0.4189404398202896,
      "relation_loss": 0.0023969741305336356,
      "selected_pairs": 26643.8,
      "step": 21500,
      "student_distance_mean": 23.36863422393799,
      "teacher_distance_mean": 0.8528348207473755,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0023969741305336356,
      "weighted_relation_loss": 0.0023969741305336356
    },
    {
      "action_query_batch_std": 0.35448374450206754,
      "action_query_norm": 41.0783332824707,
      "action_relation_loss": 0.0023501104209572078,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.12160898035547,
      "flow_matching_loss": 0.01987286191433668,
      "grad_norm": 0.09164662659168243,
      "h_stock_valid_norm": 97.67416687011719,
      "invalid_tail_samples": 22.1,
      "learning_rate": 2.0348282697176217e-05,
      "loss": 0.0222,
      "positive_teacher_pairs": 27246.9,
      "query_h_norm_ratio": 0.4205667555332184,
      "relation_loss": 0.0023501104209572078,
      "selected_pairs": 27246.9,
      "step": 21510,
      "student_distance_mean": 23.293946838378908,
      "teacher_distance_mean": 0.8552718877792358,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0023501104209572078,
      "weighted_relation_loss": 0.0023501104209572078
    },
    {
      "action_query_batch_std": 0.35192869007587435,
      "action_query_norm": 41.0475341796875,
      "action_relation_loss": 0.002545885811559856,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.130963517305894,
      "flow_matching_loss": 0.020748252235352992,
      "grad_norm": 0.10265166312456131,
      "h_stock_valid_norm": 97.62122650146485,
      "invalid_tail_samples": 22.3,
      "learning_rate": 2.030392284972964e-05,
      "loss": 0.0233,
      "positive_teacher_pairs": 27201.9,
      "query_h_norm_ratio": 0.4204801023006439,
      "relation_loss": 0.002545885811559856,
      "selected_pairs": 27201.9,
      "step": 21520,
      "student_distance_mean": 23.181599044799803,
      "teacher_distance_mean": 0.849888551235199,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.002545885811559856,
      "weighted_relation_loss": 0.002545885811559856
    },
    {
      "action_query_batch_std": 0.3510326355695724,
      "action_query_norm": 41.04510765075683,
      "action_relation_loss": 0.0023368707858026027,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.140318054256316,
      "flow_matching_loss": 0.019123362191021443,
      "grad_norm": 0.08194886147975922,
      "h_stock_valid_norm": 97.78994903564453,
      "invalid_tail_samples": 23.4,
      "learning_rate": 2.0259599085816973e-05,
      "loss": 0.0215,
      "positive_teacher_pairs": 26943.3,
      "query_h_norm_ratio": 0.41973120868206026,
      "relation_loss": 0.0023368707858026027,
      "selected_pairs": 26943.3,
      "step": 21530,
      "student_distance_mean": 23.218228340148926,
      "teacher_distance_mean": 0.8501653790473938,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0023368707858026027,
      "weighted_relation_loss": 0.0023368707858026027
    },
    {
      "action_query_batch_std": 0.35233975350856783,
      "action_query_norm": 41.028630065917966,
      "action_relation_loss": 0.0023320928448811173,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.149672591206734,
      "flow_matching_loss": 0.0203684164211154,
      "grad_norm": 0.10421299934387207,
      "h_stock_valid_norm": 98.02361679077148,
      "invalid_tail_samples": 24.9,
      "learning_rate": 2.0215311459295755e-05,
      "loss": 0.0227,
      "positive_teacher_pairs": 26604.4,
      "query_h_norm_ratio": 0.4185622215270996,
      "relation_loss": 0.0023320928448811173,
      "selected_pairs": 26604.4,
      "step": 21540,
      "student_distance_mean": 23.328689956665038,
      "teacher_distance_mean": 0.8552115499973297,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0023320928448811173,
      "weighted_relation_loss": 0.0023320928448811173
    },
    {
      "action_query_batch_std": 0.3521490156650543,
      "action_query_norm": 41.035888290405275,
      "action_relation_loss": 0.00235490829218179,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.159027128157156,
      "flow_matching_loss": 0.02091907262802124,
      "grad_norm": 0.09710785001516342,
      "h_stock_valid_norm": 98.33877258300781,
      "invalid_tail_samples": 25.7,
      "learning_rate": 2.0171060023979606e-05,
      "loss": 0.0233,
      "positive_teacher_pairs": 26413.9,
      "query_h_norm_ratio": 0.41729271709918975,
      "relation_loss": 0.00235490829218179,
      "selected_pairs": 26413.9,
      "step": 21550,
      "student_distance_mean": 23.36631488800049,
      "teacher_distance_mean": 0.8554230749607086,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.00235490829218179,
      "weighted_relation_loss": 0.00235490829218179
    },
    {
      "action_query_batch_std": 0.3525861382484436,
      "action_query_norm": 41.05697517395019,
      "action_relation_loss": 0.0022776711732149122,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.168381665107578,
      "flow_matching_loss": 0.01974712535738945,
      "grad_norm": 0.09982693195343018,
      "h_stock_valid_norm": 98.28556976318359,
      "invalid_tail_samples": 25.1,
      "learning_rate": 2.012684483363823e-05,
      "loss": 0.022,
      "positive_teacher_pairs": 26552.2,
      "query_h_norm_ratio": 0.417732959985733,
      "relation_loss": 0.0022776711732149122,
      "selected_pairs": 26552.2,
      "step": 21560,
      "student_distance_mean": 23.232019233703614,
      "teacher_distance_mean": 0.8524809360504151,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0022776711732149122,
      "weighted_relation_loss": 0.0022776711732149122
    },
    {
      "action_query_batch_std": 0.35481260120868685,
      "action_query_norm": 41.06956062316895,
      "action_relation_loss": 0.0023170117055997252,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.177736202057996,
      "flow_matching_loss": 0.018706829473376273,
      "grad_norm": 0.09853038936853409,
      "h_stock_valid_norm": 98.09704208374023,
      "invalid_tail_samples": 24.4,
      "learning_rate": 2.0082665941997237e-05,
      "loss": 0.021,
      "positive_teacher_pairs": 26717.2,
      "query_h_norm_ratio": 0.41866492927074433,
      "relation_loss": 0.0023170117055997252,
      "selected_pairs": 26717.2,
      "step": 21570,
      "student_distance_mean": 23.288229751586915,
      "teacher_distance_mean": 0.8561842918395997,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0023170117055997252,
      "weighted_relation_loss": 0.0023170117055997252
    },
    {
      "action_query_batch_std": 0.35322259962558744,
      "action_query_norm": 41.07120056152344,
      "action_relation_loss": 0.002266060374677181,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.18709073900842,
      "flow_matching_loss": 0.020013961754739285,
      "grad_norm": 0.10415330529212952,
      "h_stock_valid_norm": 98.0959243774414,
      "invalid_tail_samples": 23.1,
      "learning_rate": 2.0038523402738146e-05,
      "loss": 0.0223,
      "positive_teacher_pairs": 27012.1,
      "query_h_norm_ratio": 0.41868590414524076,
      "relation_loss": 0.002266060374677181,
      "selected_pairs": 27012.1,
      "step": 21580,
      "student_distance_mean": 23.3031436920166,
      "teacher_distance_mean": 0.8544401943683624,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002266060374677181,
      "weighted_relation_loss": 0.002266060374677181
    },
    {
      "action_query_batch_std": 0.35336015820503236,
      "action_query_norm": 41.06555862426758,
      "action_relation_loss": 0.0023508024867624046,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.19644527595884,
      "flow_matching_loss": 0.019518831558525563,
      "grad_norm": 0.1598234325647354,
      "h_stock_valid_norm": 98.20610733032227,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.99944172694983e-05,
      "loss": 0.0219,
      "positive_teacher_pairs": 26811.3,
      "query_h_norm_ratio": 0.4181589901447296,
      "relation_loss": 0.0023508024867624046,
      "selected_pairs": 26811.3,
      "step": 21590,
      "student_distance_mean": 23.330523681640624,
      "teacher_distance_mean": 0.8543040752410889,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0023508024867624046,
      "weighted_relation_loss": 0.0023508024867624046
    },
    {
      "action_query_batch_std": 0.3528548628091812,
      "action_query_norm": 41.049189758300784,
      "action_relation_loss": 0.0022195513593032956,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.205799812909262,
      "flow_matching_loss": 0.020435584336519243,
      "grad_norm": 0.09306608885526657,
      "h_stock_valid_norm": 98.1911605834961,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.99503475958708e-05,
      "loss": 0.0227,
      "positive_teacher_pairs": 27094.4,
      "query_h_norm_ratio": 0.4180596500635147,
      "relation_loss": 0.0022195513593032956,
      "selected_pairs": 27094.4,
      "step": 21600,
      "student_distance_mean": 23.27682228088379,
      "teacher_distance_mean": 0.8546176552772522,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0022195513593032956,
      "weighted_relation_loss": 0.0022195513593032956
    },
    {
      "action_query_batch_std": 0.3566500276327133,
      "action_query_norm": 41.066627502441406,
      "action_relation_loss": 0.0022668152349069716,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.21515434985968,
      "flow_matching_loss": 0.01964399553835392,
      "grad_norm": 0.09626005589962006,
      "h_stock_valid_norm": 98.62735748291016,
      "invalid_tail_samples": 24.7,
      "learning_rate": 1.9906314435404484e-05,
      "loss": 0.0219,
      "positive_teacher_pairs": 26645.8,
      "query_h_norm_ratio": 0.4163830041885376,
      "relation_loss": 0.0022668152349069716,
      "selected_pairs": 26645.8,
      "step": 21610,
      "student_distance_mean": 23.39410572052002,
      "teacher_distance_mean": 0.8569809019565582,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0022668152349069716,
      "weighted_relation_loss": 0.0022668152349069716
    },
    {
      "action_query_batch_std": 0.35052470564842225,
      "action_query_norm": 41.040060424804686,
      "action_relation_loss": 0.0023032369092106817,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.224508886810103,
      "flow_matching_loss": 0.01932609435170889,
      "grad_norm": 0.08889070898294449,
      "h_stock_valid_norm": 98.20409545898437,
      "invalid_tail_samples": 22.2,
      "learning_rate": 1.986231784160378e-05,
      "loss": 0.0216,
      "positive_teacher_pairs": 27220.9,
      "query_h_norm_ratio": 0.41790762543678284,
      "relation_loss": 0.0023032369092106817,
      "selected_pairs": 27220.9,
      "step": 21620,
      "student_distance_mean": 23.13356456756592,
      "teacher_distance_mean": 0.8488250315189362,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0023032369092106817,
      "weighted_relation_loss": 0.0023032369092106817
    },
    {
      "action_query_batch_std": 0.35540677309036256,
      "action_query_norm": 41.04880561828613,
      "action_relation_loss": 0.00236958060413599,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.233863423760525,
      "flow_matching_loss": 0.02097729854285717,
      "grad_norm": 0.0996338427066803,
      "h_stock_valid_norm": 98.2158576965332,
      "invalid_tail_samples": 24.5,
      "learning_rate": 1.9818357867928693e-05,
      "loss": 0.0233,
      "positive_teacher_pairs": 26689.7,
      "query_h_norm_ratio": 0.4179467648267746,
      "relation_loss": 0.00236958060413599,
      "selected_pairs": 26689.7,
      "step": 21630,
      "student_distance_mean": 23.370011138916016,
      "teacher_distance_mean": 0.8578679919242859,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.00236958060413599,
      "weighted_relation_loss": 0.00236958060413599
    },
    {
      "action_query_batch_std": 0.3537587523460388,
      "action_query_norm": 41.060818099975585,
      "action_relation_loss": 0.00231046078260988,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.243217960710943,
      "flow_matching_loss": 0.020845812372863294,
      "grad_norm": 0.15203358232975006,
      "h_stock_valid_norm": 98.21369400024415,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.977443456779474e-05,
      "loss": 0.0232,
      "positive_teacher_pairs": 26761.5,
      "query_h_norm_ratio": 0.41808036267757415,
      "relation_loss": 0.00231046078260988,
      "selected_pairs": 26761.5,
      "step": 21640,
      "student_distance_mean": 23.306848335266114,
      "teacher_distance_mean": 0.8533181846141815,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.00231046078260988,
      "weighted_relation_loss": 0.00231046078260988
    },
    {
      "action_query_batch_std": 0.3558522671461105,
      "action_query_norm": 41.05347442626953,
      "action_relation_loss": 0.0023099427577108144,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.252572497661365,
      "flow_matching_loss": 0.020640315767377616,
      "grad_norm": 0.09343203157186508,
      "h_stock_valid_norm": 98.21760711669921,
      "invalid_tail_samples": 24.9,
      "learning_rate": 1.9730547994572858e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 26594.9,
      "query_h_norm_ratio": 0.41798681318759917,
      "relation_loss": 0.0023099427577108144,
      "selected_pairs": 26594.9,
      "step": 21650,
      "student_distance_mean": 23.377012252807617,
      "teacher_distance_mean": 0.855999332666397,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0023099427577108144,
      "weighted_relation_loss": 0.0023099427577108144
    },
    {
      "action_query_batch_std": 0.3527176886796951,
      "action_query_norm": 41.04461441040039,
      "action_relation_loss": 0.002467458578757942,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.261927034611787,
      "flow_matching_loss": 0.020617585349828004,
      "grad_norm": 0.10113399475812912,
      "h_stock_valid_norm": 98.36528930664062,
      "invalid_tail_samples": 26.2,
      "learning_rate": 1.9686698201589393e-05,
      "loss": 0.0231,
      "positive_teacher_pairs": 26300.7,
      "query_h_norm_ratio": 0.4172679841518402,
      "relation_loss": 0.002467458578757942,
      "selected_pairs": 26300.7,
      "step": 21660,
      "student_distance_mean": 23.196188545227052,
      "teacher_distance_mean": 0.8517269551753998,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.002467458578757942,
      "weighted_relation_loss": 0.002467458578757942
    },
    {
      "action_query_batch_std": 0.35127354562282564,
      "action_query_norm": 41.02847900390625,
      "action_relation_loss": 0.0023789691273123025,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.27128157156221,
      "flow_matching_loss": 0.020791278406977653,
      "grad_norm": 0.10498250275850296,
      "h_stock_valid_norm": 98.47248229980468,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.9642885242125964e-05,
      "loss": 0.0232,
      "positive_teacher_pairs": 27103.9,
      "query_h_norm_ratio": 0.41665092706680296,
      "relation_loss": 0.0023789691273123025,
      "selected_pairs": 27103.9,
      "step": 21670,
      "student_distance_mean": 23.113460540771484,
      "teacher_distance_mean": 0.8498983800411224,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0023789691273123025,
      "weighted_relation_loss": 0.0023789691273123025
    },
    {
      "action_query_batch_std": 0.35056039690971375,
      "action_query_norm": 41.00428695678711,
      "action_relation_loss": 0.002309318934567273,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.280636108512628,
      "flow_matching_loss": 0.01997268684208393,
      "grad_norm": 0.09687624126672745,
      "h_stock_valid_norm": 98.39494934082032,
      "invalid_tail_samples": 26.6,
      "learning_rate": 1.9599109169419466e-05,
      "loss": 0.0223,
      "positive_teacher_pairs": 26213.2,
      "query_h_norm_ratio": 0.41673611402511596,
      "relation_loss": 0.002309318934567273,
      "selected_pairs": 26213.2,
      "step": 21680,
      "student_distance_mean": 23.140028190612792,
      "teacher_distance_mean": 0.8503220736980438,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.002309318934567273,
      "weighted_relation_loss": 0.002309318934567273
    },
    {
      "action_query_batch_std": 0.3582870364189148,
      "action_query_norm": 41.04278984069824,
      "action_relation_loss": 0.002376033691689372,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.28999064546305,
      "flow_matching_loss": 0.01986974198371172,
      "grad_norm": 0.1460157185792923,
      "h_stock_valid_norm": 98.14044647216797,
      "invalid_tail_samples": 26.0,
      "learning_rate": 1.9555370036661948e-05,
      "loss": 0.0222,
      "positive_teacher_pairs": 26351.1,
      "query_h_norm_ratio": 0.41820585131645205,
      "relation_loss": 0.002376033691689372,
      "selected_pairs": 26351.1,
      "step": 21690,
      "student_distance_mean": 23.6151123046875,
      "teacher_distance_mean": 0.8657501876354218,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.002376033691689372,
      "weighted_relation_loss": 0.002376033691689372
    },
    {
      "action_query_batch_std": 0.35677351951599123,
      "action_query_norm": 41.06456451416015,
      "action_relation_loss": 0.0022774202283471824,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.29934518241347,
      "flow_matching_loss": 0.01960767265409231,
      "grad_norm": 0.09120769798755646,
      "h_stock_valid_norm": 97.99837722778321,
      "invalid_tail_samples": 20.0,
      "learning_rate": 1.9511667897000574e-05,
      "loss": 0.0219,
      "positive_teacher_pairs": 27743.2,
      "query_h_norm_ratio": 0.4190338790416718,
      "relation_loss": 0.0022774202283471824,
      "selected_pairs": 27743.2,
      "step": 21700,
      "student_distance_mean": 23.475227928161623,
      "teacher_distance_mean": 0.8618841886520385,
      "valid_action_samples": 236.0,
      "weighted_auxiliary_loss": 0.0022774202283471824,
      "weighted_relation_loss": 0.0022774202283471824
    },
    {
      "action_query_batch_std": 0.35095253586769104,
      "action_query_norm": 41.01750030517578,
      "action_relation_loss": 0.0023975386517122387,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.30869971936389,
      "flow_matching_loss": 0.020070871524512767,
      "grad_norm": 0.12439413368701935,
      "h_stock_valid_norm": 98.3128448486328,
      "invalid_tail_samples": 23.5,
      "learning_rate": 1.946800280353755e-05,
      "loss": 0.0225,
      "positive_teacher_pairs": 26919.3,
      "query_h_norm_ratio": 0.41721666157245635,
      "relation_loss": 0.0023975386517122387,
      "selected_pairs": 26919.3,
      "step": 21710,
      "student_distance_mean": 23.11442279815674,
      "teacher_distance_mean": 0.8516731142997742,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0023975386517122387,
      "weighted_relation_loss": 0.0023975386517122387
    },
    {
      "action_query_batch_std": 0.3515661060810089,
      "action_query_norm": 41.009314346313474,
      "action_relation_loss": 0.002278072130866349,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.318054256314312,
      "flow_matching_loss": 0.018740891572088005,
      "grad_norm": 0.12123086303472519,
      "h_stock_valid_norm": 98.23733596801758,
      "invalid_tail_samples": 25.0,
      "learning_rate": 1.9424374809330114e-05,
      "loss": 0.021,
      "positive_teacher_pairs": 26574.7,
      "query_h_norm_ratio": 0.41745362877845765,
      "relation_loss": 0.002278072130866349,
      "selected_pairs": 26574.7,
      "step": 21720,
      "student_distance_mean": 23.07822780609131,
      "teacher_distance_mean": 0.8465249598026275,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.002278072130866349,
      "weighted_relation_loss": 0.002278072130866349
    },
    {
      "action_query_batch_std": 0.3574503302574158,
      "action_query_norm": 41.07299270629883,
      "action_relation_loss": 0.0023707979125902057,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.327408793264734,
      "flow_matching_loss": 0.019999714475125073,
      "grad_norm": 0.09489107877016068,
      "h_stock_valid_norm": 98.59155731201172,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.9380783967390382e-05,
      "loss": 0.0224,
      "positive_teacher_pairs": 26828.6,
      "query_h_norm_ratio": 0.41660008430480955,
      "relation_loss": 0.0023707979125902057,
      "selected_pairs": 26828.6,
      "step": 21730,
      "student_distance_mean": 23.59613742828369,
      "teacher_distance_mean": 0.8682154417037964,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0023707979125902057,
      "weighted_relation_loss": 0.0023707979125902057
    },
    {
      "action_query_batch_std": 0.3545467406511307,
      "action_query_norm": 41.03861427307129,
      "action_relation_loss": 0.0023393883369863034,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.336763330215156,
      "flow_matching_loss": 0.018930432479828598,
      "grad_norm": 0.10103140771389008,
      "h_stock_valid_norm": 98.2885643005371,
      "invalid_tail_samples": 23.2,
      "learning_rate": 1.933723033068533e-05,
      "loss": 0.0213,
      "positive_teacher_pairs": 26989.8,
      "query_h_norm_ratio": 0.41753461956977844,
      "relation_loss": 0.0023393883369863034,
      "selected_pairs": 26989.8,
      "step": 21740,
      "student_distance_mean": 23.331103324890137,
      "teacher_distance_mean": 0.8581591308116913,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0023393883369863034,
      "weighted_relation_loss": 0.0023393883369863034
    },
    {
      "action_query_batch_std": 0.35294999182224274,
      "action_query_norm": 41.00924644470215,
      "action_relation_loss": 0.0024237021803855895,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.346117867165574,
      "flow_matching_loss": 0.019832700304687024,
      "grad_norm": 0.16059298813343048,
      "h_stock_valid_norm": 98.28219909667969,
      "invalid_tail_samples": 25.4,
      "learning_rate": 1.929371395213674e-05,
      "loss": 0.0223,
      "positive_teacher_pairs": 26476.8,
      "query_h_norm_ratio": 0.4172612398862839,
      "relation_loss": 0.0024237021803855895,
      "selected_pairs": 26476.8,
      "step": 21750,
      "student_distance_mean": 23.237394714355467,
      "teacher_distance_mean": 0.8539641201496124,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0024237021803855895,
      "weighted_relation_loss": 0.0024237021803855895
    },
    {
      "action_query_batch_std": 0.3499897301197052,
      "action_query_norm": 41.0189094543457,
      "action_relation_loss": 0.0023466049460694194,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.355472404115996,
      "flow_matching_loss": 0.019006383046507834,
      "grad_norm": 0.11047320812940598,
      "h_stock_valid_norm": 98.1027717590332,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.9250234884621092e-05,
      "loss": 0.0214,
      "positive_teacher_pairs": 26893.6,
      "query_h_norm_ratio": 0.4181250810623169,
      "relation_loss": 0.0023466049460694194,
      "selected_pairs": 26893.6,
      "step": 21760,
      "student_distance_mean": 23.164582633972167,
      "teacher_distance_mean": 0.8489976823329926,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0023466049460694194,
      "weighted_relation_loss": 0.0023466049460694194
    },
    {
      "action_query_batch_std": 0.35447021424770353,
      "action_query_norm": 41.02370986938477,
      "action_relation_loss": 0.002385712810792029,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.36482694106642,
      "flow_matching_loss": 0.02003353536128998,
      "grad_norm": 0.08756469190120697,
      "h_stock_valid_norm": 98.12808609008789,
      "invalid_tail_samples": 24.1,
      "learning_rate": 1.920679318096959e-05,
      "loss": 0.0224,
      "positive_teacher_pairs": 26781.5,
      "query_h_norm_ratio": 0.41806638836860655,
      "relation_loss": 0.002385712810792029,
      "selected_pairs": 26781.5,
      "step": 21770,
      "student_distance_mean": 23.479391860961915,
      "teacher_distance_mean": 0.8607537090778351,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.002385712810792029,
      "weighted_relation_loss": 0.002385712810792029
    },
    {
      "action_query_batch_std": 0.3498883366584778,
      "action_query_norm": 41.005712890625,
      "action_relation_loss": 0.002345501189120114,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.374181478016837,
      "flow_matching_loss": 0.020845656096935273,
      "grad_norm": 0.18532179296016693,
      "h_stock_valid_norm": 98.06346740722657,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.9163388893967982e-05,
      "loss": 0.0232,
      "positive_teacher_pairs": 26751.3,
      "query_h_norm_ratio": 0.41816048622131347,
      "relation_loss": 0.002345501189120114,
      "selected_pairs": 26751.3,
      "step": 21780,
      "student_distance_mean": 23.090456199645995,
      "teacher_distance_mean": 0.8490921020507812,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002345501189120114,
      "weighted_relation_loss": 0.002345501189120114
    },
    {
      "action_query_batch_std": 0.3579398661851883,
      "action_query_norm": 41.050656127929685,
      "action_relation_loss": 0.0022776634665206074,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.38353601496726,
      "flow_matching_loss": 0.020457855705171824,
      "grad_norm": 0.11068091541528702,
      "h_stock_valid_norm": 98.50353546142578,
      "invalid_tail_samples": 25.0,
      "learning_rate": 1.9120022076356575e-05,
      "loss": 0.0227,
      "positive_teacher_pairs": 26578.3,
      "query_h_norm_ratio": 0.41674659252166746,
      "relation_loss": 0.0022776634665206074,
      "selected_pairs": 26578.3,
      "step": 21790,
      "student_distance_mean": 23.5451322555542,
      "teacher_distance_mean": 0.8667750298976898,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0022776634665206074,
      "weighted_relation_loss": 0.0022776634665206074
    },
    {
      "action_query_batch_std": 0.35341529846191405,
      "action_query_norm": 41.01119804382324,
      "action_relation_loss": 0.0022135223960503938,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.39289055191768,
      "flow_matching_loss": 0.019687497802078725,
      "grad_norm": 0.0991063043475151,
      "h_stock_valid_norm": 98.65928726196289,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.9076692780830114e-05,
      "loss": 0.0219,
      "positive_teacher_pairs": 26872.9,
      "query_h_norm_ratio": 0.4156925529241562,
      "relation_loss": 0.0022135223960503938,
      "selected_pairs": 26872.9,
      "step": 21800,
      "student_distance_mean": 23.209871101379395,
      "teacher_distance_mean": 0.8549434959888458,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0022135223960503938,
      "weighted_relation_loss": 0.0022135223960503938
    },
    {
      "action_query_batch_std": 0.3573841392993927,
      "action_query_norm": 41.05343132019043,
      "action_relation_loss": 0.002262817497830838,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.4022450888681,
      "flow_matching_loss": 0.018621262349188327,
      "grad_norm": 0.09832743555307388,
      "h_stock_valid_norm": 98.11487350463867,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.903340106003782e-05,
      "loss": 0.0209,
      "positive_teacher_pairs": 27103.6,
      "query_h_norm_ratio": 0.4184246152639389,
      "relation_loss": 0.002262817497830838,
      "selected_pairs": 27103.6,
      "step": 21810,
      "student_distance_mean": 23.472141075134278,
      "teacher_distance_mean": 0.8646336793899536,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.002262817497830838,
      "weighted_relation_loss": 0.002262817497830838
    },
    {
      "action_query_batch_std": 0.3508643418550491,
      "action_query_norm": 41.002814865112306,
      "action_relation_loss": 0.0022952925646677613,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.41159962581852,
      "flow_matching_loss": 0.01992616234347224,
      "grad_norm": 0.14128214120864868,
      "h_stock_valid_norm": 98.05573425292968,
      "invalid_tail_samples": 25.3,
      "learning_rate": 1.8990146966583183e-05,
      "loss": 0.0222,
      "positive_teacher_pairs": 26508.3,
      "query_h_norm_ratio": 0.41816283762454987,
      "relation_loss": 0.0022952925646677613,
      "selected_pairs": 26508.3,
      "step": 21820,
      "student_distance_mean": 23.19825973510742,
      "teacher_distance_mean": 0.8525031805038452,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0022952925646677613,
      "weighted_relation_loss": 0.0022952925646677613
    },
    {
      "action_query_batch_std": 0.35221806466579436,
      "action_query_norm": 40.99984283447266,
      "action_relation_loss": 0.0023371792165562512,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.420954162768943,
      "flow_matching_loss": 0.020800992473959924,
      "grad_norm": 0.10245228558778763,
      "h_stock_valid_norm": 98.0008659362793,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.8946930553024035e-05,
      "loss": 0.0231,
      "positive_teacher_pairs": 26831.3,
      "query_h_norm_ratio": 0.4183668166399002,
      "relation_loss": 0.0023371792165562512,
      "selected_pairs": 26831.3,
      "step": 21830,
      "student_distance_mean": 23.14894218444824,
      "teacher_distance_mean": 0.8478293538093566,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0023371792165562512,
      "weighted_relation_loss": 0.0023371792165562512
    },
    {
      "action_query_batch_std": 0.35282736718654634,
      "action_query_norm": 41.01791229248047,
      "action_relation_loss": 0.002426998899318278,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.430308699719365,
      "flow_matching_loss": 0.020878909155726433,
      "grad_norm": 0.16609908640384674,
      "h_stock_valid_norm": 98.0553352355957,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.8903751871872378e-05,
      "loss": 0.0233,
      "positive_teacher_pairs": 27083.6,
      "query_h_norm_ratio": 0.418321031332016,
      "relation_loss": 0.002426998899318278,
      "selected_pairs": 27083.6,
      "step": 21840,
      "student_distance_mean": 23.25203056335449,
      "teacher_distance_mean": 0.8534398913383484,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.002426998899318278,
      "weighted_relation_loss": 0.002426998899318278
    },
    {
      "action_query_batch_std": 0.3508568346500397,
      "action_query_norm": 41.0088996887207,
      "action_relation_loss": 0.002251979405991733,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.439663236669784,
      "flow_matching_loss": 0.019524863362312316,
      "grad_norm": 0.09449224919080734,
      "h_stock_valid_norm": 98.16452484130859,
      "invalid_tail_samples": 22.3,
      "learning_rate": 1.8860610975594382e-05,
      "loss": 0.0218,
      "positive_teacher_pairs": 27196.7,
      "query_h_norm_ratio": 0.41776070892810824,
      "relation_loss": 0.002251979405991733,
      "selected_pairs": 27196.7,
      "step": 21850,
      "student_distance_mean": 23.170781517028807,
      "teacher_distance_mean": 0.8529504060745239,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.002251979405991733,
      "weighted_relation_loss": 0.002251979405991733
    },
    {
      "action_query_batch_std": 0.35219759941101075,
      "action_query_norm": 41.0157112121582,
      "action_relation_loss": 0.0023993955925107,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.449017773620206,
      "flow_matching_loss": 0.019954023975878955,
      "grad_norm": 0.10482773184776306,
      "h_stock_valid_norm": 98.11751556396484,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.8817507916610307e-05,
      "loss": 0.0224,
      "positive_teacher_pairs": 26897.6,
      "query_h_norm_ratio": 0.4180280417203903,
      "relation_loss": 0.0023993955925107,
      "selected_pairs": 26897.6,
      "step": 21860,
      "student_distance_mean": 23.178153800964356,
      "teacher_distance_mean": 0.855341249704361,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0023993955925107,
      "weighted_relation_loss": 0.0023993955925107
    },
    {
      "action_query_batch_std": 0.34945274889469147,
      "action_query_norm": 40.99850959777832,
      "action_relation_loss": 0.00235576368868351,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.458372310570628,
      "flow_matching_loss": 0.02156717423349619,
      "grad_norm": 0.10592044144868851,
      "h_stock_valid_norm": 98.2188819885254,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.8774442747294407e-05,
      "loss": 0.0239,
      "positive_teacher_pairs": 26890.6,
      "query_h_norm_ratio": 0.41742281913757323,
      "relation_loss": 0.00235576368868351,
      "selected_pairs": 26890.6,
      "step": 21870,
      "student_distance_mean": 22.931802177429198,
      "teacher_distance_mean": 0.8434747755527496,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.00235576368868351,
      "weighted_relation_loss": 0.00235576368868351
    },
    {
      "action_query_batch_std": 0.3542636543512344,
      "action_query_norm": 41.02711029052735,
      "action_relation_loss": 0.002322632004506886,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.467726847521046,
      "flow_matching_loss": 0.021302318759262563,
      "grad_norm": 0.128797248005867,
      "h_stock_valid_norm": 98.39573822021484,
      "invalid_tail_samples": 22.9,
      "learning_rate": 1.8731415519974965e-05,
      "loss": 0.0236,
      "positive_teacher_pairs": 27057.9,
      "query_h_norm_ratio": 0.4169656813144684,
      "relation_loss": 0.002322632004506886,
      "selected_pairs": 27057.9,
      "step": 21880,
      "student_distance_mean": 23.30941276550293,
      "teacher_distance_mean": 0.8541561841964722,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.002322632004506886,
      "weighted_relation_loss": 0.002322632004506886
    },
    {
      "action_query_batch_std": 0.35380028188228607,
      "action_query_norm": 41.01491584777832,
      "action_relation_loss": 0.002303566108457744,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.477081384471468,
      "flow_matching_loss": 0.01946809673681855,
      "grad_norm": 0.13113461434841156,
      "h_stock_valid_norm": 98.6115707397461,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.86884262869341e-05,
      "loss": 0.0218,
      "positive_teacher_pairs": 26800.1,
      "query_h_norm_ratio": 0.41592669785022734,
      "relation_loss": 0.002303566108457744,
      "selected_pairs": 26800.1,
      "step": 21890,
      "student_distance_mean": 23.14449825286865,
      "teacher_distance_mean": 0.8481038689613343,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.002303566108457744,
      "weighted_relation_loss": 0.002303566108457744
    },
    {
      "action_query_batch_std": 0.3576713979244232,
      "action_query_norm": 41.01951179504395,
      "action_relation_loss": 0.002328691561706364,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.48643592142189,
      "flow_matching_loss": 0.01968366503715515,
      "grad_norm": 0.09779343754053116,
      "h_stock_valid_norm": 98.9679183959961,
      "invalid_tail_samples": 21.6,
      "learning_rate": 1.8645475100407788e-05,
      "loss": 0.022,
      "positive_teacher_pairs": 27361.3,
      "query_h_norm_ratio": 0.41447383165359497,
      "relation_loss": 0.002328691561706364,
      "selected_pairs": 27361.3,
      "step": 21900,
      "student_distance_mean": 23.506632232666014,
      "teacher_distance_mean": 0.8653329849243164,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.002328691561706364,
      "weighted_relation_loss": 0.002328691561706364
    },
    {
      "action_query_batch_std": 0.35733542740345003,
      "action_query_norm": 41.02192726135254,
      "action_relation_loss": 0.002257665176875889,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.495790458372312,
      "flow_matching_loss": 0.019364064466208218,
      "grad_norm": 0.12488632649183273,
      "h_stock_valid_norm": 98.67206573486328,
      "invalid_tail_samples": 25.9,
      "learning_rate": 1.8602562012585768e-05,
      "loss": 0.0216,
      "positive_teacher_pairs": 26368.4,
      "query_h_norm_ratio": 0.4157423049211502,
      "relation_loss": 0.002257665176875889,
      "selected_pairs": 26368.4,
      "step": 21910,
      "student_distance_mean": 23.57096462249756,
      "teacher_distance_mean": 0.8695890307426453,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.002257665176875889,
      "weighted_relation_loss": 0.002257665176875889
    },
    {
      "action_query_batch_std": 0.3499072015285492,
      "action_query_norm": 40.99232940673828,
      "action_relation_loss": 0.0023296428844332695,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.50514499532273,
      "flow_matching_loss": 0.021206812001764774,
      "grad_norm": 0.15123067796230316,
      "h_stock_valid_norm": 98.56966934204101,
      "invalid_tail_samples": 22.9,
      "learning_rate": 1.8559687075611466e-05,
      "loss": 0.0235,
      "positive_teacher_pairs": 27059.1,
      "query_h_norm_ratio": 0.4158741652965546,
      "relation_loss": 0.0023296428844332695,
      "selected_pairs": 27059.1,
      "step": 21920,
      "student_distance_mean": 22.99987907409668,
      "teacher_distance_mean": 0.843986839056015,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0023296428844332695,
      "weighted_relation_loss": 0.0023296428844332695
    },
    {
      "action_query_batch_std": 0.35259418189525604,
      "action_query_norm": 41.02521591186523,
      "action_relation_loss": 0.002399705071002245,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.514499532273152,
      "flow_matching_loss": 0.021057757921516894,
      "grad_norm": 0.08835771679878235,
      "h_stock_valid_norm": 98.32023696899414,
      "invalid_tail_samples": 26.2,
      "learning_rate": 1.8516850341582017e-05,
      "loss": 0.0235,
      "positive_teacher_pairs": 26295.7,
      "query_h_norm_ratio": 0.4172633409500122,
      "relation_loss": 0.002399705071002245,
      "selected_pairs": 26295.7,
      "step": 21930,
      "student_distance_mean": 23.3246639251709,
      "teacher_distance_mean": 0.8569990634918213,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.002399705071002245,
      "weighted_relation_loss": 0.002399705071002245
    },
    {
      "action_query_batch_std": 0.3569083124399185,
      "action_query_norm": 41.03659248352051,
      "action_relation_loss": 0.0023474615067243577,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.523854069223574,
      "flow_matching_loss": 0.01926483428105712,
      "grad_norm": 0.12467844039201736,
      "h_stock_valid_norm": 98.04558715820312,
      "invalid_tail_samples": 23.0,
      "learning_rate": 1.847405186254807e-05,
      "loss": 0.0216,
      "positive_teacher_pairs": 27030.2,
      "query_h_norm_ratio": 0.4185495346784592,
      "relation_loss": 0.0023474615067243577,
      "selected_pairs": 27030.2,
      "step": 21940,
      "student_distance_mean": 23.506885528564453,
      "teacher_distance_mean": 0.8639645993709564,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0023474615067243577,
      "weighted_relation_loss": 0.0023474615067243577
    },
    {
      "action_query_batch_std": 0.3557713538408279,
      "action_query_norm": 41.03469390869141,
      "action_relation_loss": 0.0022675345186144113,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.533208606173993,
      "flow_matching_loss": 0.019660162553191186,
      "grad_norm": 0.10677041113376617,
      "h_stock_valid_norm": 97.90414733886719,
      "invalid_tail_samples": 20.8,
      "learning_rate": 1.843129169051379e-05,
      "loss": 0.0219,
      "positive_teacher_pairs": 27551.2,
      "query_h_norm_ratio": 0.4191324770450592,
      "relation_loss": 0.0022675345186144113,
      "selected_pairs": 27551.2,
      "step": 21950,
      "student_distance_mean": 23.325975036621095,
      "teacher_distance_mean": 0.8588447511196137,
      "valid_action_samples": 235.2,
      "weighted_auxiliary_loss": 0.0022675345186144113,
      "weighted_relation_loss": 0.0022675345186144113
    },
    {
      "action_query_batch_std": 0.35472410917282104,
      "action_query_norm": 41.01180076599121,
      "action_relation_loss": 0.002272390713915229,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.542563143124415,
      "flow_matching_loss": 0.01873991722241044,
      "grad_norm": 0.11390136182308197,
      "h_stock_valid_norm": 97.77975921630859,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.8388569877436863e-05,
      "loss": 0.021,
      "positive_teacher_pairs": 27107.6,
      "query_h_norm_ratio": 0.4194324642419815,
      "relation_loss": 0.002272390713915229,
      "selected_pairs": 27107.6,
      "step": 21960,
      "student_distance_mean": 23.32289237976074,
      "teacher_distance_mean": 0.8580171883106231,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.002272390713915229,
      "weighted_relation_loss": 0.002272390713915229
    },
    {
      "action_query_batch_std": 0.3550864666700363,
      "action_query_norm": 41.00415382385254,
      "action_relation_loss": 0.002235751482658088,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.551917680074837,
      "flow_matching_loss": 0.019466070272028447,
      "grad_norm": 0.08094828575849533,
      "h_stock_valid_norm": 98.0899543762207,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.834588647522828e-05,
      "loss": 0.0217,
      "positive_teacher_pairs": 26734.5,
      "query_h_norm_ratio": 0.41803018450737,
      "relation_loss": 0.002235751482658088,
      "selected_pairs": 26734.5,
      "step": 21970,
      "student_distance_mean": 23.389320182800294,
      "teacher_distance_mean": 0.8609029829502106,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002235751482658088,
      "weighted_relation_loss": 0.002235751482658088
    },
    {
      "action_query_batch_std": 0.35177905559539796,
      "action_query_norm": 41.00231437683105,
      "action_relation_loss": 0.002496922155842185,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.56127221702526,
      "flow_matching_loss": 0.019055388495326044,
      "grad_norm": 0.08695012331008911,
      "h_stock_valid_norm": 98.21733474731445,
      "invalid_tail_samples": 25.1,
      "learning_rate": 1.8303241535752435e-05,
      "loss": 0.0216,
      "positive_teacher_pairs": 26545.4,
      "query_h_norm_ratio": 0.4174668401479721,
      "relation_loss": 0.002496922155842185,
      "selected_pairs": 26545.4,
      "step": 21980,
      "student_distance_mean": 23.086149978637696,
      "teacher_distance_mean": 0.8488863050937653,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.002496922155842185,
      "weighted_relation_loss": 0.002496922155842185
    },
    {
      "action_query_batch_std": 0.3551753133535385,
      "action_query_norm": 41.02674827575684,
      "action_relation_loss": 0.0022545970627106727,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.570626753975677,
      "flow_matching_loss": 0.019142607506364585,
      "grad_norm": 0.08928057551383972,
      "h_stock_valid_norm": 98.36990737915039,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.8260635110826934e-05,
      "loss": 0.0214,
      "positive_teacher_pairs": 26805.9,
      "query_h_norm_ratio": 0.4170715183019638,
      "relation_loss": 0.0022545970627106727,
      "selected_pairs": 26805.9,
      "step": 21990,
      "student_distance_mean": 23.47476997375488,
      "teacher_distance_mean": 0.8634514749050141,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0022545970627106727,
      "weighted_relation_loss": 0.0022545970627106727
    },
    {
      "action_query_batch_std": 0.35396202504634855,
      "action_query_norm": 41.00783157348633,
      "action_relation_loss": 0.0022688355995342134,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.5799812909261,
      "flow_matching_loss": 0.019190611969679593,
      "grad_norm": 0.11742313951253891,
      "h_stock_valid_norm": 98.59990310668945,
      "invalid_tail_samples": 22.9,
      "learning_rate": 1.8218067252222597e-05,
      "loss": 0.0215,
      "positive_teacher_pairs": 27059.3,
      "query_h_norm_ratio": 0.4159031867980957,
      "relation_loss": 0.0022688355995342134,
      "selected_pairs": 27059.3,
      "step": 22000,
      "student_distance_mean": 23.402433776855467,
      "teacher_distance_mean": 0.8597420692443848,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0022688355995342134,
      "weighted_relation_loss": 0.0022688355995342134
    },
    {
      "action_query_batch_std": 0.35301052033901215,
      "action_query_norm": 40.983435821533206,
      "action_relation_loss": 0.0021784478100016712,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.58933582787652,
      "flow_matching_loss": 0.018709355033934116,
      "grad_norm": 0.13093076646327972,
      "h_stock_valid_norm": 98.81623153686523,
      "invalid_tail_samples": 23.8,
      "learning_rate": 1.817553801166339e-05,
      "loss": 0.0209,
      "positive_teacher_pairs": 26858.9,
      "query_h_norm_ratio": 0.4147489100694656,
      "relation_loss": 0.0021784478100016712,
      "selected_pairs": 26858.9,
      "step": 22010,
      "student_distance_mean": 23.264721488952638,
      "teacher_distance_mean": 0.8536525905132294,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0021784478100016712,
      "weighted_relation_loss": 0.0021784478100016712
    },
    {
      "action_query_batch_std": 0.35222579538822174,
      "action_query_norm": 40.982536697387694,
      "action_relation_loss": 0.002298186137340963,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.59869036482694,
      "flow_matching_loss": 0.019627390056848527,
      "grad_norm": 0.1532394289970398,
      "h_stock_valid_norm": 98.66714096069336,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.8133047440826333e-05,
      "loss": 0.0219,
      "positive_teacher_pairs": 26898.8,
      "query_h_norm_ratio": 0.4153664082288742,
      "relation_loss": 0.002298186137340963,
      "selected_pairs": 26898.8,
      "step": 22020,
      "student_distance_mean": 23.12513427734375,
      "teacher_distance_mean": 0.8521599471569061,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.002298186137340963,
      "weighted_relation_loss": 0.002298186137340963
    },
    {
      "action_query_batch_std": 0.35115292072296145,
      "action_query_norm": 40.99417572021484,
      "action_relation_loss": 0.0022596195340156555,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.60804490177736,
      "flow_matching_loss": 0.019559785816818477,
      "grad_norm": 0.11176110804080963,
      "h_stock_valid_norm": 98.23292465209961,
      "invalid_tail_samples": 22.1,
      "learning_rate": 1.809059559134151e-05,
      "loss": 0.0218,
      "positive_teacher_pairs": 27251.2,
      "query_h_norm_ratio": 0.41731824576854704,
      "relation_loss": 0.0022596195340156555,
      "selected_pairs": 27251.2,
      "step": 22030,
      "student_distance_mean": 23.136780738830566,
      "teacher_distance_mean": 0.8534752070903778,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0022596195340156555,
      "weighted_relation_loss": 0.0022596195340156555
    },
    {
      "action_query_batch_std": 0.35124051570892334,
      "action_query_norm": 40.980445861816406,
      "action_relation_loss": 0.0023242352530360223,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.617399438727784,
      "flow_matching_loss": 0.02056262604892254,
      "grad_norm": 0.0869961827993393,
      "h_stock_valid_norm": 98.2149787902832,
      "invalid_tail_samples": 24.6,
      "learning_rate": 1.80481825147919e-05,
      "loss": 0.0229,
      "positive_teacher_pairs": 26663.5,
      "query_h_norm_ratio": 0.41725781559944153,
      "relation_loss": 0.0023242352530360223,
      "selected_pairs": 26663.5,
      "step": 22040,
      "student_distance_mean": 23.057643127441406,
      "teacher_distance_mean": 0.8512354791164398,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0023242352530360223,
      "weighted_relation_loss": 0.0023242352530360223
    },
    {
      "action_query_batch_std": 0.3542291700839996,
      "action_query_norm": 41.00839424133301,
      "action_relation_loss": 0.002321322006173432,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.626753975678206,
      "flow_matching_loss": 0.020648880489170553,
      "grad_norm": 0.12005643546581268,
      "h_stock_valid_norm": 98.18767013549805,
      "invalid_tail_samples": 25.6,
      "learning_rate": 1.8005808262713396e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 26439.1,
      "query_h_norm_ratio": 0.41765460968017576,
      "relation_loss": 0.002321322006173432,
      "selected_pairs": 26439.1,
      "step": 22050,
      "student_distance_mean": 23.272370719909667,
      "teacher_distance_mean": 0.8552681088447571,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.002321322006173432,
      "weighted_relation_loss": 0.002321322006173432
    },
    {
      "action_query_batch_std": 0.34936304688453673,
      "action_query_norm": 40.96491012573242,
      "action_relation_loss": 0.0024173761252313853,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.636108512628624,
      "flow_matching_loss": 0.019689891301095486,
      "grad_norm": 0.09139902144670486,
      "h_stock_valid_norm": 98.27416152954102,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.7963472886594714e-05,
      "loss": 0.0221,
      "positive_teacher_pairs": 26734.3,
      "query_h_norm_ratio": 0.416844043135643,
      "relation_loss": 0.0024173761252313853,
      "selected_pairs": 26734.3,
      "step": 22060,
      "student_distance_mean": 23.07574329376221,
      "teacher_distance_mean": 0.8498256921768188,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0024173761252313853,
      "weighted_relation_loss": 0.0024173761252313853
    },
    {
      "action_query_batch_std": 0.3479832410812378,
      "action_query_norm": 40.95675964355469,
      "action_relation_loss": 0.0024060097988694907,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.645463049579046,
      "flow_matching_loss": 0.020157302729785442,
      "grad_norm": 0.08708252757787704,
      "h_stock_valid_norm": 98.6760383605957,
      "invalid_tail_samples": 22.5,
      "learning_rate": 1.7921176437877302e-05,
      "loss": 0.0226,
      "positive_teacher_pairs": 27153.4,
      "query_h_norm_ratio": 0.4150641769170761,
      "relation_loss": 0.0024060097988694907,
      "selected_pairs": 27153.4,
      "step": 22070,
      "student_distance_mean": 22.921710205078124,
      "teacher_distance_mean": 0.8484510779380798,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0024060097988694907,
      "weighted_relation_loss": 0.0024060097988694907
    },
    {
      "action_query_batch_std": 0.3505344927310944,
      "action_query_norm": 40.975229263305664,
      "action_relation_loss": 0.00229058510158211,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.654817586529468,
      "flow_matching_loss": 0.01968748737126589,
      "grad_norm": 0.08092714846134186,
      "h_stock_valid_norm": 98.93022155761719,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.7878918967955366e-05,
      "loss": 0.022,
      "positive_teacher_pairs": 27012.4,
      "query_h_norm_ratio": 0.4141874432563782,
      "relation_loss": 0.00229058510158211,
      "selected_pairs": 27012.4,
      "step": 22080,
      "student_distance_mean": 23.01228256225586,
      "teacher_distance_mean": 0.8507722377777099,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.00229058510158211,
      "weighted_relation_loss": 0.00229058510158211
    },
    {
      "action_query_batch_std": 0.35197688937187194,
      "action_query_norm": 40.9753231048584,
      "action_relation_loss": 0.0022899986477568747,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.664172123479887,
      "flow_matching_loss": 0.0189046080224216,
      "grad_norm": 0.09612464159727097,
      "h_stock_valid_norm": 98.88132400512696,
      "invalid_tail_samples": 21.9,
      "learning_rate": 1.7836700528175693e-05,
      "loss": 0.0212,
      "positive_teacher_pairs": 27285.6,
      "query_h_norm_ratio": 0.4143903285264969,
      "relation_loss": 0.0022899986477568747,
      "selected_pairs": 27285.6,
      "step": 22090,
      "student_distance_mean": 23.143687438964843,
      "teacher_distance_mean": 0.853682941198349,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.0022899986477568747,
      "weighted_relation_loss": 0.0022899986477568747
    },
    {
      "action_query_batch_std": 0.35237019658088686,
      "action_query_norm": 40.981224822998044,
      "action_relation_loss": 0.0022858506534248592,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.67352666043031,
      "flow_matching_loss": 0.01938177701085806,
      "grad_norm": 0.08035140484571457,
      "h_stock_valid_norm": 98.4118263244629,
      "invalid_tail_samples": 26.8,
      "learning_rate": 1.779452116983769e-05,
      "loss": 0.0217,
      "positive_teacher_pairs": 26159.4,
      "query_h_norm_ratio": 0.41643557250499724,
      "relation_loss": 0.0022858506534248592,
      "selected_pairs": 26159.4,
      "step": 22100,
      "student_distance_mean": 23.212581443786622,
      "teacher_distance_mean": 0.8557725846767426,
      "valid_action_samples": 229.2,
      "weighted_auxiliary_loss": 0.0022858506534248592,
      "weighted_relation_loss": 0.0022858506534248592
    },
    {
      "action_query_batch_std": 0.3520238041877747,
      "action_query_norm": 40.97197494506836,
      "action_relation_loss": 0.002430608728900552,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.68288119738073,
      "flow_matching_loss": 0.019675070140510797,
      "grad_norm": 0.1572459638118744,
      "h_stock_valid_norm": 97.66564102172852,
      "invalid_tail_samples": 20.7,
      "learning_rate": 1.775238094419325e-05,
      "loss": 0.0221,
      "positive_teacher_pairs": 27571.0,
      "query_h_norm_ratio": 0.41951523125171664,
      "relation_loss": 0.002430608728900552,
      "selected_pairs": 27571.0,
      "step": 22110,
      "student_distance_mean": 23.109002113342285,
      "teacher_distance_mean": 0.8528156995773315,
      "valid_action_samples": 235.3,
      "weighted_auxiliary_loss": 0.002430608728900552,
      "weighted_relation_loss": 0.002430608728900552
    },
    {
      "action_query_batch_std": 0.3539894908666611,
      "action_query_norm": 41.00118560791016,
      "action_relation_loss": 0.0023286401527002456,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.69223573433115,
      "flow_matching_loss": 0.0192240659147501,
      "grad_norm": 0.10230943560600281,
      "h_stock_valid_norm": 97.09116592407227,
      "invalid_tail_samples": 23.4,
      "learning_rate": 1.7710279902446708e-05,
      "loss": 0.0216,
      "positive_teacher_pairs": 26951.5,
      "query_h_norm_ratio": 0.4223012775182724,
      "relation_loss": 0.0023286401527002456,
      "selected_pairs": 26951.5,
      "step": 22120,
      "student_distance_mean": 23.18202133178711,
      "teacher_distance_mean": 0.8533107161521911,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0023286401527002456,
      "weighted_relation_loss": 0.0023286401527002456
    },
    {
      "action_query_batch_std": 0.35128759443759916,
      "action_query_norm": 40.95638084411621,
      "action_relation_loss": 0.0023582896683365107,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.70159027128157,
      "flow_matching_loss": 0.020104354806244375,
      "grad_norm": 0.11776643246412277,
      "h_stock_valid_norm": 97.25535736083984,
      "invalid_tail_samples": 25.5,
      "learning_rate": 1.7668218095754795e-05,
      "loss": 0.0225,
      "positive_teacher_pairs": 26456.8,
      "query_h_norm_ratio": 0.4211234778165817,
      "relation_loss": 0.0023582896683365107,
      "selected_pairs": 26456.8,
      "step": 22130,
      "student_distance_mean": 23.071606826782226,
      "teacher_distance_mean": 0.8481730103492737,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0023582896683365107,
      "weighted_relation_loss": 0.0023582896683365107
    },
    {
      "action_query_batch_std": 0.3545372813940048,
      "action_query_norm": 40.964769744873045,
      "action_relation_loss": 0.0022769691655412315,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.710944808231993,
      "flow_matching_loss": 0.019368675258010625,
      "grad_norm": 0.13187062740325928,
      "h_stock_valid_norm": 97.44668884277344,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.7626195575226594e-05,
      "loss": 0.0216,
      "positive_teacher_pairs": 26898.1,
      "query_h_norm_ratio": 0.4203821510076523,
      "relation_loss": 0.0022769691655412315,
      "selected_pairs": 26898.1,
      "step": 22140,
      "student_distance_mean": 23.274655151367188,
      "teacher_distance_mean": 0.8585481703281402,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0022769691655412315,
      "weighted_relation_loss": 0.0022769691655412315
    },
    {
      "action_query_batch_std": 0.35072227716445925,
      "action_query_norm": 40.955592727661134,
      "action_relation_loss": 0.0025105864275246858,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.720299345182415,
      "flow_matching_loss": 0.020204053074121476,
      "grad_norm": 0.09722384810447693,
      "h_stock_valid_norm": 97.8525489807129,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.758421239192343e-05,
      "loss": 0.0227,
      "positive_teacher_pairs": 26824.9,
      "query_h_norm_ratio": 0.41854638457298277,
      "relation_loss": 0.0025105864275246858,
      "selected_pairs": 26824.9,
      "step": 22150,
      "student_distance_mean": 23.01792221069336,
      "teacher_distance_mean": 0.8505948185920715,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0025105864275246858,
      "weighted_relation_loss": 0.0025105864275246858
    },
    {
      "action_query_batch_std": 0.35305900275707247,
      "action_query_norm": 40.97858238220215,
      "action_relation_loss": 0.0022266840562224386,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.729653882132833,
      "flow_matching_loss": 0.020871713943779468,
      "grad_norm": 0.13183841109275818,
      "h_stock_valid_norm": 97.71117248535157,
      "invalid_tail_samples": 22.4,
      "learning_rate": 1.7542268596858814e-05,
      "loss": 0.0231,
      "positive_teacher_pairs": 27184.2,
      "query_h_norm_ratio": 0.4193859934806824,
      "relation_loss": 0.0022266840562224386,
      "selected_pairs": 27184.2,
      "step": 22160,
      "student_distance_mean": 23.237982940673827,
      "teacher_distance_mean": 0.8574661016464233,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.0022266840562224386,
      "weighted_relation_loss": 0.0022266840562224386
    },
    {
      "action_query_batch_std": 0.35392343401908877,
      "action_query_norm": 41.001324844360354,
      "action_relation_loss": 0.002318636700510979,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.739008419083255,
      "flow_matching_loss": 0.01979295173659921,
      "grad_norm": 0.09189622849225998,
      "h_stock_valid_norm": 97.49767761230468,
      "invalid_tail_samples": 23.5,
      "learning_rate": 1.750036424099841e-05,
      "loss": 0.0221,
      "positive_teacher_pairs": 26924.0,
      "query_h_norm_ratio": 0.42053923904895785,
      "relation_loss": 0.002318636700510979,
      "selected_pairs": 26924.0,
      "step": 22170,
      "student_distance_mean": 23.283037376403808,
      "teacher_distance_mean": 0.85691978931427,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.002318636700510979,
      "weighted_relation_loss": 0.002318636700510979
    },
    {
      "action_query_batch_std": 0.3521749168634415,
      "action_query_norm": 40.95079727172852,
      "action_relation_loss": 0.0022597919683903455,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.748362956033677,
      "flow_matching_loss": 0.01745082763954997,
      "grad_norm": 0.11953090131282806,
      "h_stock_valid_norm": 97.66660995483399,
      "invalid_tail_samples": 25.7,
      "learning_rate": 1.7458499375259956e-05,
      "loss": 0.0197,
      "positive_teacher_pairs": 26410.8,
      "query_h_norm_ratio": 0.4192959815263748,
      "relation_loss": 0.0022597919683903455,
      "selected_pairs": 26410.8,
      "step": 22180,
      "student_distance_mean": 23.231294059753417,
      "teacher_distance_mean": 0.8581580936908721,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0022597919683903455,
      "weighted_relation_loss": 0.0022597919683903455
    },
    {
      "action_query_batch_std": 0.3561197191476822,
      "action_query_norm": 40.99186973571777,
      "action_relation_loss": 0.002263228641822934,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.757717492984096,
      "flow_matching_loss": 0.01946650743484497,
      "grad_norm": 0.07565869390964508,
      "h_stock_valid_norm": 97.9091064453125,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.7416674050513245e-05,
      "loss": 0.0217,
      "positive_teacher_pairs": 26833.5,
      "query_h_norm_ratio": 0.4186743527650833,
      "relation_loss": 0.002263228641822934,
      "selected_pairs": 26833.5,
      "step": 22190,
      "student_distance_mean": 23.40074634552002,
      "teacher_distance_mean": 0.8637134492397308,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.002263228641822934,
      "weighted_relation_loss": 0.002263228641822934
    },
    {
      "action_query_batch_std": 0.34815865755081177,
      "action_query_norm": 40.941851425170896,
      "action_relation_loss": 0.002334009180776775,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.767072029934518,
      "flow_matching_loss": 0.021068440563976765,
      "grad_norm": 0.13262373208999634,
      "h_stock_valid_norm": 97.69076919555664,
      "invalid_tail_samples": 21.1,
      "learning_rate": 1.737488831757997e-05,
      "loss": 0.0234,
      "positive_teacher_pairs": 27476.9,
      "query_h_norm_ratio": 0.4190985202789307,
      "relation_loss": 0.002334009180776775,
      "selected_pairs": 27476.9,
      "step": 22200,
      "student_distance_mean": 22.87056484222412,
      "teacher_distance_mean": 0.8424690186977386,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.002334009180776775,
      "weighted_relation_loss": 0.002334009180776775
    },
    {
      "action_query_batch_std": 0.3512087732553482,
      "action_query_norm": 40.966880798339844,
      "action_relation_loss": 0.0021920791594311597,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.77642656688494,
      "flow_matching_loss": 0.01916970582678914,
      "grad_norm": 0.0905023142695427,
      "h_stock_valid_norm": 97.6724021911621,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.733314222723373e-05,
      "loss": 0.0214,
      "positive_teacher_pairs": 27080.9,
      "query_h_norm_ratio": 0.41943292021751405,
      "relation_loss": 0.0021920791594311597,
      "selected_pairs": 27080.9,
      "step": 22210,
      "student_distance_mean": 23.180217933654784,
      "teacher_distance_mean": 0.8545158565044403,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0021920791594311597,
      "weighted_relation_loss": 0.0021920791594311597
    },
    {
      "action_query_batch_std": 0.3514805227518082,
      "action_query_norm": 40.96918640136719,
      "action_relation_loss": 0.0022902354830875993,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.78578110383536,
      "flow_matching_loss": 0.018679173663258553,
      "grad_norm": 0.0905953124165535,
      "h_stock_valid_norm": 97.29475402832031,
      "invalid_tail_samples": 23.2,
      "learning_rate": 1.7291435830199954e-05,
      "loss": 0.021,
      "positive_teacher_pairs": 26988.9,
      "query_h_norm_ratio": 0.42108452022075654,
      "relation_loss": 0.0022902354830875993,
      "selected_pairs": 26988.9,
      "step": 22220,
      "student_distance_mean": 23.039737129211424,
      "teacher_distance_mean": 0.8488244891166687,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0022902354830875993,
      "weighted_relation_loss": 0.0022902354830875993
    },
    {
      "action_query_batch_std": 0.35884750485420225,
      "action_query_norm": 41.02574996948242,
      "action_relation_loss": 0.002285069297067821,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.79513564078578,
      "flow_matching_loss": 0.021087325178086758,
      "grad_norm": 0.10636463761329651,
      "h_stock_valid_norm": 97.1374641418457,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.7249769177155878e-05,
      "loss": 0.0234,
      "positive_teacher_pairs": 27081.1,
      "query_h_norm_ratio": 0.42234845757484435,
      "relation_loss": 0.002285069297067821,
      "selected_pairs": 27081.1,
      "step": 22230,
      "student_distance_mean": 23.64753475189209,
      "teacher_distance_mean": 0.8738548815250397,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.002285069297067821,
      "weighted_relation_loss": 0.002285069297067821
    },
    {
      "action_query_batch_std": 0.35645546317100524,
      "action_query_norm": 41.007730102539064,
      "action_relation_loss": 0.002232576999813318,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.804490177736202,
      "flow_matching_loss": 0.02082666363567114,
      "grad_norm": 0.09702441096305847,
      "h_stock_valid_norm": 97.62184371948243,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.720814231873038e-05,
      "loss": 0.0231,
      "positive_teacher_pairs": 26829.7,
      "query_h_norm_ratio": 0.4200681060552597,
      "relation_loss": 0.002232576999813318,
      "selected_pairs": 26829.7,
      "step": 22240,
      "student_distance_mean": 23.515988540649413,
      "teacher_distance_mean": 0.8653818726539612,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.002232576999813318,
      "weighted_relation_loss": 0.002232576999813318
    },
    {
      "action_query_batch_std": 0.35439781546592714,
      "action_query_norm": 40.99397773742676,
      "action_relation_loss": 0.0023359128972515464,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.813844714686624,
      "flow_matching_loss": 0.020452614687383176,
      "grad_norm": 0.10383297502994537,
      "h_stock_valid_norm": 97.73386840820312,
      "invalid_tail_samples": 21.8,
      "learning_rate": 1.716655530550405e-05,
      "loss": 0.0228,
      "positive_teacher_pairs": 27316.2,
      "query_h_norm_ratio": 0.4194455534219742,
      "relation_loss": 0.0023359128972515464,
      "selected_pairs": 27316.2,
      "step": 22250,
      "student_distance_mean": 23.41396255493164,
      "teacher_distance_mean": 0.8586272478103638,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.0023359128972515464,
      "weighted_relation_loss": 0.0023359128972515464
    },
    {
      "action_query_batch_std": 0.3557309120893478,
      "action_query_norm": 40.99952735900879,
      "action_relation_loss": 0.0023785537341609597,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.823199251637043,
      "flow_matching_loss": 0.02056278046220541,
      "grad_norm": 0.10852204263210297,
      "h_stock_valid_norm": 98.13619842529297,
      "invalid_tail_samples": 21.7,
      "learning_rate": 1.7125008188009017e-05,
      "loss": 0.0229,
      "positive_teacher_pairs": 27339.7,
      "query_h_norm_ratio": 0.4177856266498566,
      "relation_loss": 0.0023785537341609597,
      "selected_pairs": 27339.7,
      "step": 22260,
      "student_distance_mean": 23.420362091064455,
      "teacher_distance_mean": 0.8615343570709229,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0023785537341609597,
      "weighted_relation_loss": 0.0023785537341609597
    },
    {
      "action_query_batch_std": 0.35056706964969636,
      "action_query_norm": 40.961144256591794,
      "action_relation_loss": 0.0022394439205527306,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.832553788587465,
      "flow_matching_loss": 0.019292376935482025,
      "grad_norm": 0.0936623215675354,
      "h_stock_valid_norm": 98.01719589233399,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.708350101672894e-05,
      "loss": 0.0215,
      "positive_teacher_pairs": 26755.1,
      "query_h_norm_ratio": 0.417900425195694,
      "relation_loss": 0.0022394439205527306,
      "selected_pairs": 26755.1,
      "step": 22270,
      "student_distance_mean": 23.141460609436034,
      "teacher_distance_mean": 0.8533702909946441,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0022394439205527306,
      "weighted_relation_loss": 0.0022394439205527306
    },
    {
      "action_query_batch_std": 0.35202584266662595,
      "action_query_norm": 40.97019882202149,
      "action_relation_loss": 0.0021928278962150215,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.841908325537887,
      "flow_matching_loss": 0.02027285974472761,
      "grad_norm": 0.08733360469341278,
      "h_stock_valid_norm": 97.92593307495117,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.7042033842098957e-05,
      "loss": 0.0225,
      "positive_teacher_pairs": 26828.5,
      "query_h_norm_ratio": 0.4183821529150009,
      "relation_loss": 0.0021928278962150215,
      "selected_pairs": 26828.5,
      "step": 22280,
      "student_distance_mean": 23.159545135498046,
      "teacher_distance_mean": 0.8527722597122193,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0021928278962150215,
      "weighted_relation_loss": 0.0021928278962150215
    },
    {
      "action_query_batch_std": 0.35032196044921876,
      "action_query_norm": 40.95303916931152,
      "action_relation_loss": 0.0022797319805249574,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.85126286248831,
      "flow_matching_loss": 0.02163433376699686,
      "grad_norm": 0.10254623740911484,
      "h_stock_valid_norm": 97.87989807128906,
      "invalid_tail_samples": 24.8,
      "learning_rate": 1.7000606714505567e-05,
      "loss": 0.0239,
      "positive_teacher_pairs": 26630.9,
      "query_h_norm_ratio": 0.4184048682451248,
      "relation_loss": 0.0022797319805249574,
      "selected_pairs": 26630.9,
      "step": 22290,
      "student_distance_mean": 23.087105369567873,
      "teacher_distance_mean": 0.8525071322917939,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0022797319805249574,
      "weighted_relation_loss": 0.0022797319805249574
    },
    {
      "action_query_batch_std": 0.35454944968223573,
      "action_query_norm": 40.97753677368164,
      "action_relation_loss": 0.0024141873931512237,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.860617399438727,
      "flow_matching_loss": 0.021173722669482233,
      "grad_norm": 0.09345310926437378,
      "h_stock_valid_norm": 98.19568634033203,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.695921968428668e-05,
      "loss": 0.0236,
      "positive_teacher_pairs": 27014.8,
      "query_h_norm_ratio": 0.41730855107307435,
      "relation_loss": 0.0024141873931512237,
      "selected_pairs": 27014.8,
      "step": 22300,
      "student_distance_mean": 23.31320972442627,
      "teacher_distance_mean": 0.858293479681015,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0024141873931512237,
      "weighted_relation_loss": 0.0024141873931512237
    },
    {
      "action_query_batch_std": 0.3555565059185028,
      "action_query_norm": 40.99645080566406,
      "action_relation_loss": 0.0023094174335710703,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.86997193638915,
      "flow_matching_loss": 0.01819348894059658,
      "grad_norm": 0.10672001540660858,
      "h_stock_valid_norm": 98.08112716674805,
      "invalid_tail_samples": 25.7,
      "learning_rate": 1.6917872801731416e-05,
      "loss": 0.0205,
      "positive_teacher_pairs": 26419.6,
      "query_h_norm_ratio": 0.41798787415027616,
      "relation_loss": 0.0023094174335710703,
      "selected_pairs": 26419.6,
      "step": 22310,
      "student_distance_mean": 23.331517219543457,
      "teacher_distance_mean": 0.8599628210067749,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0023094174335710703,
      "weighted_relation_loss": 0.0023094174335710703
    },
    {
      "action_query_batch_std": 0.348847559094429,
      "action_query_norm": 40.93336791992188,
      "action_relation_loss": 0.002300865715369582,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.87932647333957,
      "flow_matching_loss": 0.01945893606171012,
      "grad_norm": 0.08025160431861877,
      "h_stock_valid_norm": 97.97822494506836,
      "invalid_tail_samples": 25.3,
      "learning_rate": 1.687656611708014e-05,
      "loss": 0.0218,
      "positive_teacher_pairs": 26506.9,
      "query_h_norm_ratio": 0.41778424978256223,
      "relation_loss": 0.002300865715369582,
      "selected_pairs": 26506.9,
      "step": 22320,
      "student_distance_mean": 22.983300018310548,
      "teacher_distance_mean": 0.848354059457779,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.002300865715369582,
      "weighted_relation_loss": 0.002300865715369582
    },
    {
      "action_query_batch_std": 0.35123060643672943,
      "action_query_norm": 40.95010414123535,
      "action_relation_loss": 0.0022245598724111916,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.88868101028999,
      "flow_matching_loss": 0.02028316156938672,
      "grad_norm": 0.10445397347211838,
      "h_stock_valid_norm": 98.14135971069337,
      "invalid_tail_samples": 24.6,
      "learning_rate": 1.6835299680524368e-05,
      "loss": 0.0225,
      "positive_teacher_pairs": 26667.1,
      "query_h_norm_ratio": 0.41725927889347075,
      "relation_loss": 0.0022245598724111916,
      "selected_pairs": 26667.1,
      "step": 22330,
      "student_distance_mean": 23.14187698364258,
      "teacher_distance_mean": 0.8556691944599152,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0022245598724111916,
      "weighted_relation_loss": 0.0022245598724111916
    },
    {
      "action_query_batch_std": 0.3501691222190857,
      "action_query_norm": 40.94533271789551,
      "action_relation_loss": 0.002272215159609914,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.89803554724041,
      "flow_matching_loss": 0.018493746314197777,
      "grad_norm": 0.10506627708673477,
      "h_stock_valid_norm": 98.1951515197754,
      "invalid_tail_samples": 24.9,
      "learning_rate": 1.679407354220669e-05,
      "loss": 0.0208,
      "positive_teacher_pairs": 26601.8,
      "query_h_norm_ratio": 0.41698208451271057,
      "relation_loss": 0.002272215159609914,
      "selected_pairs": 26601.8,
      "step": 22340,
      "student_distance_mean": 23.03660430908203,
      "teacher_distance_mean": 0.8514031410217285,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.002272215159609914,
      "weighted_relation_loss": 0.002272215159609914
    },
    {
      "action_query_batch_std": 0.35293697416782377,
      "action_query_norm": 40.98249626159668,
      "action_relation_loss": 0.0022539522498846052,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.907390084190833,
      "flow_matching_loss": 0.017676498368382455,
      "grad_norm": 0.11603834480047226,
      "h_stock_valid_norm": 98.1182991027832,
      "invalid_tail_samples": 25.2,
      "learning_rate": 1.6752887752220793e-05,
      "loss": 0.0199,
      "positive_teacher_pairs": 26535.4,
      "query_h_norm_ratio": 0.4176895350217819,
      "relation_loss": 0.0022539522498846052,
      "selected_pairs": 26535.4,
      "step": 22350,
      "student_distance_mean": 23.261747550964355,
      "teacher_distance_mean": 0.8595798730850219,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0022539522498846052,
      "weighted_relation_loss": 0.0022539522498846052
    },
    {
      "action_query_batch_std": 0.349379700422287,
      "action_query_norm": 40.93867111206055,
      "action_relation_loss": 0.002320609032176435,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.916744621141255,
      "flow_matching_loss": 0.017561376560479403,
      "grad_norm": 0.13117852807044983,
      "h_stock_valid_norm": 98.13078308105469,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.6711742360611277e-05,
      "loss": 0.0199,
      "positive_teacher_pairs": 26734.7,
      "query_h_norm_ratio": 0.4171887397766113,
      "relation_loss": 0.002320609032176435,
      "selected_pairs": 26734.7,
      "step": 22360,
      "student_distance_mean": 23.017044067382812,
      "teacher_distance_mean": 0.8479538261890411,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002320609032176435,
      "weighted_relation_loss": 0.002320609032176435
    },
    {
      "action_query_batch_std": 0.35321756899356843,
      "action_query_norm": 40.956913757324216,
      "action_relation_loss": 0.00236212695017457,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.926099158091674,
      "flow_matching_loss": 0.0191637403331697,
      "grad_norm": 0.08029670268297195,
      "h_stock_valid_norm": 97.96337051391602,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.6670637417373652e-05,
      "loss": 0.0215,
      "positive_teacher_pairs": 26741.2,
      "query_h_norm_ratio": 0.41809155642986295,
      "relation_loss": 0.00236212695017457,
      "selected_pairs": 26741.2,
      "step": 22370,
      "student_distance_mean": 23.211198616027833,
      "teacher_distance_mean": 0.8550454199314117,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.00236212695017457,
      "weighted_relation_loss": 0.00236212695017457
    },
    {
      "action_query_batch_std": 0.3529053807258606,
      "action_query_norm": 40.954559326171875,
      "action_relation_loss": 0.002260146103799343,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.935453695042096,
      "flow_matching_loss": 0.020575348660349844,
      "grad_norm": 0.12312287092208862,
      "h_stock_valid_norm": 97.99020080566406,
      "invalid_tail_samples": 20.3,
      "learning_rate": 1.6629572972454333e-05,
      "loss": 0.0228,
      "positive_teacher_pairs": 27665.1,
      "query_h_norm_ratio": 0.4179489254951477,
      "relation_loss": 0.002260146103799343,
      "selected_pairs": 27665.1,
      "step": 22380,
      "student_distance_mean": 23.16234760284424,
      "teacher_distance_mean": 0.8544592797756195,
      "valid_action_samples": 235.7,
      "weighted_auxiliary_loss": 0.002260146103799343,
      "weighted_relation_loss": 0.002260146103799343
    },
    {
      "action_query_batch_std": 0.3473969131708145,
      "action_query_norm": 40.90746726989746,
      "action_relation_loss": 0.0022533552721142767,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.944808231992518,
      "flow_matching_loss": 0.01787747424095869,
      "grad_norm": 0.11216370761394501,
      "h_stock_valid_norm": 98.11271743774414,
      "invalid_tail_samples": 25.5,
      "learning_rate": 1.6588549075750464e-05,
      "loss": 0.0201,
      "positive_teacher_pairs": 26462.8,
      "query_h_norm_ratio": 0.41694530844688416,
      "relation_loss": 0.0022533552721142767,
      "selected_pairs": 26462.8,
      "step": 22390,
      "student_distance_mean": 22.88265914916992,
      "teacher_distance_mean": 0.8458777129650116,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0022533552721142767,
      "weighted_relation_loss": 0.0022533552721142767
    },
    {
      "action_query_batch_std": 0.35026181042194365,
      "action_query_norm": 40.92576370239258,
      "action_relation_loss": 0.0022530496353283525,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.954162768942936,
      "flow_matching_loss": 0.01860562600195408,
      "grad_norm": 0.09687870740890503,
      "h_stock_valid_norm": 98.10309600830078,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.6547565777109976e-05,
      "loss": 0.0209,
      "positive_teacher_pairs": 26876.1,
      "query_h_norm_ratio": 0.4171731799840927,
      "relation_loss": 0.0022530496353283525,
      "selected_pairs": 26876.1,
      "step": 22400,
      "student_distance_mean": 23.079208183288575,
      "teacher_distance_mean": 0.8541008412837983,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0022530496353283525,
      "weighted_relation_loss": 0.0022530496353283525
    },
    {
      "action_query_batch_std": 0.3512772560119629,
      "action_query_norm": 40.942766952514646,
      "action_relation_loss": 0.002432092116214335,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.963517305893358,
      "flow_matching_loss": 0.018826870061457158,
      "grad_norm": 0.09080616384744644,
      "h_stock_valid_norm": 98.32205505371094,
      "invalid_tail_samples": 24.9,
      "learning_rate": 1.650662312633143e-05,
      "loss": 0.0213,
      "positive_teacher_pairs": 26590.7,
      "query_h_norm_ratio": 0.4164161145687103,
      "relation_loss": 0.002432092116214335,
      "selected_pairs": 26590.7,
      "step": 22410,
      "student_distance_mean": 23.020368385314942,
      "teacher_distance_mean": 0.8528976678848267,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.002432092116214335,
      "weighted_relation_loss": 0.002432092116214335
    },
    {
      "action_query_batch_std": 0.3527579367160797,
      "action_query_norm": 40.95445709228515,
      "action_relation_loss": 0.002303655620198697,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.97287184284378,
      "flow_matching_loss": 0.01906510116532445,
      "grad_norm": 0.08756162226200104,
      "h_stock_valid_norm": 98.28914337158203,
      "invalid_tail_samples": 26.7,
      "learning_rate": 1.6465721173164002e-05,
      "loss": 0.0214,
      "positive_teacher_pairs": 26179.4,
      "query_h_norm_ratio": 0.416674530506134,
      "relation_loss": 0.002303655620198697,
      "selected_pairs": 26179.4,
      "step": 22420,
      "student_distance_mean": 23.20816135406494,
      "teacher_distance_mean": 0.8565963208675385,
      "valid_action_samples": 229.3,
      "weighted_auxiliary_loss": 0.002303655620198697,
      "weighted_relation_loss": 0.002303655620198697
    },
    {
      "action_query_batch_std": 0.35016959607601167,
      "action_query_norm": 40.91701049804688,
      "action_relation_loss": 0.0024361380375921724,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.9822263797942,
      "flow_matching_loss": 0.020290889870375396,
      "grad_norm": 0.14950154721736908,
      "h_stock_valid_norm": 98.25290603637696,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.6424859967307428e-05,
      "loss": 0.0227,
      "positive_teacher_pairs": 26741.2,
      "query_h_norm_ratio": 0.41644900739192964,
      "relation_loss": 0.0024361380375921724,
      "selected_pairs": 26741.2,
      "step": 22430,
      "student_distance_mean": 23.07914218902588,
      "teacher_distance_mean": 0.8517954170703887,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0024361380375921724,
      "weighted_relation_loss": 0.0024361380375921724
    },
    {
      "action_query_batch_std": 0.3524070084095001,
      "action_query_norm": 40.9715274810791,
      "action_relation_loss": 0.00226714804302901,
      "auxiliary_loss_weight": 1.0,
      "epoch": 20.99158091674462,
      "flow_matching_loss": 0.01830556197091937,
      "grad_norm": 0.13892601430416107,
      "h_stock_valid_norm": 98.5342010498047,
      "invalid_tail_samples": 22.4,
      "learning_rate": 1.6384039558411903e-05,
      "loss": 0.0206,
      "positive_teacher_pairs": 27182.4,
      "query_h_norm_ratio": 0.4158117711544037,
      "relation_loss": 0.00226714804302901,
      "selected_pairs": 27182.4,
      "step": 22440,
      "student_distance_mean": 23.23599147796631,
      "teacher_distance_mean": 0.8572915852069855,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.00226714804302901,
      "weighted_relation_loss": 0.00226714804302901
    },
    {
      "action_query_batch_std": 0.3483229994773865,
      "action_query_norm": 40.92371597290039,
      "action_relation_loss": 0.002235451294109225,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.000935453695043,
      "flow_matching_loss": 0.01873501380905509,
      "grad_norm": 0.10522134602069855,
      "h_stock_valid_norm": 98.31050567626953,
      "invalid_tail_samples": 22.1,
      "learning_rate": 1.634325999607811e-05,
      "loss": 0.021,
      "positive_teacher_pairs": 24271.0,
      "query_h_norm_ratio": 0.41627306342124937,
      "relation_loss": 0.002235451294109225,
      "selected_pairs": 24271.0,
      "step": 22450,
      "student_distance_mean": 22.85482406616211,
      "teacher_distance_mean": 0.844015383720398,
      "valid_action_samples": 214.0,
      "weighted_auxiliary_loss": 0.002235451294109225,
      "weighted_relation_loss": 0.002235451294109225
    },
    {
      "action_query_batch_std": 0.35296079218387605,
      "action_query_norm": 40.951611328125,
      "action_relation_loss": 0.002238153188955039,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.010289990645465,
      "flow_matching_loss": 0.01956739742308855,
      "grad_norm": 0.09858188778162003,
      "h_stock_valid_norm": 98.29468688964843,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.6302521329857045e-05,
      "loss": 0.0218,
      "positive_teacher_pairs": 27089.3,
      "query_h_norm_ratio": 0.41662359833717344,
      "relation_loss": 0.002238153188955039,
      "selected_pairs": 27089.3,
      "step": 22460,
      "student_distance_mean": 23.191672134399415,
      "teacher_distance_mean": 0.857388299703598,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.002238153188955039,
      "weighted_relation_loss": 0.002238153188955039
    },
    {
      "action_query_batch_std": 0.35436886847019194,
      "action_query_norm": 40.962345886230466,
      "action_relation_loss": 0.002173441497143358,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.019644527595883,
      "flow_matching_loss": 0.019914102740585804,
      "grad_norm": 0.12154101580381393,
      "h_stock_valid_norm": 97.85830230712891,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.6261823609250027e-05,
      "loss": 0.0221,
      "positive_teacher_pairs": 27010.6,
      "query_h_norm_ratio": 0.41859171688556673,
      "relation_loss": 0.002173441497143358,
      "selected_pairs": 27010.6,
      "step": 22470,
      "student_distance_mean": 23.318238830566408,
      "teacher_distance_mean": 0.860969877243042,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002173441497143358,
      "weighted_relation_loss": 0.002173441497143358
    },
    {
      "action_query_batch_std": 0.3550168961286545,
      "action_query_norm": 40.955804443359376,
      "action_relation_loss": 0.0022227165172807872,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.028999064546305,
      "flow_matching_loss": 0.01767159104347229,
      "grad_norm": 0.09910617768764496,
      "h_stock_valid_norm": 97.83523406982422,
      "invalid_tail_samples": 22.2,
      "learning_rate": 1.6221166883708627e-05,
      "loss": 0.0199,
      "positive_teacher_pairs": 27222.1,
      "query_h_norm_ratio": 0.4186233103275299,
      "relation_loss": 0.0022227165172807872,
      "selected_pairs": 27222.1,
      "step": 22480,
      "student_distance_mean": 23.239179611206055,
      "teacher_distance_mean": 0.8579420566558837,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0022227165172807872,
      "weighted_relation_loss": 0.0022227165172807872
    },
    {
      "action_query_batch_std": 0.3500003606081009,
      "action_query_norm": 40.95580368041992,
      "action_relation_loss": 0.0022366487653926017,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.038353601496727,
      "flow_matching_loss": 0.019305373076349497,
      "grad_norm": 0.10824110358953476,
      "h_stock_valid_norm": 97.88977661132813,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.6180551202634603e-05,
      "loss": 0.0215,
      "positive_teacher_pairs": 26906.2,
      "query_h_norm_ratio": 0.41838937997817993,
      "relation_loss": 0.0022366487653926017,
      "selected_pairs": 26906.2,
      "step": 22490,
      "student_distance_mean": 23.026094436645508,
      "teacher_distance_mean": 0.8498268187046051,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0022366487653926017,
      "weighted_relation_loss": 0.0022366487653926017
    },
    {
      "action_query_batch_std": 0.35213756263256074,
      "action_query_norm": 40.93923797607422,
      "action_relation_loss": 0.0022745259338989854,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.047708138447145,
      "flow_matching_loss": 0.01791217029094696,
      "grad_norm": 0.1527150720357895,
      "h_stock_valid_norm": 98.04242477416992,
      "invalid_tail_samples": 23.2,
      "learning_rate": 1.613997661537981e-05,
      "loss": 0.0202,
      "positive_teacher_pairs": 26996.3,
      "query_h_norm_ratio": 0.4175682157278061,
      "relation_loss": 0.0022745259338989854,
      "selected_pairs": 26996.3,
      "step": 22500,
      "student_distance_mean": 23.19380111694336,
      "teacher_distance_mean": 0.8583734571933747,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0022745259338989854,
      "weighted_relation_loss": 0.0022745259338989854
    },
    {
      "action_query_batch_std": 0.34881835877895356,
      "action_query_norm": 40.91412467956543,
      "action_relation_loss": 0.0023192269261926414,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.057062675397567,
      "flow_matching_loss": 0.01877179937437177,
      "grad_norm": 0.09712641686201096,
      "h_stock_valid_norm": 98.2964698791504,
      "invalid_tail_samples": 22.9,
      "learning_rate": 1.609944317124624e-05,
      "loss": 0.0211,
      "positive_teacher_pairs": 27057.3,
      "query_h_norm_ratio": 0.41623497009277344,
      "relation_loss": 0.0023192269261926414,
      "selected_pairs": 27057.3,
      "step": 22510,
      "student_distance_mean": 22.999904441833497,
      "teacher_distance_mean": 0.8511442601680755,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0023192269261926414,
      "weighted_relation_loss": 0.0023192269261926414
    },
    {
      "action_query_batch_std": 0.35575630366802213,
      "action_query_norm": 40.97647705078125,
      "action_relation_loss": 0.0022168920142576097,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.06641721234799,
      "flow_matching_loss": 0.018523948453366756,
      "grad_norm": 0.12874625623226166,
      "h_stock_valid_norm": 98.20391006469727,
      "invalid_tail_samples": 24.7,
      "learning_rate": 1.605895091948582e-05,
      "loss": 0.0207,
      "positive_teacher_pairs": 26638.9,
      "query_h_norm_ratio": 0.4172610968351364,
      "relation_loss": 0.0022168920142576097,
      "selected_pairs": 26638.9,
      "step": 22520,
      "student_distance_mean": 23.278434181213377,
      "teacher_distance_mean": 0.8578325390815735,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0022168920142576097,
      "weighted_relation_loss": 0.0022168920142576097
    },
    {
      "action_query_batch_std": 0.3524499982595444,
      "action_query_norm": 40.944759368896484,
      "action_relation_loss": 0.0022598672658205032,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.07577174929841,
      "flow_matching_loss": 0.018351737316697835,
      "grad_norm": 0.08552693575620651,
      "h_stock_valid_norm": 98.20086669921875,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.6018499909300477e-05,
      "loss": 0.0206,
      "positive_teacher_pairs": 27110.1,
      "query_h_norm_ratio": 0.4169521301984787,
      "relation_loss": 0.0022598672658205032,
      "selected_pairs": 27110.1,
      "step": 22530,
      "student_distance_mean": 23.122834777832033,
      "teacher_distance_mean": 0.8524871408939362,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0022598672658205032,
      "weighted_relation_loss": 0.0022598672658205032
    },
    {
      "action_query_batch_std": 0.3501693785190582,
      "action_query_norm": 40.93105201721191,
      "action_relation_loss": 0.0023011012934148313,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.08512628624883,
      "flow_matching_loss": 0.01790554225444794,
      "grad_norm": 0.10420790314674377,
      "h_stock_valid_norm": 98.274853515625,
      "invalid_tail_samples": 22.5,
      "learning_rate": 1.597809018984199e-05,
      "loss": 0.0202,
      "positive_teacher_pairs": 27153.8,
      "query_h_norm_ratio": 0.4165007770061493,
      "relation_loss": 0.0023011012934148313,
      "selected_pairs": 27153.8,
      "step": 22540,
      "student_distance_mean": 23.038796043395998,
      "teacher_distance_mean": 0.8515476763248444,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0023011012934148313,
      "weighted_relation_loss": 0.0023011012934148313
    },
    {
      "action_query_batch_std": 0.34696981906890867,
      "action_query_norm": 40.90302696228027,
      "action_relation_loss": 0.0022526882123202084,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.09448082319925,
      "flow_matching_loss": 0.019514921214431526,
      "grad_norm": 0.1151537224650383,
      "h_stock_valid_norm": 98.11539230346679,
      "invalid_tail_samples": 22.5,
      "learning_rate": 1.593772181021196e-05,
      "loss": 0.0218,
      "positive_teacher_pairs": 27152.7,
      "query_h_norm_ratio": 0.41689073145389555,
      "relation_loss": 0.0022526882123202084,
      "selected_pairs": 27152.7,
      "step": 22550,
      "student_distance_mean": 22.83361473083496,
      "teacher_distance_mean": 0.8447019338607789,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0022526882123202084,
      "weighted_relation_loss": 0.0022526882123202084
    },
    {
      "action_query_batch_std": 0.3533198952674866,
      "action_query_norm": 40.9704833984375,
      "action_relation_loss": 0.0021763896220363677,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.103835360149674,
      "flow_matching_loss": 0.0190091154538095,
      "grad_norm": 0.08335302770137787,
      "h_stock_valid_norm": 97.91030502319336,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.5897394819461813e-05,
      "loss": 0.0212,
      "positive_teacher_pairs": 27009.7,
      "query_h_norm_ratio": 0.41845051646232606,
      "relation_loss": 0.0021763896220363677,
      "selected_pairs": 27009.7,
      "step": 22560,
      "student_distance_mean": 23.221919059753418,
      "teacher_distance_mean": 0.8585618019104004,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0021763896220363677,
      "weighted_relation_loss": 0.0021763896220363677
    },
    {
      "action_query_batch_std": 0.35359120070934297,
      "action_query_norm": 40.956132125854495,
      "action_relation_loss": 0.002189308358356357,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.113189897100092,
      "flow_matching_loss": 0.01808615503832698,
      "grad_norm": 0.09213841706514359,
      "h_stock_valid_norm": 98.1633201599121,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.5857109266592625e-05,
      "loss": 0.0203,
      "positive_teacher_pairs": 27103.4,
      "query_h_norm_ratio": 0.4172267585992813,
      "relation_loss": 0.002189308358356357,
      "selected_pairs": 27103.4,
      "step": 22570,
      "student_distance_mean": 23.207831192016602,
      "teacher_distance_mean": 0.858489853143692,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.002189308358356357,
      "weighted_relation_loss": 0.002189308358356357
    },
    {
      "action_query_batch_std": 0.35030455589294435,
      "action_query_norm": 40.90352363586426,
      "action_relation_loss": 0.0022807331755757333,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.122544434050514,
      "flow_matching_loss": 0.01946716532111168,
      "grad_norm": 0.10002139955759048,
      "h_stock_valid_norm": 98.27940444946289,
      "invalid_tail_samples": 26.5,
      "learning_rate": 1.581686520055514e-05,
      "loss": 0.0217,
      "positive_teacher_pairs": 26226.8,
      "query_h_norm_ratio": 0.41619797646999357,
      "relation_loss": 0.0022807331755757333,
      "selected_pairs": 26226.8,
      "step": 22580,
      "student_distance_mean": 23.044969177246095,
      "teacher_distance_mean": 0.8551080346107482,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.0022807331755757333,
      "weighted_relation_loss": 0.0022807331755757333
    },
    {
      "action_query_batch_std": 0.353768527507782,
      "action_query_norm": 40.938972854614256,
      "action_relation_loss": 0.0022029954358004033,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.131898971000936,
      "flow_matching_loss": 0.01989644467830658,
      "grad_norm": 0.10191550105810165,
      "h_stock_valid_norm": 98.3064178466797,
      "invalid_tail_samples": 25.0,
      "learning_rate": 1.5776662670249704e-05,
      "loss": 0.0221,
      "positive_teacher_pairs": 26575.8,
      "query_h_norm_ratio": 0.41644780039787294,
      "relation_loss": 0.0022029954358004033,
      "selected_pairs": 26575.8,
      "step": 22590,
      "student_distance_mean": 23.336264610290527,
      "teacher_distance_mean": 0.862355375289917,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0022029954358004033,
      "weighted_relation_loss": 0.0022029954358004033
    },
    {
      "action_query_batch_std": 0.35786037147045135,
      "action_query_norm": 40.97918434143067,
      "action_relation_loss": 0.002175613527651876,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.141253507951358,
      "flow_matching_loss": 0.019294323958456518,
      "grad_norm": 0.12248951941728592,
      "h_stock_valid_norm": 98.28541107177735,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.573650172452615e-05,
      "loss": 0.0215,
      "positive_teacher_pairs": 26731.9,
      "query_h_norm_ratio": 0.4169442743062973,
      "relation_loss": 0.002175613527651876,
      "selected_pairs": 26731.9,
      "step": 22600,
      "student_distance_mean": 23.476053619384764,
      "teacher_distance_mean": 0.8660019338130951,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002175613527651876,
      "weighted_relation_loss": 0.002175613527651876
    },
    {
      "action_query_batch_std": 0.35107546448707583,
      "action_query_norm": 40.92860336303711,
      "action_relation_loss": 0.002245326666161418,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.150608044901777,
      "flow_matching_loss": 0.01735308105126023,
      "grad_norm": 0.11721895635128021,
      "h_stock_valid_norm": 98.05516891479492,
      "invalid_tail_samples": 22.6,
      "learning_rate": 1.5696382412183852e-05,
      "loss": 0.0196,
      "positive_teacher_pairs": 27125.9,
      "query_h_norm_ratio": 0.4174077302217484,
      "relation_loss": 0.002245326666161418,
      "selected_pairs": 27125.9,
      "step": 22610,
      "student_distance_mean": 23.164657402038575,
      "teacher_distance_mean": 0.8560261487960815,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.002245326666161418,
      "weighted_relation_loss": 0.002245326666161418
    },
    {
      "action_query_batch_std": 0.3482648730278015,
      "action_query_norm": 40.9119514465332,
      "action_relation_loss": 0.0021692376816645265,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.1599625818522,
      "flow_matching_loss": 0.019656293746083974,
      "grad_norm": 0.10905440896749496,
      "h_stock_valid_norm": 97.92607192993164,
      "invalid_tail_samples": 23.3,
      "learning_rate": 1.565630478197155e-05,
      "loss": 0.0218,
      "positive_teacher_pairs": 26965.0,
      "query_h_norm_ratio": 0.41778606474399566,
      "relation_loss": 0.0021692376816645265,
      "selected_pairs": 26965.0,
      "step": 22620,
      "student_distance_mean": 22.93798999786377,
      "teacher_distance_mean": 0.8477884113788605,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0021692376816645265,
      "weighted_relation_loss": 0.0021692376816645265
    },
    {
      "action_query_batch_std": 0.34926846623420715,
      "action_query_norm": 40.92271461486816,
      "action_relation_loss": 0.0022364453179761767,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.16931711880262,
      "flow_matching_loss": 0.019594068825244903,
      "grad_norm": 0.12013394385576248,
      "h_stock_valid_norm": 97.7669288635254,
      "invalid_tail_samples": 20.2,
      "learning_rate": 1.5616268882587332e-05,
      "loss": 0.0218,
      "positive_teacher_pairs": 27688.0,
      "query_h_norm_ratio": 0.4185789316892624,
      "relation_loss": 0.0022364453179761767,
      "selected_pairs": 27688.0,
      "step": 22630,
      "student_distance_mean": 22.997855567932127,
      "teacher_distance_mean": 0.8505739867687225,
      "valid_action_samples": 235.8,
      "weighted_auxiliary_loss": 0.0022364453179761767,
      "weighted_relation_loss": 0.0022364453179761767
    },
    {
      "action_query_batch_std": 0.3494103759527206,
      "action_query_norm": 40.90648422241211,
      "action_relation_loss": 0.0021635858807712792,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.17867165575304,
      "flow_matching_loss": 0.019025259464979172,
      "grad_norm": 0.10209611058235168,
      "h_stock_valid_norm": 97.76040573120117,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.5576274762678594e-05,
      "loss": 0.0212,
      "positive_teacher_pairs": 26874.7,
      "query_h_norm_ratio": 0.4184385806322098,
      "relation_loss": 0.0021635858807712792,
      "selected_pairs": 26874.7,
      "step": 22640,
      "student_distance_mean": 23.012207984924316,
      "teacher_distance_mean": 0.8487162470817566,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0021635858807712792,
      "weighted_relation_loss": 0.0021635858807712792
    },
    {
      "action_query_batch_std": 0.34952176809310914,
      "action_query_norm": 40.90249443054199,
      "action_relation_loss": 0.0022221044171601536,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.18802619270346,
      "flow_matching_loss": 0.01823324868455529,
      "grad_norm": 0.08942200243473053,
      "h_stock_valid_norm": 97.71342010498047,
      "invalid_tail_samples": 25.0,
      "learning_rate": 1.5536322470841952e-05,
      "loss": 0.0205,
      "positive_teacher_pairs": 26579.3,
      "query_h_norm_ratio": 0.4185983031988144,
      "relation_loss": 0.0022221044171601536,
      "selected_pairs": 26579.3,
      "step": 22650,
      "student_distance_mean": 23.003540802001954,
      "teacher_distance_mean": 0.850579684972763,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0022221044171601536,
      "weighted_relation_loss": 0.0022221044171601536
    },
    {
      "action_query_batch_std": 0.3490843653678894,
      "action_query_norm": 40.906215286254884,
      "action_relation_loss": 0.0022405080730095507,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.197380729653883,
      "flow_matching_loss": 0.019975644443184136,
      "grad_norm": 0.10367430001497269,
      "h_stock_valid_norm": 97.97197570800782,
      "invalid_tail_samples": 24.6,
      "learning_rate": 1.549641205562324e-05,
      "loss": 0.0222,
      "positive_teacher_pairs": 26673.1,
      "query_h_norm_ratio": 0.41753241419792175,
      "relation_loss": 0.0022405080730095507,
      "selected_pairs": 26673.1,
      "step": 22660,
      "student_distance_mean": 23.082332801818847,
      "teacher_distance_mean": 0.8532684981822968,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0022405080730095507,
      "weighted_relation_loss": 0.0022405080730095507
    },
    {
      "action_query_batch_std": 0.35119680166244505,
      "action_query_norm": 40.9122615814209,
      "action_relation_loss": 0.002161772735416889,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.2067352666043,
      "flow_matching_loss": 0.019552530907094478,
      "grad_norm": 0.09070755541324615,
      "h_stock_valid_norm": 97.92974472045898,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.5456543565517334e-05,
      "loss": 0.0217,
      "positive_teacher_pairs": 26879.4,
      "query_h_norm_ratio": 0.4177752614021301,
      "relation_loss": 0.002161772735416889,
      "selected_pairs": 26879.4,
      "step": 22670,
      "student_distance_mean": 23.114950752258302,
      "teacher_distance_mean": 0.8536933779716491,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.002161772735416889,
      "weighted_relation_loss": 0.002161772735416889
    },
    {
      "action_query_batch_std": 0.35223848223686216,
      "action_query_norm": 40.92066612243652,
      "action_relation_loss": 0.0022837764117866755,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.216089803554723,
      "flow_matching_loss": 0.018382636923342945,
      "grad_norm": 0.09176210314035416,
      "h_stock_valid_norm": 97.86638488769532,
      "invalid_tail_samples": 26.8,
      "learning_rate": 1.541671704896825e-05,
      "loss": 0.0207,
      "positive_teacher_pairs": 26164.2,
      "query_h_norm_ratio": 0.4181288093328476,
      "relation_loss": 0.0022837764117866755,
      "selected_pairs": 26164.2,
      "step": 22680,
      "student_distance_mean": 23.1839168548584,
      "teacher_distance_mean": 0.8543808698654175,
      "valid_action_samples": 229.2,
      "weighted_auxiliary_loss": 0.0022837764117866755,
      "weighted_relation_loss": 0.0022837764117866755
    },
    {
      "action_query_batch_std": 0.35078323185443877,
      "action_query_norm": 40.92591018676758,
      "action_relation_loss": 0.0022802296094596385,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.225444340505145,
      "flow_matching_loss": 0.020386040583252905,
      "grad_norm": 0.19413813948631287,
      "h_stock_valid_norm": 97.74632110595704,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.5376932554368938e-05,
      "loss": 0.0227,
      "positive_teacher_pairs": 27015.0,
      "query_h_norm_ratio": 0.4186978995800018,
      "relation_loss": 0.0022802296094596385,
      "selected_pairs": 27015.0,
      "step": 22690,
      "student_distance_mean": 23.048692893981933,
      "teacher_distance_mean": 0.8490243494510651,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0022802296094596385,
      "weighted_relation_loss": 0.0022802296094596385
    },
    {
      "action_query_batch_std": 0.3488989472389221,
      "action_query_norm": 40.91690673828125,
      "action_relation_loss": 0.002259928616695106,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.234798877455567,
      "flow_matching_loss": 0.018743183091282846,
      "grad_norm": 0.27576932311058044,
      "h_stock_valid_norm": 98.06450958251953,
      "invalid_tail_samples": 21.4,
      "learning_rate": 1.5337190130061318e-05,
      "loss": 0.021,
      "positive_teacher_pairs": 27407.3,
      "query_h_norm_ratio": 0.41724573373794555,
      "relation_loss": 0.002259928616695106,
      "selected_pairs": 27407.3,
      "step": 22700,
      "student_distance_mean": 22.917891693115234,
      "teacher_distance_mean": 0.8454705774784088,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.002259928616695106,
      "weighted_relation_loss": 0.002259928616695106
    },
    {
      "action_query_batch_std": 0.35486293137073516,
      "action_query_norm": 40.950544357299805,
      "action_relation_loss": 0.0022241999628022312,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.244153414405986,
      "flow_matching_loss": 0.01810530060902238,
      "grad_norm": 0.10478392243385315,
      "h_stock_valid_norm": 98.30147171020508,
      "invalid_tail_samples": 25.3,
      "learning_rate": 1.5297489824336165e-05,
      "loss": 0.0203,
      "positive_teacher_pairs": 26504.0,
      "query_h_norm_ratio": 0.4165847599506378,
      "relation_loss": 0.0022241999628022312,
      "selected_pairs": 26504.0,
      "step": 22710,
      "student_distance_mean": 23.25147819519043,
      "teacher_distance_mean": 0.8603722155094147,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0022241999628022312,
      "weighted_relation_loss": 0.0022241999628022312
    },
    {
      "action_query_batch_std": 0.34935817718505857,
      "action_query_norm": 40.920935440063474,
      "action_relation_loss": 0.002159284777007997,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.253507951356408,
      "flow_matching_loss": 0.017448091972619297,
      "grad_norm": 0.10147882997989655,
      "h_stock_valid_norm": 98.0514747619629,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.5257831685433121e-05,
      "loss": 0.0196,
      "positive_teacher_pairs": 27012.1,
      "query_h_norm_ratio": 0.41734306514263153,
      "relation_loss": 0.002159284777007997,
      "selected_pairs": 27012.1,
      "step": 22720,
      "student_distance_mean": 23.037029647827147,
      "teacher_distance_mean": 0.8520468533039093,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002159284777007997,
      "weighted_relation_loss": 0.002159284777007997
    },
    {
      "action_query_batch_std": 0.34670062363147736,
      "action_query_norm": 40.894738006591794,
      "action_relation_loss": 0.0022326358128339054,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.26286248830683,
      "flow_matching_loss": 0.018794896174222232,
      "grad_norm": 0.11605657637119293,
      "h_stock_valid_norm": 98.49314956665039,
      "invalid_tail_samples": 25.0,
      "learning_rate": 1.521821576154055e-05,
      "loss": 0.021,
      "positive_teacher_pairs": 26580.1,
      "query_h_norm_ratio": 0.415206116437912,
      "relation_loss": 0.0022326358128339054,
      "selected_pairs": 26580.1,
      "step": 22730,
      "student_distance_mean": 22.83160572052002,
      "teacher_distance_mean": 0.8469596922397613,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0022326358128339054,
      "weighted_relation_loss": 0.0022326358128339054
    },
    {
      "action_query_batch_std": 0.35368116199970245,
      "action_query_norm": 40.91840934753418,
      "action_relation_loss": 0.0022519099758937953,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.27221702525725,
      "flow_matching_loss": 0.01762992814183235,
      "grad_norm": 0.10241924971342087,
      "h_stock_valid_norm": 98.4147560119629,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.5178642100795542e-05,
      "loss": 0.0199,
      "positive_teacher_pairs": 26806.9,
      "query_h_norm_ratio": 0.41577699184417727,
      "relation_loss": 0.0022519099758937953,
      "selected_pairs": 26806.9,
      "step": 22740,
      "student_distance_mean": 23.16023998260498,
      "teacher_distance_mean": 0.8576762735843658,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0022519099758937953,
      "weighted_relation_loss": 0.0022519099758937953
    },
    {
      "action_query_batch_std": 0.35059289932250975,
      "action_query_norm": 40.89923782348633,
      "action_relation_loss": 0.002193937113042921,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.28157156220767,
      "flow_matching_loss": 0.018616346083581446,
      "grad_norm": 0.09690862894058228,
      "h_stock_valid_norm": 98.52701416015626,
      "invalid_tail_samples": 23.8,
      "learning_rate": 1.5139110751283819e-05,
      "loss": 0.0208,
      "positive_teacher_pairs": 26852.8,
      "query_h_norm_ratio": 0.415110844373703,
      "relation_loss": 0.002193937113042921,
      "selected_pairs": 26852.8,
      "step": 22750,
      "student_distance_mean": 23.050198554992676,
      "teacher_distance_mean": 0.8533592879772186,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.002193937113042921,
      "weighted_relation_loss": 0.002193937113042921
    },
    {
      "action_query_batch_std": 0.34779036343097686,
      "action_query_norm": 40.894575119018555,
      "action_relation_loss": 0.002226987574249506,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.290926099158092,
      "flow_matching_loss": 0.02093381164595485,
      "grad_norm": 0.0948602706193924,
      "h_stock_valid_norm": 98.47644805908203,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.5099621761039683e-05,
      "loss": 0.0232,
      "positive_teacher_pairs": 27105.1,
      "query_h_norm_ratio": 0.415274515748024,
      "relation_loss": 0.002226987574249506,
      "selected_pairs": 27105.1,
      "step": 22760,
      "student_distance_mean": 22.849188995361327,
      "teacher_distance_mean": 0.8437100291252136,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.002226987574249506,
      "weighted_relation_loss": 0.002226987574249506
    },
    {
      "action_query_batch_std": 0.3510949581861496,
      "action_query_norm": 40.929095077514646,
      "action_relation_loss": 0.002206914231646806,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.300280636108514,
      "flow_matching_loss": 0.019292285945266486,
      "grad_norm": 0.10223927348852158,
      "h_stock_valid_norm": 98.04336166381836,
      "invalid_tail_samples": 21.2,
      "learning_rate": 1.5060175178046016e-05,
      "loss": 0.0215,
      "positive_teacher_pairs": 27452.8,
      "query_h_norm_ratio": 0.4174619674682617,
      "relation_loss": 0.002206914231646806,
      "selected_pairs": 27452.8,
      "step": 22770,
      "student_distance_mean": 23.08407516479492,
      "teacher_distance_mean": 0.8515381217002869,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.002206914231646806,
      "weighted_relation_loss": 0.002206914231646806
    },
    {
      "action_query_batch_std": 0.35583006143569945,
      "action_query_norm": 40.95743827819824,
      "action_relation_loss": 0.0020904222852550446,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.309635173058933,
      "flow_matching_loss": 0.0201387002132833,
      "grad_norm": 0.1167697086930275,
      "h_stock_valid_norm": 97.86989212036133,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.5020771050234117e-05,
      "loss": 0.0222,
      "positive_teacher_pairs": 26911.3,
      "query_h_norm_ratio": 0.4184907883405685,
      "relation_loss": 0.0020904222852550446,
      "selected_pairs": 26911.3,
      "step": 22780,
      "student_distance_mean": 23.371187019348145,
      "teacher_distance_mean": 0.863712626695633,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0020904222852550446,
      "weighted_relation_loss": 0.0020904222852550446
    },
    {
      "action_query_batch_std": 0.3507529318332672,
      "action_query_norm": 40.90351867675781,
      "action_relation_loss": 0.002163197030313313,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.318989710009355,
      "flow_matching_loss": 0.018885099422186614,
      "grad_norm": 0.147997185587883,
      "h_stock_valid_norm": 97.7584846496582,
      "invalid_tail_samples": 26.9,
      "learning_rate": 1.4981409425483716e-05,
      "loss": 0.021,
      "positive_teacher_pairs": 26138.1,
      "query_h_norm_ratio": 0.4184186518192291,
      "relation_loss": 0.002163197030313313,
      "selected_pairs": 26138.1,
      "step": 22790,
      "student_distance_mean": 23.09907474517822,
      "teacher_distance_mean": 0.8521228432655334,
      "valid_action_samples": 229.1,
      "weighted_auxiliary_loss": 0.002163197030313313,
      "weighted_relation_loss": 0.002163197030313313
    },
    {
      "action_query_batch_std": 0.3497656971216202,
      "action_query_norm": 40.90317840576172,
      "action_relation_loss": 0.002169678499922156,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.328344246959777,
      "flow_matching_loss": 0.017253033723682164,
      "grad_norm": 0.08664190024137497,
      "h_stock_valid_norm": 97.99648742675781,
      "invalid_tail_samples": 24.6,
      "learning_rate": 1.4942090351622883e-05,
      "loss": 0.0194,
      "positive_teacher_pairs": 26670.9,
      "query_h_norm_ratio": 0.4173961341381073,
      "relation_loss": 0.002169678499922156,
      "selected_pairs": 26670.9,
      "step": 22800,
      "student_distance_mean": 23.141951179504396,
      "teacher_distance_mean": 0.8547110676765441,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.002169678499922156,
      "weighted_relation_loss": 0.002169678499922156
    },
    {
      "action_query_batch_std": 0.3491384118795395,
      "action_query_norm": 40.88763694763183,
      "action_relation_loss": 0.00226747109554708,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.337698783910195,
      "flow_matching_loss": 0.017619803827255965,
      "grad_norm": 0.10323385894298553,
      "h_stock_valid_norm": 98.16284942626953,
      "invalid_tail_samples": 25.5,
      "learning_rate": 1.490281387642804e-05,
      "loss": 0.0199,
      "positive_teacher_pairs": 26455.4,
      "query_h_norm_ratio": 0.41653220653533934,
      "relation_loss": 0.00226747109554708,
      "selected_pairs": 26455.4,
      "step": 22810,
      "student_distance_mean": 22.91903190612793,
      "teacher_distance_mean": 0.849175900220871,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.00226747109554708,
      "weighted_relation_loss": 0.00226747109554708
    },
    {
      "action_query_batch_std": 0.3479614436626434,
      "action_query_norm": 40.899970626831056,
      "action_relation_loss": 0.0022447420051321387,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.347053320860617,
      "flow_matching_loss": 0.0175293886102736,
      "grad_norm": 0.10585862398147583,
      "h_stock_valid_norm": 98.0390853881836,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.4863580047623765e-05,
      "loss": 0.0198,
      "positive_teacher_pairs": 26869.5,
      "query_h_norm_ratio": 0.41718530654907227,
      "relation_loss": 0.0022447420051321387,
      "selected_pairs": 26869.5,
      "step": 22820,
      "student_distance_mean": 22.914368057250975,
      "teacher_distance_mean": 0.8484698235988617,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0022447420051321387,
      "weighted_relation_loss": 0.0022447420051321387
    },
    {
      "action_query_batch_std": 0.35122097134590147,
      "action_query_norm": 40.904660034179685,
      "action_relation_loss": 0.0021403686492703854,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.35640785781104,
      "flow_matching_loss": 0.01686710650101304,
      "grad_norm": 0.0960199385881424,
      "h_stock_valid_norm": 98.38430862426758,
      "invalid_tail_samples": 24.9,
      "learning_rate": 1.4824388912882897e-05,
      "loss": 0.019,
      "positive_teacher_pairs": 26595.4,
      "query_h_norm_ratio": 0.41576640009880067,
      "relation_loss": 0.0021403686492703854,
      "selected_pairs": 26595.4,
      "step": 22830,
      "student_distance_mean": 23.082257652282713,
      "teacher_distance_mean": 0.854604172706604,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0021403686492703854,
      "weighted_relation_loss": 0.0021403686492703854
    },
    {
      "action_query_batch_std": 0.34973866641521456,
      "action_query_norm": 40.90168380737305,
      "action_relation_loss": 0.0021970242960378528,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.36576239476146,
      "flow_matching_loss": 0.018333970755338668,
      "grad_norm": 0.10541188716888428,
      "h_stock_valid_norm": 98.16550216674804,
      "invalid_tail_samples": 22.1,
      "learning_rate": 1.4785240519826344e-05,
      "loss": 0.0205,
      "positive_teacher_pairs": 27242.7,
      "query_h_norm_ratio": 0.4166644632816315,
      "relation_loss": 0.0021970242960378528,
      "selected_pairs": 27242.7,
      "step": 22840,
      "student_distance_mean": 22.902908897399904,
      "teacher_distance_mean": 0.8468108654022217,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0021970242960378528,
      "weighted_relation_loss": 0.0021970242960378528
    },
    {
      "action_query_batch_std": 0.3520269632339478,
      "action_query_norm": 40.91297378540039,
      "action_relation_loss": 0.0022273391019552945,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.37511693171188,
      "flow_matching_loss": 0.01734156599268317,
      "grad_norm": 0.12159529328346252,
      "h_stock_valid_norm": 97.94789581298828,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.4746134916023097e-05,
      "loss": 0.0196,
      "positive_teacher_pairs": 26755.4,
      "query_h_norm_ratio": 0.41770492792129515,
      "relation_loss": 0.0022273391019552945,
      "selected_pairs": 26755.4,
      "step": 22850,
      "student_distance_mean": 23.08509178161621,
      "teacher_distance_mean": 0.8531951546669007,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0022273391019552945,
      "weighted_relation_loss": 0.0022273391019552945
    },
    {
      "action_query_batch_std": 0.34726912081241607,
      "action_query_norm": 40.86095924377442,
      "action_relation_loss": 0.0022204141831025483,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.3844714686623,
      "flow_matching_loss": 0.01835748106241226,
      "grad_norm": 0.09926709532737732,
      "h_stock_valid_norm": 98.1355224609375,
      "invalid_tail_samples": 25.6,
      "learning_rate": 1.4707072148990141e-05,
      "loss": 0.0206,
      "positive_teacher_pairs": 26438.2,
      "query_h_norm_ratio": 0.4163740485906601,
      "relation_loss": 0.0022204141831025483,
      "selected_pairs": 26438.2,
      "step": 22860,
      "student_distance_mean": 22.867959594726564,
      "teacher_distance_mean": 0.8463808774948121,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0022204141831025483,
      "weighted_relation_loss": 0.0022204141831025483
    },
    {
      "action_query_batch_std": 0.35062127709388735,
      "action_query_norm": 40.89724845886231,
      "action_relation_loss": 0.002243541181087494,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.393826005612723,
      "flow_matching_loss": 0.019162527844309808,
      "grad_norm": 0.10249580442905426,
      "h_stock_valid_norm": 97.99340896606445,
      "invalid_tail_samples": 24.8,
      "learning_rate": 1.4668052266192423e-05,
      "loss": 0.0214,
      "positive_teacher_pairs": 26616.0,
      "query_h_norm_ratio": 0.41734840273857116,
      "relation_loss": 0.002243541181087494,
      "selected_pairs": 26616.0,
      "step": 22870,
      "student_distance_mean": 23.04878749847412,
      "teacher_distance_mean": 0.8514677226543427,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.002243541181087494,
      "weighted_relation_loss": 0.002243541181087494
    },
    {
      "action_query_batch_std": 0.3518178880214691,
      "action_query_norm": 40.89996070861817,
      "action_relation_loss": 0.002161222184076905,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.403180542563142,
      "flow_matching_loss": 0.01957376580685377,
      "grad_norm": 0.10232657194137573,
      "h_stock_valid_norm": 98.01754684448242,
      "invalid_tail_samples": 21.1,
      "learning_rate": 1.4629075315042795e-05,
      "loss": 0.0217,
      "positive_teacher_pairs": 27481.5,
      "query_h_norm_ratio": 0.4172767102718353,
      "relation_loss": 0.002161222184076905,
      "selected_pairs": 27481.5,
      "step": 22880,
      "student_distance_mean": 23.224177360534668,
      "teacher_distance_mean": 0.8572475194931031,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.002161222184076905,
      "weighted_relation_loss": 0.002161222184076905
    },
    {
      "action_query_batch_std": 0.3504867136478424,
      "action_query_norm": 40.89320106506348,
      "action_relation_loss": 0.002164178225211799,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.412535079513564,
      "flow_matching_loss": 0.01927729919552803,
      "grad_norm": 0.10128524154424667,
      "h_stock_valid_norm": 97.71122817993164,
      "invalid_tail_samples": 23.0,
      "learning_rate": 1.4590141342901925e-05,
      "loss": 0.0214,
      "positive_teacher_pairs": 27038.0,
      "query_h_norm_ratio": 0.4185125470161438,
      "relation_loss": 0.002164178225211799,
      "selected_pairs": 27038.0,
      "step": 22890,
      "student_distance_mean": 23.07155246734619,
      "teacher_distance_mean": 0.8523436665534974,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.002164178225211799,
      "weighted_relation_loss": 0.002164178225211799
    },
    {
      "action_query_batch_std": 0.3565142720937729,
      "action_query_norm": 40.9425350189209,
      "action_relation_loss": 0.002054353873245418,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.421889616463986,
      "flow_matching_loss": 0.019389237463474273,
      "grad_norm": 0.10256616026163101,
      "h_stock_valid_norm": 97.86513290405273,
      "invalid_tail_samples": 23.4,
      "learning_rate": 1.4551250397078253e-05,
      "loss": 0.0214,
      "positive_teacher_pairs": 26945.3,
      "query_h_norm_ratio": 0.41836076974868774,
      "relation_loss": 0.002054353873245418,
      "selected_pairs": 26945.3,
      "step": 22900,
      "student_distance_mean": 23.495555305480956,
      "teacher_distance_mean": 0.8691042780876159,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002054353873245418,
      "weighted_relation_loss": 0.002054353873245418
    },
    {
      "action_query_batch_std": 0.3553712725639343,
      "action_query_norm": 40.949758529663086,
      "action_relation_loss": 0.0022733663441613316,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.431244153414404,
      "flow_matching_loss": 0.01781175211071968,
      "grad_norm": 0.14570240676403046,
      "h_stock_valid_norm": 98.08903427124024,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.4512402524827945e-05,
      "loss": 0.0201,
      "positive_teacher_pairs": 26899.5,
      "query_h_norm_ratio": 0.4174776881933212,
      "relation_loss": 0.0022733663441613316,
      "selected_pairs": 26899.5,
      "step": 22910,
      "student_distance_mean": 23.387875747680663,
      "teacher_distance_mean": 0.8660071969032288,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0022733663441613316,
      "weighted_relation_loss": 0.0022733663441613316
    },
    {
      "action_query_batch_std": 0.35583025217056274,
      "action_query_norm": 40.9258472442627,
      "action_relation_loss": 0.0020992020959965884,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.440598690364826,
      "flow_matching_loss": 0.01825886024162173,
      "grad_norm": 0.13868993520736694,
      "h_stock_valid_norm": 98.00345077514649,
      "invalid_tail_samples": 23.0,
      "learning_rate": 1.447359777335482e-05,
      "loss": 0.0204,
      "positive_teacher_pairs": 27034.0,
      "query_h_norm_ratio": 0.4175970911979675,
      "relation_loss": 0.0020992020959965884,
      "selected_pairs": 27034.0,
      "step": 22920,
      "student_distance_mean": 23.31498622894287,
      "teacher_distance_mean": 0.8613941669464111,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0020992020959965884,
      "weighted_relation_loss": 0.0020992020959965884
    },
    {
      "action_query_batch_std": 0.3539259731769562,
      "action_query_norm": 40.907819747924805,
      "action_relation_loss": 0.002168287057429552,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.44995322731525,
      "flow_matching_loss": 0.017918859701603652,
      "grad_norm": 0.12192896008491516,
      "h_stock_valid_norm": 98.11310195922852,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.4434836189810335e-05,
      "loss": 0.0201,
      "positive_teacher_pairs": 26735.0,
      "query_h_norm_ratio": 0.4169480711221695,
      "relation_loss": 0.002168287057429552,
      "selected_pairs": 26735.0,
      "step": 22930,
      "student_distance_mean": 23.253621101379395,
      "teacher_distance_mean": 0.8580859303474426,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002168287057429552,
      "weighted_relation_loss": 0.002168287057429552
    },
    {
      "action_query_batch_std": 0.35356192886829374,
      "action_query_norm": 40.902819061279295,
      "action_relation_loss": 0.002212131896521896,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.45930776426567,
      "flow_matching_loss": 0.0187641903758049,
      "grad_norm": 0.08194923400878906,
      "h_stock_valid_norm": 97.8932373046875,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.4396117821293453e-05,
      "loss": 0.021,
      "positive_teacher_pairs": 26759.8,
      "query_h_norm_ratio": 0.41783430576324465,
      "relation_loss": 0.002212131896521896,
      "selected_pairs": 26759.8,
      "step": 22940,
      "student_distance_mean": 23.229980659484863,
      "teacher_distance_mean": 0.8574157238006592,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.002212131896521896,
      "weighted_relation_loss": 0.002212131896521896
    },
    {
      "action_query_batch_std": 0.35089996457099915,
      "action_query_norm": 40.89215927124023,
      "action_relation_loss": 0.002120430988725275,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.46866230121609,
      "flow_matching_loss": 0.018794584833085538,
      "grad_norm": 0.10181162506341934,
      "h_stock_valid_norm": 97.90436782836915,
      "invalid_tail_samples": 25.8,
      "learning_rate": 1.4357442714850633e-05,
      "loss": 0.0209,
      "positive_teacher_pairs": 26393.2,
      "query_h_norm_ratio": 0.41767561435699463,
      "relation_loss": 0.002120430988725275,
      "selected_pairs": 26393.2,
      "step": 22950,
      "student_distance_mean": 23.04529857635498,
      "teacher_distance_mean": 0.8535067737102509,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.002120430988725275,
      "weighted_relation_loss": 0.002120430988725275
    },
    {
      "action_query_batch_std": 0.3507977396249771,
      "action_query_norm": 40.903128433227536,
      "action_relation_loss": 0.0021786822588182988,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.47801683816651,
      "flow_matching_loss": 0.018224022909998895,
      "grad_norm": 0.08724480867385864,
      "h_stock_valid_norm": 97.90445556640626,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.43188109174758e-05,
      "loss": 0.0204,
      "positive_teacher_pairs": 27010.7,
      "query_h_norm_ratio": 0.41778793931007385,
      "relation_loss": 0.0021786822588182988,
      "selected_pairs": 27010.7,
      "step": 22960,
      "student_distance_mean": 23.168931198120116,
      "teacher_distance_mean": 0.8578958332538604,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0021786822588182988,
      "weighted_relation_loss": 0.0021786822588182988
    },
    {
      "action_query_batch_std": 0.3523415446281433,
      "action_query_norm": 40.898827743530276,
      "action_relation_loss": 0.0022133432561531664,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.487371375116933,
      "flow_matching_loss": 0.02077936930581927,
      "grad_norm": 0.12260377407073975,
      "h_stock_valid_norm": 97.87756576538087,
      "invalid_tail_samples": 25.9,
      "learning_rate": 1.4280222476110206e-05,
      "loss": 0.023,
      "positive_teacher_pairs": 26364.4,
      "query_h_norm_ratio": 0.41785991191864014,
      "relation_loss": 0.0022133432561531664,
      "selected_pairs": 26364.4,
      "step": 22970,
      "student_distance_mean": 23.218113517761232,
      "teacher_distance_mean": 0.8575498819351196,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.0022133432561531664,
      "weighted_relation_loss": 0.0022133432561531664
    },
    {
      "action_query_batch_std": 0.35104221999645235,
      "action_query_norm": 40.896803283691405,
      "action_relation_loss": 0.002148125623352826,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.49672591206735,
      "flow_matching_loss": 0.018021642789244653,
      "grad_norm": 0.09153969585895538,
      "h_stock_valid_norm": 98.09562606811524,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.4241677437642469e-05,
      "loss": 0.0202,
      "positive_teacher_pairs": 26737.7,
      "query_h_norm_ratio": 0.41691436171531676,
      "relation_loss": 0.002148125623352826,
      "selected_pairs": 26737.7,
      "step": 22980,
      "student_distance_mean": 23.139046096801756,
      "teacher_distance_mean": 0.8570938885211945,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002148125623352826,
      "weighted_relation_loss": 0.002148125623352826
    },
    {
      "action_query_batch_std": 0.3501764565706253,
      "action_query_norm": 40.889212417602536,
      "action_relation_loss": 0.0022166750626638533,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.506080449017773,
      "flow_matching_loss": 0.01764487735927105,
      "grad_norm": 0.08970385044813156,
      "h_stock_valid_norm": 97.99965286254883,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.420317584890844e-05,
      "loss": 0.0199,
      "positive_teacher_pairs": 26805.0,
      "query_h_norm_ratio": 0.4172394037246704,
      "relation_loss": 0.0022166750626638533,
      "selected_pairs": 26805.0,
      "step": 22990,
      "student_distance_mean": 23.03099308013916,
      "teacher_distance_mean": 0.8536903560161591,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0022166750626638533,
      "weighted_relation_loss": 0.0022166750626638533
    },
    {
      "action_query_batch_std": 0.34744271636009216,
      "action_query_norm": 40.8553825378418,
      "action_relation_loss": 0.00217963510658592,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.515434985968195,
      "flow_matching_loss": 0.018177211098372937,
      "grad_norm": 0.10532987117767334,
      "h_stock_valid_norm": 97.8814826965332,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.4164717756691176e-05,
      "loss": 0.0204,
      "positive_teacher_pairs": 27086.5,
      "query_h_norm_ratio": 0.41739759743213656,
      "relation_loss": 0.00217963510658592,
      "selected_pairs": 27086.5,
      "step": 23000,
      "student_distance_mean": 22.8771764755249,
      "teacher_distance_mean": 0.847283661365509,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.00217963510658592,
      "weighted_relation_loss": 0.00217963510658592
    },
    {
      "action_query_batch_std": 0.35396081805229185,
      "action_query_norm": 40.91129112243652,
      "action_relation_loss": 0.0021050618845038114,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.524789522918617,
      "flow_matching_loss": 0.01988883838057518,
      "grad_norm": 0.09990138560533524,
      "h_stock_valid_norm": 97.98846054077148,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.4126303207720882e-05,
      "loss": 0.022,
      "positive_teacher_pairs": 27081.2,
      "query_h_norm_ratio": 0.4175130039453506,
      "relation_loss": 0.0021050618845038114,
      "selected_pairs": 27081.2,
      "step": 23010,
      "student_distance_mean": 23.26800193786621,
      "teacher_distance_mean": 0.8617390334606171,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0021050618845038114,
      "weighted_relation_loss": 0.0021050618845038114
    },
    {
      "action_query_batch_std": 0.35003008544445036,
      "action_query_norm": 40.885487747192386,
      "action_relation_loss": 0.002161786938086152,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.534144059869035,
      "flow_matching_loss": 0.01943021295592189,
      "grad_norm": 0.10527092218399048,
      "h_stock_valid_norm": 97.77840194702148,
      "invalid_tail_samples": 24.4,
      "learning_rate": 1.408793224867484e-05,
      "loss": 0.0216,
      "positive_teacher_pairs": 26716.1,
      "query_h_norm_ratio": 0.4181497871875763,
      "relation_loss": 0.002161786938086152,
      "selected_pairs": 26716.1,
      "step": 23020,
      "student_distance_mean": 23.034885978698732,
      "teacher_distance_mean": 0.8543200135231018,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.002161786938086152,
      "weighted_relation_loss": 0.002161786938086152
    },
    {
      "action_query_batch_std": 0.3506717950105667,
      "action_query_norm": 40.887638854980466,
      "action_relation_loss": 0.0021961990278214216,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.543498596819457,
      "flow_matching_loss": 0.019520097598433494,
      "grad_norm": 0.11279764026403427,
      "h_stock_valid_norm": 97.73698501586914,
      "invalid_tail_samples": 24.9,
      "learning_rate": 1.4049604926177423e-05,
      "loss": 0.0217,
      "positive_teacher_pairs": 26599.0,
      "query_h_norm_ratio": 0.4183459997177124,
      "relation_loss": 0.0021961990278214216,
      "selected_pairs": 26599.0,
      "step": 23030,
      "student_distance_mean": 23.037625885009767,
      "teacher_distance_mean": 0.8548133671283722,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0021961990278214216,
      "weighted_relation_loss": 0.0021961990278214216
    },
    {
      "action_query_batch_std": 0.3495196431875229,
      "action_query_norm": 40.88341865539551,
      "action_relation_loss": 0.0022237901110202076,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.55285313376988,
      "flow_matching_loss": 0.01960750427097082,
      "grad_norm": 0.09889258444309235,
      "h_stock_valid_norm": 97.84570007324218,
      "invalid_tail_samples": 22.5,
      "learning_rate": 1.4011321286799916e-05,
      "loss": 0.0218,
      "positive_teacher_pairs": 27154.9,
      "query_h_norm_ratio": 0.41783835291862487,
      "relation_loss": 0.0022237901110202076,
      "selected_pairs": 27154.9,
      "step": 23040,
      "student_distance_mean": 22.925423049926756,
      "teacher_distance_mean": 0.8492319762706757,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0022237901110202076,
      "weighted_relation_loss": 0.0022237901110202076
    },
    {
      "action_query_batch_std": 0.3494176954030991,
      "action_query_norm": 40.87905921936035,
      "action_relation_loss": 0.00220481741707772,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.562207670720298,
      "flow_matching_loss": 0.016897210106253623,
      "grad_norm": 0.10214468091726303,
      "h_stock_valid_norm": 98.12181549072265,
      "invalid_tail_samples": 26.0,
      "learning_rate": 1.3973081377060565e-05,
      "loss": 0.0191,
      "positive_teacher_pairs": 26348.0,
      "query_h_norm_ratio": 0.4166155606508255,
      "relation_loss": 0.00220481741707772,
      "selected_pairs": 26348.0,
      "step": 23050,
      "student_distance_mean": 23.051347923278808,
      "teacher_distance_mean": 0.8540005505084991,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.00220481741707772,
      "weighted_relation_loss": 0.00220481741707772
    },
    {
      "action_query_batch_std": 0.34955317080020903,
      "action_query_norm": 40.88945960998535,
      "action_relation_loss": 0.0021925556473433973,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.57156220767072,
      "flow_matching_loss": 0.020281597319990397,
      "grad_norm": 0.09566525369882584,
      "h_stock_valid_norm": 97.9874282836914,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.393488524342445e-05,
      "loss": 0.0225,
      "positive_teacher_pairs": 27112.2,
      "query_h_norm_ratio": 0.4172962874174118,
      "relation_loss": 0.0021925556473433973,
      "selected_pairs": 27112.2,
      "step": 23060,
      "student_distance_mean": 23.014820861816407,
      "teacher_distance_mean": 0.8522536396980286,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0021925556473433973,
      "weighted_relation_loss": 0.0021925556473433973
    },
    {
      "action_query_batch_std": 0.349824795126915,
      "action_query_norm": 40.87125091552734,
      "action_relation_loss": 0.002104581333696842,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.580916744621142,
      "flow_matching_loss": 0.020589852519333363,
      "grad_norm": 0.14249886572360992,
      "h_stock_valid_norm": 98.04321060180663,
      "invalid_tail_samples": 26.5,
      "learning_rate": 1.3896732932303486e-05,
      "loss": 0.0227,
      "positive_teacher_pairs": 26231.7,
      "query_h_norm_ratio": 0.41687228679656985,
      "relation_loss": 0.002104581333696842,
      "selected_pairs": 26231.7,
      "step": 23070,
      "student_distance_mean": 23.154132652282716,
      "teacher_distance_mean": 0.8577911078929901,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.002104581333696842,
      "weighted_relation_loss": 0.002104581333696842
    },
    {
      "action_query_batch_std": 0.3503269016742706,
      "action_query_norm": 40.86998405456543,
      "action_relation_loss": 0.0021774732624180613,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.590271281571564,
      "flow_matching_loss": 0.01872373130172491,
      "grad_norm": 0.09604749828577042,
      "h_stock_valid_norm": 97.94187545776367,
      "invalid_tail_samples": 22.9,
      "learning_rate": 1.3858624490056304e-05,
      "loss": 0.0209,
      "positive_teacher_pairs": 27064.2,
      "query_h_norm_ratio": 0.41728905141353606,
      "relation_loss": 0.0021774732624180613,
      "selected_pairs": 27064.2,
      "step": 23080,
      "student_distance_mean": 23.025790786743165,
      "teacher_distance_mean": 0.8555888950824737,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0021774732624180613,
      "weighted_relation_loss": 0.0021774732624180613
    },
    {
      "action_query_batch_std": 0.3525254726409912,
      "action_query_norm": 40.88857727050781,
      "action_relation_loss": 0.002101389970630407,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.599625818521982,
      "flow_matching_loss": 0.01746738739311695,
      "grad_norm": 0.09945331513881683,
      "h_stock_valid_norm": 97.83227081298828,
      "invalid_tail_samples": 23.5,
      "learning_rate": 1.3820559962988299e-05,
      "loss": 0.0196,
      "positive_teacher_pairs": 26916.9,
      "query_h_norm_ratio": 0.41794741451740264,
      "relation_loss": 0.002101389970630407,
      "selected_pairs": 26916.9,
      "step": 23090,
      "student_distance_mean": 23.165944480895995,
      "teacher_distance_mean": 0.8593064248561859,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.002101389970630407,
      "weighted_relation_loss": 0.002101389970630407
    },
    {
      "action_query_batch_std": 0.3486390024423599,
      "action_query_norm": 40.84654350280762,
      "action_relation_loss": 0.002329082367941737,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.608980355472404,
      "flow_matching_loss": 0.018892497196793555,
      "grad_norm": 0.1337936520576477,
      "h_stock_valid_norm": 97.94718704223632,
      "invalid_tail_samples": 24.8,
      "learning_rate": 1.3782539397351418e-05,
      "loss": 0.0212,
      "positive_teacher_pairs": 26618.8,
      "query_h_norm_ratio": 0.4170305341482162,
      "relation_loss": 0.002329082367941737,
      "selected_pairs": 26618.8,
      "step": 23100,
      "student_distance_mean": 22.963363456726075,
      "teacher_distance_mean": 0.8528796970844269,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.002329082367941737,
      "weighted_relation_loss": 0.002329082367941737
    },
    {
      "action_query_batch_std": 0.3509960174560547,
      "action_query_norm": 40.883404922485354,
      "action_relation_loss": 0.002136921836063266,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.618334892422826,
      "flow_matching_loss": 0.019923126325011253,
      "grad_norm": 0.0940355584025383,
      "h_stock_valid_norm": 97.7161018371582,
      "invalid_tail_samples": 21.8,
      "learning_rate": 1.3744562839344266e-05,
      "loss": 0.0221,
      "positive_teacher_pairs": 27317.3,
      "query_h_norm_ratio": 0.41839563846588135,
      "relation_loss": 0.002136921836063266,
      "selected_pairs": 27317.3,
      "step": 23110,
      "student_distance_mean": 23.081598854064943,
      "teacher_distance_mean": 0.8571161389350891,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.002136921836063266,
      "weighted_relation_loss": 0.002136921836063266
    },
    {
      "action_query_batch_std": 0.34915315806865693,
      "action_query_norm": 40.8625789642334,
      "action_relation_loss": 0.0022382883937098084,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.627689429373245,
      "flow_matching_loss": 0.018692814279347658,
      "grad_norm": 0.17483709752559662,
      "h_stock_valid_norm": 97.87484664916992,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.370663033511193e-05,
      "loss": 0.0209,
      "positive_teacher_pairs": 26872.3,
      "query_h_norm_ratio": 0.4175028294324875,
      "relation_loss": 0.0022382883937098084,
      "selected_pairs": 26872.3,
      "step": 23120,
      "student_distance_mean": 22.936262130737305,
      "teacher_distance_mean": 0.8523586332798004,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0022382883937098084,
      "weighted_relation_loss": 0.0022382883937098084
    },
    {
      "action_query_batch_std": 0.3512852698564529,
      "action_query_norm": 40.85760765075683,
      "action_relation_loss": 0.0021481756353750824,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.637043966323667,
      "flow_matching_loss": 0.018985367752611636,
      "grad_norm": 0.12445438653230667,
      "h_stock_valid_norm": 97.84795684814453,
      "invalid_tail_samples": 23.2,
      "learning_rate": 1.3668741930745964e-05,
      "loss": 0.0211,
      "positive_teacher_pairs": 26998.6,
      "query_h_norm_ratio": 0.41756452322006227,
      "relation_loss": 0.0021481756353750824,
      "selected_pairs": 26998.6,
      "step": 23130,
      "student_distance_mean": 23.073838233947754,
      "teacher_distance_mean": 0.8554975330829621,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0021481756353750824,
      "weighted_relation_loss": 0.0021481756353750824
    },
    {
      "action_query_batch_std": 0.3504486113786697,
      "action_query_norm": 40.87649307250977,
      "action_relation_loss": 0.002149263839237392,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.64639850327409,
      "flow_matching_loss": 0.01755537400022149,
      "grad_norm": 0.09201261401176453,
      "h_stock_valid_norm": 97.64300918579102,
      "invalid_tail_samples": 23.5,
      "learning_rate": 1.3630897672284382e-05,
      "loss": 0.0197,
      "positive_teacher_pairs": 26917.9,
      "query_h_norm_ratio": 0.4186328053474426,
      "relation_loss": 0.002149263839237392,
      "selected_pairs": 26917.9,
      "step": 23140,
      "student_distance_mean": 23.03529644012451,
      "teacher_distance_mean": 0.8544784009456634,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.002149263839237392,
      "weighted_relation_loss": 0.002149263839237392
    },
    {
      "action_query_batch_std": 0.34834648966789244,
      "action_query_norm": 40.848776245117186,
      "action_relation_loss": 0.0021635104902088644,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.65575304022451,
      "flow_matching_loss": 0.018220537807792425,
      "grad_norm": 0.08738987147808075,
      "h_stock_valid_norm": 97.66350173950195,
      "invalid_tail_samples": 26.9,
      "learning_rate": 1.3593097605711508e-05,
      "loss": 0.0204,
      "positive_teacher_pairs": 26144.3,
      "query_h_norm_ratio": 0.41826215386390686,
      "relation_loss": 0.0021635104902088644,
      "selected_pairs": 26144.3,
      "step": 23150,
      "student_distance_mean": 22.956068229675292,
      "teacher_distance_mean": 0.8533890604972839,
      "valid_action_samples": 229.1,
      "weighted_auxiliary_loss": 0.0021635104902088644,
      "weighted_relation_loss": 0.0021635104902088644
    },
    {
      "action_query_batch_std": 0.3480793833732605,
      "action_query_norm": 40.85136947631836,
      "action_relation_loss": 0.002361361775547266,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.66510757717493,
      "flow_matching_loss": 0.020768039487302305,
      "grad_norm": 0.11209594458341599,
      "h_stock_valid_norm": 97.72668838500977,
      "invalid_tail_samples": 23.3,
      "learning_rate": 1.3555341776957992e-05,
      "loss": 0.0231,
      "positive_teacher_pairs": 26964.3,
      "query_h_norm_ratio": 0.4180198162794113,
      "relation_loss": 0.002361361775547266,
      "selected_pairs": 26964.3,
      "step": 23160,
      "student_distance_mean": 22.851109313964844,
      "teacher_distance_mean": 0.8463043570518494,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.002361361775547266,
      "weighted_relation_loss": 0.002361361775547266
    },
    {
      "action_query_batch_std": 0.351250496506691,
      "action_query_norm": 40.867314147949216,
      "action_relation_loss": 0.002198648639023304,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.67446211412535,
      "flow_matching_loss": 0.01781523898243904,
      "grad_norm": 0.08798828721046448,
      "h_stock_valid_norm": 97.73321685791015,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.3517630231900724e-05,
      "loss": 0.02,
      "positive_teacher_pairs": 26755.4,
      "query_h_norm_ratio": 0.41815229058265685,
      "relation_loss": 0.002198648639023304,
      "selected_pairs": 26755.4,
      "step": 23170,
      "student_distance_mean": 23.100949096679688,
      "teacher_distance_mean": 0.8562601506710052,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.002198648639023304,
      "weighted_relation_loss": 0.002198648639023304
    },
    {
      "action_query_batch_std": 0.3538341671228409,
      "action_query_norm": 40.89773674011231,
      "action_relation_loss": 0.00210999611299485,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.683816651075773,
      "flow_matching_loss": 0.017706008162349463,
      "grad_norm": 0.09504697471857071,
      "h_stock_valid_norm": 97.96786270141601,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.3479963016362767e-05,
      "loss": 0.0198,
      "positive_teacher_pairs": 26801.5,
      "query_h_norm_ratio": 0.41746210753917695,
      "relation_loss": 0.00210999611299485,
      "selected_pairs": 26801.5,
      "step": 23180,
      "student_distance_mean": 23.270135116577148,
      "teacher_distance_mean": 0.8620006740093231,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.00210999611299485,
      "weighted_relation_loss": 0.00210999611299485
    },
    {
      "action_query_batch_std": 0.35049317479133607,
      "action_query_norm": 40.86059036254883,
      "action_relation_loss": 0.002316607069224119,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.69317118802619,
      "flow_matching_loss": 0.019070213567465542,
      "grad_norm": 0.0920209214091301,
      "h_stock_valid_norm": 97.6926513671875,
      "invalid_tail_samples": 22.6,
      "learning_rate": 1.3442340176113377e-05,
      "loss": 0.0214,
      "positive_teacher_pairs": 27126.0,
      "query_h_norm_ratio": 0.41825818419456484,
      "relation_loss": 0.002316607069224119,
      "selected_pairs": 27126.0,
      "step": 23190,
      "student_distance_mean": 23.049904441833498,
      "teacher_distance_mean": 0.8556494593620301,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.002316607069224119,
      "weighted_relation_loss": 0.002316607069224119
    },
    {
      "action_query_batch_std": 0.3476662069559097,
      "action_query_norm": 40.84772720336914,
      "action_relation_loss": 0.0022534484043717383,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.702525724976613,
      "flow_matching_loss": 0.01970300544053316,
      "grad_norm": 0.08945652842521667,
      "h_stock_valid_norm": 97.58172073364258,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.340476175686784e-05,
      "loss": 0.022,
      "positive_teacher_pairs": 27014.2,
      "query_h_norm_ratio": 0.418604364991188,
      "relation_loss": 0.0022534484043717383,
      "selected_pairs": 27014.2,
      "step": 23200,
      "student_distance_mean": 22.84386920928955,
      "teacher_distance_mean": 0.8509067952632904,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0022534484043717383,
      "weighted_relation_loss": 0.0022534484043717383
    },
    {
      "action_query_batch_std": 0.34931340515613557,
      "action_query_norm": 40.83986892700195,
      "action_relation_loss": 0.0022185014095157385,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.711880261927035,
      "flow_matching_loss": 0.019179464876651765,
      "grad_norm": 0.07999581843614578,
      "h_stock_valid_norm": 97.50111694335938,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.3367227804287469e-05,
      "loss": 0.0214,
      "positive_teacher_pairs": 26758.7,
      "query_h_norm_ratio": 0.418867164850235,
      "relation_loss": 0.0022185014095157385,
      "selected_pairs": 26758.7,
      "step": 23210,
      "student_distance_mean": 23.037190246582032,
      "teacher_distance_mean": 0.8586019515991211,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0022185014095157385,
      "weighted_relation_loss": 0.0022185014095157385
    },
    {
      "action_query_batch_std": 0.35467382669448855,
      "action_query_norm": 40.92348098754883,
      "action_relation_loss": 0.0021312486031092705,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.721234798877454,
      "flow_matching_loss": 0.01831276835873723,
      "grad_norm": 0.08317180722951889,
      "h_stock_valid_norm": 97.86179275512696,
      "invalid_tail_samples": 22.3,
      "learning_rate": 1.332973836397956e-05,
      "loss": 0.0204,
      "positive_teacher_pairs": 27199.2,
      "query_h_norm_ratio": 0.41817925572395326,
      "relation_loss": 0.0021312486031092705,
      "selected_pairs": 27199.2,
      "step": 23220,
      "student_distance_mean": 23.28612937927246,
      "teacher_distance_mean": 0.8661631107330322,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0021312486031092705,
      "weighted_relation_loss": 0.0021312486031092705
    },
    {
      "action_query_batch_std": 0.3486302852630615,
      "action_query_norm": 40.86497192382812,
      "action_relation_loss": 0.0021966884727589787,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.730589335827876,
      "flow_matching_loss": 0.01900133416056633,
      "grad_norm": 0.09736490249633789,
      "h_stock_valid_norm": 97.64623794555663,
      "invalid_tail_samples": 22.1,
      "learning_rate": 1.329229348149731e-05,
      "loss": 0.0212,
      "positive_teacher_pairs": 27246.6,
      "query_h_norm_ratio": 0.4185032159090042,
      "relation_loss": 0.0021966884727589787,
      "selected_pairs": 27246.6,
      "step": 23230,
      "student_distance_mean": 22.956842994689943,
      "teacher_distance_mean": 0.8508980929851532,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0021966884727589787,
      "weighted_relation_loss": 0.0021966884727589787
    },
    {
      "action_query_batch_std": 0.34883267879486085,
      "action_query_norm": 40.853098678588864,
      "action_relation_loss": 0.0022570531349629165,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.739943872778298,
      "flow_matching_loss": 0.018345562182366847,
      "grad_norm": 0.10256943106651306,
      "h_stock_valid_norm": 97.39312133789062,
      "invalid_tail_samples": 22.9,
      "learning_rate": 1.3254893202339796e-05,
      "loss": 0.0206,
      "positive_teacher_pairs": 27058.4,
      "query_h_norm_ratio": 0.41947194635868074,
      "relation_loss": 0.0022570531349629165,
      "selected_pairs": 27058.4,
      "step": 23240,
      "student_distance_mean": 22.989525032043456,
      "teacher_distance_mean": 0.8535291910171509,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0022570531349629165,
      "weighted_relation_loss": 0.0022570531349629165
    },
    {
      "action_query_batch_std": 0.34744167029857637,
      "action_query_norm": 40.841617584228516,
      "action_relation_loss": 0.0021962311118841173,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.74929840972872,
      "flow_matching_loss": 0.019906277395784856,
      "grad_norm": 0.12863558530807495,
      "h_stock_valid_norm": 97.55293121337891,
      "invalid_tail_samples": 24.8,
      "learning_rate": 1.3217537571951872e-05,
      "loss": 0.0221,
      "positive_teacher_pairs": 26618.5,
      "query_h_norm_ratio": 0.4186625599861145,
      "relation_loss": 0.0021962311118841173,
      "selected_pairs": 26618.5,
      "step": 23250,
      "student_distance_mean": 22.86318359375,
      "teacher_distance_mean": 0.8483159661293029,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0021962311118841173,
      "weighted_relation_loss": 0.0021962311118841173
    },
    {
      "action_query_batch_std": 0.35212617218494413,
      "action_query_norm": 40.889739990234375,
      "action_relation_loss": 0.002109903935343027,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.75865294667914,
      "flow_matching_loss": 0.01769296173006296,
      "grad_norm": 0.11628998070955276,
      "h_stock_valid_norm": 97.7650032043457,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.3180226635724169e-05,
      "loss": 0.0198,
      "positive_teacher_pairs": 26738.3,
      "query_h_norm_ratio": 0.41824604868888854,
      "relation_loss": 0.002109903935343027,
      "selected_pairs": 26738.3,
      "step": 23260,
      "student_distance_mean": 23.11638240814209,
      "teacher_distance_mean": 0.8605891466140747,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002109903935343027,
      "weighted_relation_loss": 0.002109903935343027
    },
    {
      "action_query_batch_std": 0.3528038144111633,
      "action_query_norm": 40.89388580322266,
      "action_relation_loss": 0.002090930484700948,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.76800748362956,
      "flow_matching_loss": 0.018321341648697854,
      "grad_norm": 0.0763324499130249,
      "h_stock_valid_norm": 98.00179061889648,
      "invalid_tail_samples": 21.7,
      "learning_rate": 1.314296043899298e-05,
      "loss": 0.0204,
      "positive_teacher_pairs": 27341.7,
      "query_h_norm_ratio": 0.4172857582569122,
      "relation_loss": 0.002090930484700948,
      "selected_pairs": 27341.7,
      "step": 23270,
      "student_distance_mean": 23.158841133117676,
      "teacher_distance_mean": 0.8613059461116791,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.002090930484700948,
      "weighted_relation_loss": 0.002090930484700948
    },
    {
      "action_query_batch_std": 0.3521402269601822,
      "action_query_norm": 40.887291717529294,
      "action_relation_loss": 0.0021974863833747805,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.777362020579982,
      "flow_matching_loss": 0.018499866873025895,
      "grad_norm": 0.12625135481357574,
      "h_stock_valid_norm": 97.97281799316406,
      "invalid_tail_samples": 23.2,
      "learning_rate": 1.3105739027040248e-05,
      "loss": 0.0207,
      "positive_teacher_pairs": 26989.5,
      "query_h_norm_ratio": 0.417336693406105,
      "relation_loss": 0.0021974863833747805,
      "selected_pairs": 26989.5,
      "step": 23280,
      "student_distance_mean": 23.101320266723633,
      "teacher_distance_mean": 0.8602680206298828,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0021974863833747805,
      "weighted_relation_loss": 0.0021974863833747805
    },
    {
      "action_query_batch_std": 0.34905995428562164,
      "action_query_norm": 40.850547790527344,
      "action_relation_loss": 0.002107081178110093,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.7867165575304,
      "flow_matching_loss": 0.018507501389831305,
      "grad_norm": 0.07752898335456848,
      "h_stock_valid_norm": 97.65296096801758,
      "invalid_tail_samples": 25.4,
      "learning_rate": 1.3068562445093491e-05,
      "loss": 0.0206,
      "positive_teacher_pairs": 26484.2,
      "query_h_norm_ratio": 0.4183258295059204,
      "relation_loss": 0.002107081178110093,
      "selected_pairs": 26484.2,
      "step": 23290,
      "student_distance_mean": 23.0070592880249,
      "teacher_distance_mean": 0.8554937720298768,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.002107081178110093,
      "weighted_relation_loss": 0.002107081178110093
    },
    {
      "action_query_batch_std": 0.3549045234918594,
      "action_query_norm": 40.90382270812988,
      "action_relation_loss": 0.002147181355394423,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.796071094480823,
      "flow_matching_loss": 0.01954119438305497,
      "grad_norm": 0.08109335601329803,
      "h_stock_valid_norm": 97.72986907958985,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.3031430738325778e-05,
      "loss": 0.0217,
      "positive_teacher_pairs": 26737.0,
      "query_h_norm_ratio": 0.4185417205095291,
      "relation_loss": 0.002147181355394423,
      "selected_pairs": 26737.0,
      "step": 23300,
      "student_distance_mean": 23.369583320617675,
      "teacher_distance_mean": 0.8681683242321014,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002147181355394423,
      "weighted_relation_loss": 0.002147181355394423
    },
    {
      "action_query_batch_std": 0.34890898764133454,
      "action_query_norm": 40.86053047180176,
      "action_relation_loss": 0.002242671023122966,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.805425631431245,
      "flow_matching_loss": 0.018921780027449132,
      "grad_norm": 0.09445303678512573,
      "h_stock_valid_norm": 97.36440811157226,
      "invalid_tail_samples": 25.9,
      "learning_rate": 1.299434395185563e-05,
      "loss": 0.0212,
      "positive_teacher_pairs": 26374.4,
      "query_h_norm_ratio": 0.419667637348175,
      "relation_loss": 0.002242671023122966,
      "selected_pairs": 26374.4,
      "step": 23310,
      "student_distance_mean": 23.04708023071289,
      "teacher_distance_mean": 0.8547684967517852,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.002242671023122966,
      "weighted_relation_loss": 0.002242671023122966
    },
    {
      "action_query_batch_std": 0.34798086285591123,
      "action_query_norm": 40.85361785888672,
      "action_relation_loss": 0.0022517408709973098,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.814780168381667,
      "flow_matching_loss": 0.01736700991168618,
      "grad_norm": 0.13987086713314056,
      "h_stock_valid_norm": 97.6569694519043,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.295730213074699e-05,
      "loss": 0.0196,
      "positive_teacher_pairs": 26757.3,
      "query_h_norm_ratio": 0.4183399647474289,
      "relation_loss": 0.0022517408709973098,
      "selected_pairs": 26757.3,
      "step": 23320,
      "student_distance_mean": 22.905315017700197,
      "teacher_distance_mean": 0.8511657953262329,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0022517408709973098,
      "weighted_relation_loss": 0.0022517408709973098
    },
    {
      "action_query_batch_std": 0.3531468540430069,
      "action_query_norm": 40.8733211517334,
      "action_relation_loss": 0.0021323674125596883,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.824134705332085,
      "flow_matching_loss": 0.018242873810231687,
      "grad_norm": 0.09627649933099747,
      "h_stock_valid_norm": 97.71111526489258,
      "invalid_tail_samples": 25.0,
      "learning_rate": 1.2920305320009152e-05,
      "loss": 0.0204,
      "positive_teacher_pairs": 26567.8,
      "query_h_norm_ratio": 0.4183108687400818,
      "relation_loss": 0.0021323674125596883,
      "selected_pairs": 26567.8,
      "step": 23330,
      "student_distance_mean": 23.196446228027344,
      "teacher_distance_mean": 0.8628508269786834,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0021323674125596883,
      "weighted_relation_loss": 0.0021323674125596883
    },
    {
      "action_query_batch_std": 0.3474243015050888,
      "action_query_norm": 40.84400978088379,
      "action_relation_loss": 0.002261191071011126,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.833489242282507,
      "flow_matching_loss": 0.018420317582786082,
      "grad_norm": 0.09037171304225922,
      "h_stock_valid_norm": 97.68903427124023,
      "invalid_tail_samples": 23.5,
      "learning_rate": 1.2883353564596729e-05,
      "loss": 0.0207,
      "positive_teacher_pairs": 26917.6,
      "query_h_norm_ratio": 0.41810397803783417,
      "relation_loss": 0.002261191071011126,
      "selected_pairs": 26917.6,
      "step": 23340,
      "student_distance_mean": 22.8634090423584,
      "teacher_distance_mean": 0.8484084844589234,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.002261191071011126,
      "weighted_relation_loss": 0.002261191071011126
    },
    {
      "action_query_batch_std": 0.347142431139946,
      "action_query_norm": 40.83272438049316,
      "action_relation_loss": 0.0021275987848639486,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.84284377923293,
      "flow_matching_loss": 0.019460823759436606,
      "grad_norm": 0.09289294481277466,
      "h_stock_valid_norm": 97.65790328979492,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.2846446909409604e-05,
      "loss": 0.0216,
      "positive_teacher_pairs": 26803.6,
      "query_h_norm_ratio": 0.41812337636947633,
      "relation_loss": 0.0021275987848639486,
      "selected_pairs": 26803.6,
      "step": 23350,
      "student_distance_mean": 22.904920387268067,
      "teacher_distance_mean": 0.8523203790187835,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0021275987848639486,
      "weighted_relation_loss": 0.0021275987848639486
    },
    {
      "action_query_batch_std": 0.35249671041965486,
      "action_query_norm": 40.86802520751953,
      "action_relation_loss": 0.0021040193736553193,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.852198316183348,
      "flow_matching_loss": 0.01857465049251914,
      "grad_norm": 0.1033901646733284,
      "h_stock_valid_norm": 97.58963165283203,
      "invalid_tail_samples": 24.8,
      "learning_rate": 1.280958539929284e-05,
      "loss": 0.0207,
      "positive_teacher_pairs": 26615.1,
      "query_h_norm_ratio": 0.4187764197587967,
      "relation_loss": 0.0021040193736553193,
      "selected_pairs": 26615.1,
      "step": 23360,
      "student_distance_mean": 23.289059257507326,
      "teacher_distance_mean": 0.8622651517391204,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0021040193736553193,
      "weighted_relation_loss": 0.0021040193736553193
    },
    {
      "action_query_batch_std": 0.35358574986457825,
      "action_query_norm": 40.86834182739258,
      "action_relation_loss": 0.0021265996154397728,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.86155285313377,
      "flow_matching_loss": 0.017366335727274418,
      "grad_norm": 0.0925690308213234,
      "h_stock_valid_norm": 97.88720169067383,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.277276907903664e-05,
      "loss": 0.0195,
      "positive_teacher_pairs": 26872.7,
      "query_h_norm_ratio": 0.41750648319721223,
      "relation_loss": 0.0021265996154397728,
      "selected_pairs": 26872.7,
      "step": 23370,
      "student_distance_mean": 23.2574520111084,
      "teacher_distance_mean": 0.8617159605026246,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0021265996154397728,
      "weighted_relation_loss": 0.0021265996154397728
    },
    {
      "action_query_batch_std": 0.34679156839847564,
      "action_query_norm": 40.828453826904294,
      "action_relation_loss": 0.0021275726146996023,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.87090739008419,
      "flow_matching_loss": 0.01977771520614624,
      "grad_norm": 0.07143034040927887,
      "h_stock_valid_norm": 97.88933181762695,
      "invalid_tail_samples": 25.2,
      "learning_rate": 1.2735997993376298e-05,
      "loss": 0.0219,
      "positive_teacher_pairs": 26530.6,
      "query_h_norm_ratio": 0.41709230840206146,
      "relation_loss": 0.0021275726146996023,
      "selected_pairs": 26530.6,
      "step": 23380,
      "student_distance_mean": 22.810803604125976,
      "teacher_distance_mean": 0.8470286905765534,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0021275726146996023,
      "weighted_relation_loss": 0.0021275726146996023
    },
    {
      "action_query_batch_std": 0.3513321757316589,
      "action_query_norm": 40.85137672424317,
      "action_relation_loss": 0.0021435156697407363,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.880261927034613,
      "flow_matching_loss": 0.01729458700865507,
      "grad_norm": 0.08204533159732819,
      "h_stock_valid_norm": 97.90290069580078,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.2699272186992166e-05,
      "loss": 0.0194,
      "positive_teacher_pairs": 26803.8,
      "query_h_norm_ratio": 0.4172664791345596,
      "relation_loss": 0.0021435156697407363,
      "selected_pairs": 26803.8,
      "step": 23390,
      "student_distance_mean": 23.140047454833983,
      "teacher_distance_mean": 0.8611633896827697,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0021435156697407363,
      "weighted_relation_loss": 0.0021435156697407363
    },
    {
      "action_query_batch_std": 0.35005209147930144,
      "action_query_norm": 40.853820419311525,
      "action_relation_loss": 0.0022213489864952862,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.889616463985032,
      "flow_matching_loss": 0.016761256940662862,
      "grad_norm": 0.09611459821462631,
      "h_stock_valid_norm": 97.87066268920898,
      "invalid_tail_samples": 25.2,
      "learning_rate": 1.2662591704509547e-05,
      "loss": 0.019,
      "positive_teacher_pairs": 26523.3,
      "query_h_norm_ratio": 0.4174271017313004,
      "relation_loss": 0.0022213489864952862,
      "selected_pairs": 26523.3,
      "step": 23400,
      "student_distance_mean": 23.122623443603516,
      "teacher_distance_mean": 0.8606972992420197,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0022213489864952862,
      "weighted_relation_loss": 0.0022213489864952862
    },
    {
      "action_query_batch_std": 0.3488642543554306,
      "action_query_norm": 40.84860687255859,
      "action_relation_loss": 0.002113549644127488,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.898971000935454,
      "flow_matching_loss": 0.01971525847911835,
      "grad_norm": 0.08096697926521301,
      "h_stock_valid_norm": 97.86753845214844,
      "invalid_tail_samples": 25.3,
      "learning_rate": 1.2625956590498711e-05,
      "loss": 0.0218,
      "positive_teacher_pairs": 26500.1,
      "query_h_norm_ratio": 0.41738922595977784,
      "relation_loss": 0.002113549644127488,
      "selected_pairs": 26500.1,
      "step": 23410,
      "student_distance_mean": 22.957968521118165,
      "teacher_distance_mean": 0.8516542434692382,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.002113549644127488,
      "weighted_relation_loss": 0.002113549644127488
    },
    {
      "action_query_batch_std": 0.35291163325309755,
      "action_query_norm": 40.86743659973145,
      "action_relation_loss": 0.0020805982290767132,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.908325537885876,
      "flow_matching_loss": 0.016964821983128785,
      "grad_norm": 0.11043213307857513,
      "h_stock_valid_norm": 97.95081481933593,
      "invalid_tail_samples": 23.4,
      "learning_rate": 1.2589366889474757e-05,
      "loss": 0.019,
      "positive_teacher_pairs": 26940.0,
      "query_h_norm_ratio": 0.4172269105911255,
      "relation_loss": 0.0020805982290767132,
      "selected_pairs": 26940.0,
      "step": 23420,
      "student_distance_mean": 23.215331840515137,
      "teacher_distance_mean": 0.8606866657733917,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0020805982290767132,
      "weighted_relation_loss": 0.0020805982290767132
    },
    {
      "action_query_batch_std": 0.35295838713645933,
      "action_query_norm": 40.886904907226565,
      "action_relation_loss": 0.0021905751316808166,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.917680074836294,
      "flow_matching_loss": 0.019893867336213587,
      "grad_norm": 0.21620045602321625,
      "h_stock_valid_norm": 97.9907356262207,
      "invalid_tail_samples": 22.4,
      "learning_rate": 1.2552822645897621e-05,
      "loss": 0.0221,
      "positive_teacher_pairs": 27174.7,
      "query_h_norm_ratio": 0.41725540459156035,
      "relation_loss": 0.0021905751316808166,
      "selected_pairs": 27174.7,
      "step": 23430,
      "student_distance_mean": 23.16678066253662,
      "teacher_distance_mean": 0.8617683351039886,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.0021905751316808166,
      "weighted_relation_loss": 0.0021905751316808166
    },
    {
      "action_query_batch_std": 0.3510841131210327,
      "action_query_norm": 40.8488037109375,
      "action_relation_loss": 0.002126326283905655,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.927034611786716,
      "flow_matching_loss": 0.018229990452528,
      "grad_norm": 0.08855490386486053,
      "h_stock_valid_norm": 97.97025299072266,
      "invalid_tail_samples": 21.8,
      "learning_rate": 1.2516323904172e-05,
      "loss": 0.0204,
      "positive_teacher_pairs": 27317.0,
      "query_h_norm_ratio": 0.4169537752866745,
      "relation_loss": 0.002126326283905655,
      "selected_pairs": 27317.0,
      "step": 23440,
      "student_distance_mean": 23.0488618850708,
      "teacher_distance_mean": 0.8581844627857208,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.002126326283905655,
      "weighted_relation_loss": 0.002126326283905655
    },
    {
      "action_query_batch_std": 0.3490076869726181,
      "action_query_norm": 40.82139930725098,
      "action_relation_loss": 0.0021556376479566096,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.93638914873714,
      "flow_matching_loss": 0.01891662711277604,
      "grad_norm": 0.11134834587574005,
      "h_stock_valid_norm": 98.0601303100586,
      "invalid_tail_samples": 22.1,
      "learning_rate": 1.2479870708647324e-05,
      "loss": 0.0211,
      "positive_teacher_pairs": 27244.7,
      "query_h_norm_ratio": 0.4162931263446808,
      "relation_loss": 0.0021556376479566096,
      "selected_pairs": 27244.7,
      "step": 23450,
      "student_distance_mean": 22.86545124053955,
      "teacher_distance_mean": 0.8488706409931183,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0021556376479566096,
      "weighted_relation_loss": 0.0021556376479566096
    },
    {
      "action_query_batch_std": 0.34624482691287994,
      "action_query_norm": 40.811064910888675,
      "action_relation_loss": 0.002221098146401346,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.94574368568756,
      "flow_matching_loss": 0.018714678660035133,
      "grad_norm": 0.09067696332931519,
      "h_stock_valid_norm": 98.02334899902344,
      "invalid_tail_samples": 23.0,
      "learning_rate": 1.2443463103617658e-05,
      "loss": 0.0209,
      "positive_teacher_pairs": 27040.2,
      "query_h_norm_ratio": 0.41634093821048734,
      "relation_loss": 0.002221098146401346,
      "selected_pairs": 27040.2,
      "step": 23460,
      "student_distance_mean": 22.796988487243652,
      "teacher_distance_mean": 0.8469535052776337,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.002221098146401346,
      "weighted_relation_loss": 0.002221098146401346
    },
    {
      "action_query_batch_std": 0.3502708196640015,
      "action_query_norm": 40.8501277923584,
      "action_relation_loss": 0.0021044312859885396,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.95509822263798,
      "flow_matching_loss": 0.01668113274499774,
      "grad_norm": 0.08773232251405716,
      "h_stock_valid_norm": 97.91693267822265,
      "invalid_tail_samples": 24.6,
      "learning_rate": 1.240710113332167e-05,
      "loss": 0.0188,
      "positive_teacher_pairs": 26668.8,
      "query_h_norm_ratio": 0.41719402074813844,
      "relation_loss": 0.0021044312859885396,
      "selected_pairs": 26668.8,
      "step": 23470,
      "student_distance_mean": 23.155811882019044,
      "teacher_distance_mean": 0.8579237222671509,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0021044312859885396,
      "weighted_relation_loss": 0.0021044312859885396
    },
    {
      "action_query_batch_std": 0.35168923139572145,
      "action_query_norm": 40.86979293823242,
      "action_relation_loss": 0.002208888262975961,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.9644527595884,
      "flow_matching_loss": 0.019055926240980625,
      "grad_norm": 0.12544003129005432,
      "h_stock_valid_norm": 97.98461456298828,
      "invalid_tail_samples": 22.4,
      "learning_rate": 1.2370784841942584e-05,
      "loss": 0.0213,
      "positive_teacher_pairs": 27175.5,
      "query_h_norm_ratio": 0.417107367515564,
      "relation_loss": 0.002208888262975961,
      "selected_pairs": 27175.5,
      "step": 23480,
      "student_distance_mean": 23.026560592651368,
      "teacher_distance_mean": 0.8549975335597992,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.002208888262975961,
      "weighted_relation_loss": 0.002208888262975961
    },
    {
      "action_query_batch_std": 0.35054963231086733,
      "action_query_norm": 40.840664291381835,
      "action_relation_loss": 0.002165077475365251,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.973807296538823,
      "flow_matching_loss": 0.016729397140443324,
      "grad_norm": 0.09174078702926636,
      "h_stock_valid_norm": 97.8623275756836,
      "invalid_tail_samples": 25.6,
      "learning_rate": 1.2334514273608117e-05,
      "loss": 0.0189,
      "positive_teacher_pairs": 26445.2,
      "query_h_norm_ratio": 0.41732960343360903,
      "relation_loss": 0.002165077475365251,
      "selected_pairs": 26445.2,
      "step": 23490,
      "student_distance_mean": 22.98121910095215,
      "teacher_distance_mean": 0.8546140730381012,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.002165077475365251,
      "weighted_relation_loss": 0.002165077475365251
    },
    {
      "action_query_batch_std": 0.34643651843070983,
      "action_query_norm": 40.81309928894043,
      "action_relation_loss": 0.0023318052291870115,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.98316183348924,
      "flow_matching_loss": 0.01841773958876729,
      "grad_norm": 0.09526622295379639,
      "h_stock_valid_norm": 97.75305709838867,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.2298289472390417e-05,
      "loss": 0.0207,
      "positive_teacher_pairs": 26881.2,
      "query_h_norm_ratio": 0.4175169825553894,
      "relation_loss": 0.0023318052291870115,
      "selected_pairs": 26881.2,
      "step": 23500,
      "student_distance_mean": 22.80832290649414,
      "teacher_distance_mean": 0.8471233963966369,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0023318052291870115,
      "weighted_relation_loss": 0.0023318052291870115
    },
    {
      "action_query_batch_std": 0.3489471793174744,
      "action_query_norm": 40.843936157226565,
      "action_relation_loss": 0.0022580217802897094,
      "auxiliary_loss_weight": 1.0,
      "epoch": 21.992516370439663,
      "flow_matching_loss": 0.018747478444129228,
      "grad_norm": 0.09469529241323471,
      "h_stock_valid_norm": 97.92053604125977,
      "invalid_tail_samples": 21.2,
      "learning_rate": 1.226211048230606e-05,
      "loss": 0.021,
      "positive_teacher_pairs": 27470.3,
      "query_h_norm_ratio": 0.41711496412754057,
      "relation_loss": 0.0022580217802897094,
      "selected_pairs": 27470.3,
      "step": 23510,
      "student_distance_mean": 22.875692939758302,
      "teacher_distance_mean": 0.8514962911605835,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.0022580217802897094,
      "weighted_relation_loss": 0.0022580217802897094
    },
    {
      "action_query_batch_std": 0.34816305339336395,
      "action_query_norm": 40.83118896484375,
      "action_relation_loss": 0.002241098484955728,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.001870907390085,
      "flow_matching_loss": 0.0174473924562335,
      "grad_norm": 0.08970163762569427,
      "h_stock_valid_norm": 97.82058639526367,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.2225977347315921e-05,
      "loss": 0.0197,
      "positive_teacher_pairs": 24058.7,
      "query_h_norm_ratio": 0.41740956008434293,
      "relation_loss": 0.002241098484955728,
      "selected_pairs": 24058.7,
      "step": 23520,
      "student_distance_mean": 22.794245529174805,
      "teacher_distance_mean": 0.8477723836898804,
      "valid_action_samples": 213.3,
      "weighted_auxiliary_loss": 0.002241098484955728,
      "weighted_relation_loss": 0.002241098484955728
    },
    {
      "action_query_batch_std": 0.3459670215845108,
      "action_query_norm": 40.802667236328126,
      "action_relation_loss": 0.0020568782347254455,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.011225444340504,
      "flow_matching_loss": 0.01854914063587785,
      "grad_norm": 0.12785950303077698,
      "h_stock_valid_norm": 97.7497459411621,
      "invalid_tail_samples": 28.0,
      "learning_rate": 1.2189890111325148e-05,
      "loss": 0.0206,
      "positive_teacher_pairs": 25891.5,
      "query_h_norm_ratio": 0.4174245119094849,
      "relation_loss": 0.0020568782347254455,
      "selected_pairs": 25891.5,
      "step": 23530,
      "student_distance_mean": 22.800192642211915,
      "teacher_distance_mean": 0.85164195895195,
      "valid_action_samples": 228.0,
      "weighted_auxiliary_loss": 0.0020568782347254455,
      "weighted_relation_loss": 0.0020568782347254455
    },
    {
      "action_query_batch_std": 0.3449630498886108,
      "action_query_norm": 40.81236953735352,
      "action_relation_loss": 0.002061582857277244,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.020579981290926,
      "flow_matching_loss": 0.0190609241835773,
      "grad_norm": 0.1273183673620224,
      "h_stock_valid_norm": 97.86057052612304,
      "invalid_tail_samples": 27.1,
      "learning_rate": 1.2153848818183161e-05,
      "loss": 0.0211,
      "positive_teacher_pairs": 26090.6,
      "query_h_norm_ratio": 0.41704751253128053,
      "relation_loss": 0.002061582857277244,
      "selected_pairs": 26090.6,
      "step": 23540,
      "student_distance_mean": 22.791987419128418,
      "teacher_distance_mean": 0.8503153681755066,
      "valid_action_samples": 228.9,
      "weighted_auxiliary_loss": 0.002061582857277244,
      "weighted_relation_loss": 0.002061582857277244
    },
    {
      "action_query_batch_std": 0.35463556051254275,
      "action_query_norm": 40.87304344177246,
      "action_relation_loss": 0.0020514488220214843,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.029934518241348,
      "flow_matching_loss": 0.017627991177141665,
      "grad_norm": 0.112796850502491,
      "h_stock_valid_norm": 98.0605369567871,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.2117853511683507e-05,
      "loss": 0.0197,
      "positive_teacher_pairs": 26874.1,
      "query_h_norm_ratio": 0.4168174505233765,
      "relation_loss": 0.0020514488220214843,
      "selected_pairs": 26874.1,
      "step": 23550,
      "student_distance_mean": 23.16886672973633,
      "teacher_distance_mean": 0.8596763372421264,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0020514488220214843,
      "weighted_relation_loss": 0.0020514488220214843
    },
    {
      "action_query_batch_std": 0.35117517709732055,
      "action_query_norm": 40.86205978393555,
      "action_relation_loss": 0.002099504030775279,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.03928905519177,
      "flow_matching_loss": 0.01680407002568245,
      "grad_norm": 0.09841073304414749,
      "h_stock_valid_norm": 97.99096374511718,
      "invalid_tail_samples": 21.7,
      "learning_rate": 1.2081904235563906e-05,
      "loss": 0.0189,
      "positive_teacher_pairs": 27342.8,
      "query_h_norm_ratio": 0.41700004041194916,
      "relation_loss": 0.002099504030775279,
      "selected_pairs": 27342.8,
      "step": 23560,
      "student_distance_mean": 23.14405002593994,
      "teacher_distance_mean": 0.8590806543827056,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.002099504030775279,
      "weighted_relation_loss": 0.002099504030775279
    },
    {
      "action_query_batch_std": 0.3481352686882019,
      "action_query_norm": 40.823530197143555,
      "action_relation_loss": 0.0021180459531024098,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.048643592142188,
      "flow_matching_loss": 0.017924181651324034,
      "grad_norm": 0.11005187034606934,
      "h_stock_valid_norm": 98.13203125,
      "invalid_tail_samples": 24.1,
      "learning_rate": 1.20460010335061e-05,
      "loss": 0.02,
      "positive_teacher_pairs": 26782.3,
      "query_h_norm_ratio": 0.4160117894411087,
      "relation_loss": 0.0021180459531024098,
      "selected_pairs": 26782.3,
      "step": 23570,
      "student_distance_mean": 22.87143211364746,
      "teacher_distance_mean": 0.8512828946113586,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0021180459531024098,
      "weighted_relation_loss": 0.0021180459531024098
    },
    {
      "action_query_batch_std": 0.3516850918531418,
      "action_query_norm": 40.86639060974121,
      "action_relation_loss": 0.002089414303191006,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.05799812909261,
      "flow_matching_loss": 0.017744565661996604,
      "grad_norm": 0.09214925020933151,
      "h_stock_valid_norm": 98.29550552368164,
      "invalid_tail_samples": 25.1,
      "learning_rate": 1.2010143949135865e-05,
      "loss": 0.0198,
      "positive_teacher_pairs": 26564.5,
      "query_h_norm_ratio": 0.4157532602548599,
      "relation_loss": 0.002089414303191006,
      "selected_pairs": 26564.5,
      "step": 23580,
      "student_distance_mean": 23.191164016723633,
      "teacher_distance_mean": 0.8636395275592804,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.002089414303191006,
      "weighted_relation_loss": 0.002089414303191006
    },
    {
      "action_query_batch_std": 0.355782949924469,
      "action_query_norm": 40.88807373046875,
      "action_relation_loss": 0.0021135336137376726,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.067352666043032,
      "flow_matching_loss": 0.018294787406921385,
      "grad_norm": 0.0801146849989891,
      "h_stock_valid_norm": 97.92133026123047,
      "invalid_tail_samples": 21.2,
      "learning_rate": 1.1974333026022938e-05,
      "loss": 0.0204,
      "positive_teacher_pairs": 27461.7,
      "query_h_norm_ratio": 0.4175622969865799,
      "relation_loss": 0.0021135336137376726,
      "selected_pairs": 27461.7,
      "step": 23590,
      "student_distance_mean": 23.397561836242676,
      "teacher_distance_mean": 0.8694544672966004,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.0021135336137376726,
      "weighted_relation_loss": 0.0021135336137376726
    },
    {
      "action_query_batch_std": 0.35244332253932953,
      "action_query_norm": 40.88188934326172,
      "action_relation_loss": 0.0020334429922513665,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.07670720299345,
      "flow_matching_loss": 0.019759645126760005,
      "grad_norm": 0.0848734974861145,
      "h_stock_valid_norm": 97.84278030395508,
      "invalid_tail_samples": 22.4,
      "learning_rate": 1.1938568307680964e-05,
      "loss": 0.0218,
      "positive_teacher_pairs": 27181.0,
      "query_h_norm_ratio": 0.417834410071373,
      "relation_loss": 0.0020334429922513665,
      "selected_pairs": 27181.0,
      "step": 23600,
      "student_distance_mean": 23.06207981109619,
      "teacher_distance_mean": 0.8568421840667725,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.0020334429922513665,
      "weighted_relation_loss": 0.0020334429922513665
    },
    {
      "action_query_batch_std": 0.35247364938259124,
      "action_query_norm": 40.8460807800293,
      "action_relation_loss": 0.00208442643051967,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.086061739943872,
      "flow_matching_loss": 0.016840056516230106,
      "grad_norm": 0.12679696083068848,
      "h_stock_valid_norm": 97.58981781005859,
      "invalid_tail_samples": 24.6,
      "learning_rate": 1.1902849837567464e-05,
      "loss": 0.0189,
      "positive_teacher_pairs": 26660.5,
      "query_h_norm_ratio": 0.41855316162109374,
      "relation_loss": 0.00208442643051967,
      "selected_pairs": 26660.5,
      "step": 23610,
      "student_distance_mean": 23.23484573364258,
      "teacher_distance_mean": 0.8624653458595276,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.00208442643051967,
      "weighted_relation_loss": 0.00208442643051967
    },
    {
      "action_query_batch_std": 0.3493127405643463,
      "action_query_norm": 40.82159233093262,
      "action_relation_loss": 0.0021661870181560517,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.095416276894294,
      "flow_matching_loss": 0.017891449481248857,
      "grad_norm": 0.10024784505367279,
      "h_stock_valid_norm": 97.53485794067383,
      "invalid_tail_samples": 25.0,
      "learning_rate": 1.1867177659083739e-05,
      "loss": 0.0201,
      "positive_teacher_pairs": 26571.1,
      "query_h_norm_ratio": 0.41853408217430116,
      "relation_loss": 0.0021661870181560517,
      "selected_pairs": 26571.1,
      "step": 23620,
      "student_distance_mean": 22.957920455932616,
      "teacher_distance_mean": 0.852611380815506,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0021661870181560517,
      "weighted_relation_loss": 0.0021661870181560517
    },
    {
      "action_query_batch_std": 0.3504002243280411,
      "action_query_norm": 40.86529006958008,
      "action_relation_loss": 0.0021562979905866086,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.104770813844716,
      "flow_matching_loss": 0.018855020962655546,
      "grad_norm": 0.09824316203594208,
      "h_stock_valid_norm": 97.62019882202148,
      "invalid_tail_samples": 20.0,
      "learning_rate": 1.1831551815574848e-05,
      "loss": 0.021,
      "positive_teacher_pairs": 27736.0,
      "query_h_norm_ratio": 0.4186177760362625,
      "relation_loss": 0.0021562979905866086,
      "selected_pairs": 27736.0,
      "step": 23630,
      "student_distance_mean": 23.02950973510742,
      "teacher_distance_mean": 0.8554836869239807,
      "valid_action_samples": 236.0,
      "weighted_auxiliary_loss": 0.0021562979905866086,
      "weighted_relation_loss": 0.0021562979905866086
    },
    {
      "action_query_batch_std": 0.3495037168264389,
      "action_query_norm": 40.84339637756348,
      "action_relation_loss": 0.002245269645936787,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.114125350795135,
      "flow_matching_loss": 0.018861996568739416,
      "grad_norm": 0.12703627347946167,
      "h_stock_valid_norm": 97.86536331176758,
      "invalid_tail_samples": 25.0,
      "learning_rate": 1.1795972350329554e-05,
      "loss": 0.0211,
      "positive_teacher_pairs": 26579.5,
      "query_h_norm_ratio": 0.41734724342823026,
      "relation_loss": 0.002245269645936787,
      "selected_pairs": 26579.5,
      "step": 23640,
      "student_distance_mean": 22.939640617370607,
      "teacher_distance_mean": 0.853944593667984,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.002245269645936787,
      "weighted_relation_loss": 0.002245269645936787
    },
    {
      "action_query_batch_std": 0.3459760695695877,
      "action_query_norm": 40.8090160369873,
      "action_relation_loss": 0.0021129284985363485,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.123479887745557,
      "flow_matching_loss": 0.018138814717531204,
      "grad_norm": 0.09378980845212936,
      "h_stock_valid_norm": 97.84596633911133,
      "invalid_tail_samples": 23.3,
      "learning_rate": 1.176043930658024e-05,
      "loss": 0.0203,
      "positive_teacher_pairs": 26965.6,
      "query_h_norm_ratio": 0.4170774966478348,
      "relation_loss": 0.0021129284985363485,
      "selected_pairs": 26965.6,
      "step": 23650,
      "student_distance_mean": 22.808216285705566,
      "teacher_distance_mean": 0.84664067029953,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0021129284985363485,
      "weighted_relation_loss": 0.0021129284985363485
    },
    {
      "action_query_batch_std": 0.34931338131427764,
      "action_query_norm": 40.858218383789065,
      "action_relation_loss": 0.0020962335867807267,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.13283442469598,
      "flow_matching_loss": 0.018010623659938572,
      "grad_norm": 0.09712062031030655,
      "h_stock_valid_norm": 97.83157730102539,
      "invalid_tail_samples": 24.6,
      "learning_rate": 1.1724952727502942e-05,
      "loss": 0.0201,
      "positive_teacher_pairs": 26669.5,
      "query_h_norm_ratio": 0.4176399827003479,
      "relation_loss": 0.0020962335867807267,
      "selected_pairs": 26669.5,
      "step": 23660,
      "student_distance_mean": 23.07593193054199,
      "teacher_distance_mean": 0.8559212267398835,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0020962335867807267,
      "weighted_relation_loss": 0.0020962335867807267
    },
    {
      "action_query_batch_std": 0.3513458877801895,
      "action_query_norm": 40.85997123718262,
      "action_relation_loss": 0.002087342622689903,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.142188961646397,
      "flow_matching_loss": 0.018416247516870498,
      "grad_norm": 0.08661138266324997,
      "h_stock_valid_norm": 97.86880111694336,
      "invalid_tail_samples": 23.4,
      "learning_rate": 1.1689512656217178e-05,
      "loss": 0.0205,
      "positive_teacher_pairs": 26948.6,
      "query_h_norm_ratio": 0.41749849617481233,
      "relation_loss": 0.002087342622689903,
      "selected_pairs": 26948.6,
      "step": 23670,
      "student_distance_mean": 23.082146644592285,
      "teacher_distance_mean": 0.8560546994209289,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002087342622689903,
      "weighted_relation_loss": 0.002087342622689903
    },
    {
      "action_query_batch_std": 0.3494957000017166,
      "action_query_norm": 40.83638153076172,
      "action_relation_loss": 0.002108856337144971,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.15154349859682,
      "flow_matching_loss": 0.01867959937080741,
      "grad_norm": 0.11444606631994247,
      "h_stock_valid_norm": 97.9418556213379,
      "invalid_tail_samples": 25.1,
      "learning_rate": 1.1654119135785962e-05,
      "loss": 0.0208,
      "positive_teacher_pairs": 26555.9,
      "query_h_norm_ratio": 0.41695054471492765,
      "relation_loss": 0.002108856337144971,
      "selected_pairs": 26555.9,
      "step": 23680,
      "student_distance_mean": 23.019787216186522,
      "teacher_distance_mean": 0.8550256669521332,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.002108856337144971,
      "weighted_relation_loss": 0.002108856337144971
    },
    {
      "action_query_batch_std": 0.3478525787591934,
      "action_query_norm": 40.82752571105957,
      "action_relation_loss": 0.0021398998913355173,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.16089803554724,
      "flow_matching_loss": 0.01848315093666315,
      "grad_norm": 0.13836920261383057,
      "h_stock_valid_norm": 98.24667892456054,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.1618772209215794e-05,
      "loss": 0.0206,
      "positive_teacher_pairs": 27019.2,
      "query_h_norm_ratio": 0.4155635148286819,
      "relation_loss": 0.0021398998913355173,
      "selected_pairs": 27019.2,
      "step": 23690,
      "student_distance_mean": 22.85799560546875,
      "teacher_distance_mean": 0.8487030029296875,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0021398998913355173,
      "weighted_relation_loss": 0.0021398998913355173
    },
    {
      "action_query_batch_std": 0.35055847764015197,
      "action_query_norm": 40.84348678588867,
      "action_relation_loss": 0.002060342812910676,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.170252572497663,
      "flow_matching_loss": 0.016916164290159942,
      "grad_norm": 0.09674052894115448,
      "h_stock_valid_norm": 98.36694259643555,
      "invalid_tail_samples": 22.6,
      "learning_rate": 1.1583471919456506e-05,
      "loss": 0.019,
      "positive_teacher_pairs": 27126.0,
      "query_h_norm_ratio": 0.41521660685539247,
      "relation_loss": 0.002060342812910676,
      "selected_pairs": 27126.0,
      "step": 23700,
      "student_distance_mean": 22.99634552001953,
      "teacher_distance_mean": 0.8545035779476166,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.002060342812910676,
      "weighted_relation_loss": 0.002060342812910676
    },
    {
      "action_query_batch_std": 0.34810554683208467,
      "action_query_norm": 40.823981857299806,
      "action_relation_loss": 0.0021361228660680353,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.17960710944808,
      "flow_matching_loss": 0.017457714769989252,
      "grad_norm": 0.1026804968714714,
      "h_stock_valid_norm": 98.1904182434082,
      "invalid_tail_samples": 25.4,
      "learning_rate": 1.1548218309401266e-05,
      "loss": 0.0196,
      "positive_teacher_pairs": 26476.7,
      "query_h_norm_ratio": 0.4157676786184311,
      "relation_loss": 0.0021361228660680353,
      "selected_pairs": 26476.7,
      "step": 23710,
      "student_distance_mean": 22.784664154052734,
      "teacher_distance_mean": 0.846165931224823,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0021361228660680353,
      "weighted_relation_loss": 0.0021361228660680353
    },
    {
      "action_query_batch_std": 0.3503779500722885,
      "action_query_norm": 40.823097229003906,
      "action_relation_loss": 0.002129070193041116,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.188961646398504,
      "flow_matching_loss": 0.01733510522171855,
      "grad_norm": 0.11400020867586136,
      "h_stock_valid_norm": 97.80561218261718,
      "invalid_tail_samples": 25.1,
      "learning_rate": 1.1513011421886555e-05,
      "loss": 0.0195,
      "positive_teacher_pairs": 26556.4,
      "query_h_norm_ratio": 0.41739384829998016,
      "relation_loss": 0.002129070193041116,
      "selected_pairs": 26556.4,
      "step": 23720,
      "student_distance_mean": 23.047933769226074,
      "teacher_distance_mean": 0.8566198825836182,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.002129070193041116,
      "weighted_relation_loss": 0.002129070193041116
    },
    {
      "action_query_batch_std": 0.3513912230730057,
      "action_query_norm": 40.84076843261719,
      "action_relation_loss": 0.0020430605043657126,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.198316183348926,
      "flow_matching_loss": 0.01596337528899312,
      "grad_norm": 0.2114395797252655,
      "h_stock_valid_norm": 97.53727340698242,
      "invalid_tail_samples": 21.8,
      "learning_rate": 1.1477851299692056e-05,
      "loss": 0.018,
      "positive_teacher_pairs": 27320.2,
      "query_h_norm_ratio": 0.41872344315052035,
      "relation_loss": 0.0020430605043657126,
      "selected_pairs": 27320.2,
      "step": 23730,
      "student_distance_mean": 23.014277267456055,
      "teacher_distance_mean": 0.8551732838153839,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.0020430605043657126,
      "weighted_relation_loss": 0.0020430605043657126
    },
    {
      "action_query_batch_std": 0.34895937144756317,
      "action_query_norm": 40.838177871704104,
      "action_relation_loss": 0.0021325009525753557,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.207670720299344,
      "flow_matching_loss": 0.017085214238613845,
      "grad_norm": 0.15325678884983063,
      "h_stock_valid_norm": 97.5036994934082,
      "invalid_tail_samples": 21.9,
      "learning_rate": 1.144273798554063e-05,
      "loss": 0.0192,
      "positive_teacher_pairs": 27289.5,
      "query_h_norm_ratio": 0.41883830428123475,
      "relation_loss": 0.0021325009525753557,
      "selected_pairs": 27289.5,
      "step": 23740,
      "student_distance_mean": 22.907831764221193,
      "teacher_distance_mean": 0.8521166265010833,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.0021325009525753557,
      "weighted_relation_loss": 0.0021325009525753557
    },
    {
      "action_query_batch_std": 0.35270353555679324,
      "action_query_norm": 40.83035392761231,
      "action_relation_loss": 0.00212514519225806,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.217025257249766,
      "flow_matching_loss": 0.01798565676435828,
      "grad_norm": 0.09319111704826355,
      "h_stock_valid_norm": 97.54648361206054,
      "invalid_tail_samples": 23.8,
      "learning_rate": 1.1407671522098262e-05,
      "loss": 0.0201,
      "positive_teacher_pairs": 26846.0,
      "query_h_norm_ratio": 0.4185762584209442,
      "relation_loss": 0.00212514519225806,
      "selected_pairs": 26846.0,
      "step": 23750,
      "student_distance_mean": 23.099261474609374,
      "teacher_distance_mean": 0.861338347196579,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.00212514519225806,
      "weighted_relation_loss": 0.00212514519225806
    },
    {
      "action_query_batch_std": 0.34930036067962644,
      "action_query_norm": 40.798352813720705,
      "action_relation_loss": 0.0020494975382462145,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.226379794200188,
      "flow_matching_loss": 0.017683976236730813,
      "grad_norm": 0.09460613131523132,
      "h_stock_valid_norm": 97.75336532592773,
      "invalid_tail_samples": 22.5,
      "learning_rate": 1.1372651951974e-05,
      "loss": 0.0197,
      "positive_teacher_pairs": 27152.3,
      "query_h_norm_ratio": 0.4173611342906952,
      "relation_loss": 0.0020494975382462145,
      "selected_pairs": 27152.3,
      "step": 23760,
      "student_distance_mean": 23.009911346435548,
      "teacher_distance_mean": 0.8581305623054505,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0020494975382462145,
      "weighted_relation_loss": 0.0020494975382462145
    },
    {
      "action_query_batch_std": 0.34615033864974976,
      "action_query_norm": 40.79229316711426,
      "action_relation_loss": 0.0020997258950956168,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.235734331150606,
      "flow_matching_loss": 0.017177801020443438,
      "grad_norm": 0.08339061588048935,
      "h_stock_valid_norm": 97.8672248840332,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.1337679317719952e-05,
      "loss": 0.0193,
      "positive_teacher_pairs": 26909.5,
      "query_h_norm_ratio": 0.4168133556842804,
      "relation_loss": 0.0020997258950956168,
      "selected_pairs": 26909.5,
      "step": 23770,
      "student_distance_mean": 22.78512496948242,
      "teacher_distance_mean": 0.8480682730674743,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0020997258950956168,
      "weighted_relation_loss": 0.0020997258950956168
    },
    {
      "action_query_batch_std": 0.3493582189083099,
      "action_query_norm": 40.81434020996094,
      "action_relation_loss": 0.002094562549609691,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.24508886810103,
      "flow_matching_loss": 0.01614360688254237,
      "grad_norm": 0.08019273728132248,
      "h_stock_valid_norm": 97.75819244384766,
      "invalid_tail_samples": 26.2,
      "learning_rate": 1.1302753661831151e-05,
      "loss": 0.0182,
      "positive_teacher_pairs": 26299.2,
      "query_h_norm_ratio": 0.41750659644603727,
      "relation_loss": 0.002094562549609691,
      "selected_pairs": 26299.2,
      "step": 23780,
      "student_distance_mean": 22.88293399810791,
      "teacher_distance_mean": 0.8506523907184601,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.002094562549609691,
      "weighted_relation_loss": 0.002094562549609691
    },
    {
      "action_query_batch_std": 0.34621461629867556,
      "action_query_norm": 40.818972778320315,
      "action_relation_loss": 0.002194523252546787,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.25444340505145,
      "flow_matching_loss": 0.017745980620384218,
      "grad_norm": 0.17296725511550903,
      "h_stock_valid_norm": 97.71716842651367,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.1267875026745561e-05,
      "loss": 0.0199,
      "positive_teacher_pairs": 26805.6,
      "query_h_norm_ratio": 0.41772725582122805,
      "relation_loss": 0.002194523252546787,
      "selected_pairs": 26805.6,
      "step": 23790,
      "student_distance_mean": 22.78506736755371,
      "teacher_distance_mean": 0.847198623418808,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.002194523252546787,
      "weighted_relation_loss": 0.002194523252546787
    },
    {
      "action_query_batch_std": 0.3573063939809799,
      "action_query_norm": 40.87803001403809,
      "action_relation_loss": 0.0020510048838332296,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.263797942001872,
      "flow_matching_loss": 0.017380001209676265,
      "grad_norm": 0.10977038741111755,
      "h_stock_valid_norm": 97.56794891357421,
      "invalid_tail_samples": 23.3,
      "learning_rate": 1.1233043454844016e-05,
      "loss": 0.0194,
      "positive_teacher_pairs": 26963.9,
      "query_h_norm_ratio": 0.41897078454494474,
      "relation_loss": 0.0020510048838332296,
      "selected_pairs": 26963.9,
      "step": 23800,
      "student_distance_mean": 23.437066078186035,
      "teacher_distance_mean": 0.8721794664859772,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0020510048838332296,
      "weighted_relation_loss": 0.0020510048838332296
    },
    {
      "action_query_batch_std": 0.354949751496315,
      "action_query_norm": 40.84057121276855,
      "action_relation_loss": 0.002112687425687909,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.27315247895229,
      "flow_matching_loss": 0.01850593900308013,
      "grad_norm": 0.12428075075149536,
      "h_stock_valid_norm": 97.33687591552734,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.1198258988450144e-05,
      "loss": 0.0206,
      "positive_teacher_pairs": 27015.4,
      "query_h_norm_ratio": 0.4195818483829498,
      "relation_loss": 0.002112687425687909,
      "selected_pairs": 27015.4,
      "step": 23810,
      "student_distance_mean": 23.31623935699463,
      "teacher_distance_mean": 0.8694539546966553,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002112687425687909,
      "weighted_relation_loss": 0.002112687425687909
    },
    {
      "action_query_batch_std": 0.34941443502902986,
      "action_query_norm": 40.826150512695314,
      "action_relation_loss": 0.00211502630263567,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.282507015902713,
      "flow_matching_loss": 0.01831147205084562,
      "grad_norm": 0.11009105294942856,
      "h_stock_valid_norm": 97.37522964477539,
      "invalid_tail_samples": 21.4,
      "learning_rate": 1.1163521669830368e-05,
      "loss": 0.0204,
      "positive_teacher_pairs": 27405.1,
      "query_h_norm_ratio": 0.4192686527967453,
      "relation_loss": 0.00211502630263567,
      "selected_pairs": 27405.1,
      "step": 23820,
      "student_distance_mean": 22.94960823059082,
      "teacher_distance_mean": 0.8559471309185028,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.00211502630263567,
      "weighted_relation_loss": 0.00211502630263567
    },
    {
      "action_query_batch_std": 0.346015653014183,
      "action_query_norm": 40.786949920654294,
      "action_relation_loss": 0.0021731693064793943,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.291861552853135,
      "flow_matching_loss": 0.01859332751482725,
      "grad_norm": 0.10916482657194138,
      "h_stock_valid_norm": 97.82630233764648,
      "invalid_tail_samples": 27.3,
      "learning_rate": 1.1128831541193818e-05,
      "loss": 0.0208,
      "positive_teacher_pairs": 26051.9,
      "query_h_norm_ratio": 0.41693301498889923,
      "relation_loss": 0.0021731693064793943,
      "selected_pairs": 26051.9,
      "step": 23830,
      "student_distance_mean": 22.866211700439454,
      "teacher_distance_mean": 0.8502493321895599,
      "valid_action_samples": 228.7,
      "weighted_auxiliary_loss": 0.0021731693064793943,
      "weighted_relation_loss": 0.0021731693064793943
    },
    {
      "action_query_batch_std": 0.35114347338676455,
      "action_query_norm": 40.822030258178714,
      "action_relation_loss": 0.0022131934762001037,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.301216089803553,
      "flow_matching_loss": 0.018245735485106707,
      "grad_norm": 0.11392652243375778,
      "h_stock_valid_norm": 97.9726547241211,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.1094188644692255e-05,
      "loss": 0.0205,
      "positive_teacher_pairs": 27009.0,
      "query_h_norm_ratio": 0.4166688293218613,
      "relation_loss": 0.0022131934762001037,
      "selected_pairs": 27009.0,
      "step": 23840,
      "student_distance_mean": 23.006905174255373,
      "teacher_distance_mean": 0.8565351903438568,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0022131934762001037,
      "weighted_relation_loss": 0.0022131934762001037
    },
    {
      "action_query_batch_std": 0.34840327203273774,
      "action_query_norm": 40.78988761901856,
      "action_relation_loss": 0.002131305949296802,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.310570626753975,
      "flow_matching_loss": 0.01839590920135379,
      "grad_norm": 0.13701027631759644,
      "h_stock_valid_norm": 97.64288635253907,
      "invalid_tail_samples": 26.7,
      "learning_rate": 1.1059593022420067e-05,
      "loss": 0.0205,
      "positive_teacher_pairs": 26184.5,
      "query_h_norm_ratio": 0.41774965524673463,
      "relation_loss": 0.002131305949296802,
      "selected_pairs": 26184.5,
      "step": 23850,
      "student_distance_mean": 22.99454116821289,
      "teacher_distance_mean": 0.8580619692802429,
      "valid_action_samples": 229.3,
      "weighted_auxiliary_loss": 0.002131305949296802,
      "weighted_relation_loss": 0.002131305949296802
    },
    {
      "action_query_batch_std": 0.35093735456466674,
      "action_query_norm": 40.82414283752441,
      "action_relation_loss": 0.002150159887969494,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.319925163704397,
      "flow_matching_loss": 0.018067675922065974,
      "grad_norm": 0.1090458482503891,
      "h_stock_valid_norm": 97.67624130249024,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.1025044716414184e-05,
      "loss": 0.0202,
      "positive_teacher_pairs": 26757.7,
      "query_h_norm_ratio": 0.4179555684328079,
      "relation_loss": 0.002150159887969494,
      "selected_pairs": 26757.7,
      "step": 23860,
      "student_distance_mean": 23.124706268310547,
      "teacher_distance_mean": 0.8608139157295227,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.002150159887969494,
      "weighted_relation_loss": 0.002150159887969494
    },
    {
      "action_query_batch_std": 0.3506525486707687,
      "action_query_norm": 40.80496406555176,
      "action_relation_loss": 0.0020113020786084236,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.32927970065482,
      "flow_matching_loss": 0.016997635271400212,
      "grad_norm": 0.10058266669511795,
      "h_stock_valid_norm": 97.92334213256837,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.0990543768654082e-05,
      "loss": 0.019,
      "positive_teacher_pairs": 26804.4,
      "query_h_norm_ratio": 0.41670753955841067,
      "relation_loss": 0.0020113020786084236,
      "selected_pairs": 26804.4,
      "step": 23870,
      "student_distance_mean": 23.03685932159424,
      "teacher_distance_mean": 0.8588102400302887,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0020113020786084236,
      "weighted_relation_loss": 0.0020113020786084236
    },
    {
      "action_query_batch_std": 0.346627289056778,
      "action_query_norm": 40.75433158874512,
      "action_relation_loss": 0.002179406792856753,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.338634237605238,
      "flow_matching_loss": 0.016878704261034726,
      "grad_norm": 0.10829087346792221,
      "h_stock_valid_norm": 97.80125198364257,
      "invalid_tail_samples": 24.5,
      "learning_rate": 1.0956090221061655e-05,
      "loss": 0.0191,
      "positive_teacher_pairs": 26689.7,
      "query_h_norm_ratio": 0.41671003103256227,
      "relation_loss": 0.002179406792856753,
      "selected_pairs": 26689.7,
      "step": 23880,
      "student_distance_mean": 22.825977897644044,
      "teacher_distance_mean": 0.8518024265766144,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.002179406792856753,
      "weighted_relation_loss": 0.002179406792856753
    },
    {
      "action_query_batch_std": 0.34948924481868743,
      "action_query_norm": 40.78249397277832,
      "action_relation_loss": 0.002036551584023982,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.34798877455566,
      "flow_matching_loss": 0.01806167159229517,
      "grad_norm": 0.09938054531812668,
      "h_stock_valid_norm": 97.83184432983398,
      "invalid_tail_samples": 23.8,
      "learning_rate": 1.0921684115501208e-05,
      "loss": 0.0201,
      "positive_teacher_pairs": 26859.9,
      "query_h_norm_ratio": 0.41686452329158785,
      "relation_loss": 0.002036551584023982,
      "selected_pairs": 26859.9,
      "step": 23890,
      "student_distance_mean": 22.956085777282716,
      "teacher_distance_mean": 0.8553228437900543,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.002036551584023982,
      "weighted_relation_loss": 0.002036551584023982
    },
    {
      "action_query_batch_std": 0.3475064069032669,
      "action_query_norm": 40.790840530395506,
      "action_relation_loss": 0.002176578948274255,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.35734331150608,
      "flow_matching_loss": 0.018374201469123365,
      "grad_norm": 0.11478500068187714,
      "h_stock_valid_norm": 97.82283401489258,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.0887325493779405e-05,
      "loss": 0.0206,
      "positive_teacher_pairs": 26838.5,
      "query_h_norm_ratio": 0.4169889628887177,
      "relation_loss": 0.002176578948274255,
      "selected_pairs": 26838.5,
      "step": 23900,
      "student_distance_mean": 22.90255661010742,
      "teacher_distance_mean": 0.8536721050739289,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.002176578948274255,
      "weighted_relation_loss": 0.002176578948274255
    },
    {
      "action_query_batch_std": 0.35070980489254,
      "action_query_norm": 40.79458084106445,
      "action_relation_loss": 0.0020393421058543027,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.3666978484565,
      "flow_matching_loss": 0.016406848561018705,
      "grad_norm": 0.12017332762479782,
      "h_stock_valid_norm": 97.96829147338867,
      "invalid_tail_samples": 24.5,
      "learning_rate": 1.0853014397645207e-05,
      "loss": 0.0184,
      "positive_teacher_pairs": 26695.7,
      "query_h_norm_ratio": 0.4164064019918442,
      "relation_loss": 0.0020393421058543027,
      "selected_pairs": 26695.7,
      "step": 23910,
      "student_distance_mean": 22.99672927856445,
      "teacher_distance_mean": 0.8573339462280274,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0020393421058543027,
      "weighted_relation_loss": 0.0020393421058543027
    },
    {
      "action_query_batch_std": 0.34781189560890197,
      "action_query_norm": 40.787496185302736,
      "action_relation_loss": 0.0021246668184176087,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.376052385406922,
      "flow_matching_loss": 0.017447217460721733,
      "grad_norm": 0.11833827942609787,
      "h_stock_valid_norm": 97.74444732666015,
      "invalid_tail_samples": 25.6,
      "learning_rate": 1.0818750868789829e-05,
      "loss": 0.0196,
      "positive_teacher_pairs": 26433.6,
      "query_h_norm_ratio": 0.41728958785533904,
      "relation_loss": 0.0021246668184176087,
      "selected_pairs": 26433.6,
      "step": 23920,
      "student_distance_mean": 22.81471996307373,
      "teacher_distance_mean": 0.8489904463291168,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0021246668184176087,
      "weighted_relation_loss": 0.0021246668184176087
    },
    {
      "action_query_batch_std": 0.3522741585969925,
      "action_query_norm": 40.828162384033206,
      "action_relation_loss": 0.002020621451083571,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.385406922357344,
      "flow_matching_loss": 0.017598217446357013,
      "grad_norm": 0.13214924931526184,
      "h_stock_valid_norm": 98.00770492553711,
      "invalid_tail_samples": 22.1,
      "learning_rate": 1.0784534948846703e-05,
      "loss": 0.0196,
      "positive_teacher_pairs": 27245.9,
      "query_h_norm_ratio": 0.4165827870368958,
      "relation_loss": 0.002020621451083571,
      "selected_pairs": 27245.9,
      "step": 23930,
      "student_distance_mean": 23.142811584472657,
      "teacher_distance_mean": 0.8632162034511566,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.002020621451083571,
      "weighted_relation_loss": 0.002020621451083571
    },
    {
      "action_query_batch_std": 0.3517685145139694,
      "action_query_norm": 40.805152130126956,
      "action_relation_loss": 0.0021143052959814667,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.394761459307766,
      "flow_matching_loss": 0.01881320970132947,
      "grad_norm": 0.13990046083927155,
      "h_stock_valid_norm": 98.0085578918457,
      "invalid_tail_samples": 25.6,
      "learning_rate": 1.0750366679391393e-05,
      "loss": 0.0209,
      "positive_teacher_pairs": 26440.8,
      "query_h_norm_ratio": 0.4163470357656479,
      "relation_loss": 0.0021143052959814667,
      "selected_pairs": 26440.8,
      "step": 23940,
      "student_distance_mean": 23.146740341186522,
      "teacher_distance_mean": 0.8619961440563202,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0021143052959814667,
      "weighted_relation_loss": 0.0021143052959814667
    },
    {
      "action_query_batch_std": 0.34771160781383514,
      "action_query_norm": 40.8330738067627,
      "action_relation_loss": 0.0020110207144170998,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.404115996258184,
      "flow_matching_loss": 0.01780571509152651,
      "grad_norm": 0.10087810456752777,
      "h_stock_valid_norm": 97.84759292602538,
      "invalid_tail_samples": 22.5,
      "learning_rate": 1.0716246101941557e-05,
      "loss": 0.0198,
      "positive_teacher_pairs": 27157.4,
      "query_h_norm_ratio": 0.41731628179550173,
      "relation_loss": 0.0020110207144170998,
      "selected_pairs": 27157.4,
      "step": 23950,
      "student_distance_mean": 22.828743171691894,
      "teacher_distance_mean": 0.850716894865036,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0020110207144170998,
      "weighted_relation_loss": 0.0020110207144170998
    },
    {
      "action_query_batch_std": 0.3494638502597809,
      "action_query_norm": 40.776310729980466,
      "action_relation_loss": 0.002100602281279862,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.413470533208606,
      "flow_matching_loss": 0.01677681440487504,
      "grad_norm": 0.09767525643110275,
      "h_stock_valid_norm": 97.82458953857422,
      "invalid_tail_samples": 24.4,
      "learning_rate": 1.0682173257956934e-05,
      "loss": 0.0189,
      "positive_teacher_pairs": 26709.9,
      "query_h_norm_ratio": 0.41683806478977203,
      "relation_loss": 0.002100602281279862,
      "selected_pairs": 26709.9,
      "step": 23960,
      "student_distance_mean": 22.942413330078125,
      "teacher_distance_mean": 0.852373731136322,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.002100602281279862,
      "weighted_relation_loss": 0.002100602281279862
    },
    {
      "action_query_batch_std": 0.3499838560819626,
      "action_query_norm": 40.822357940673825,
      "action_relation_loss": 0.0021021330147050323,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.42282507015903,
      "flow_matching_loss": 0.016180775500833987,
      "grad_norm": 0.0949501097202301,
      "h_stock_valid_norm": 98.04284439086913,
      "invalid_tail_samples": 25.3,
      "learning_rate": 1.064814818883922e-05,
      "loss": 0.0183,
      "positive_teacher_pairs": 26502.5,
      "query_h_norm_ratio": 0.41637648940086364,
      "relation_loss": 0.0021021330147050323,
      "selected_pairs": 26502.5,
      "step": 23970,
      "student_distance_mean": 23.071041107177734,
      "teacher_distance_mean": 0.8569958508014679,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0021021330147050323,
      "weighted_relation_loss": 0.0021021330147050323
    },
    {
      "action_query_batch_std": 0.3514868974685669,
      "action_query_norm": 40.815674209594725,
      "action_relation_loss": 0.002168475661892444,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.432179607109447,
      "flow_matching_loss": 0.018531528860330583,
      "grad_norm": 0.11732813715934753,
      "h_stock_valid_norm": 97.90873565673829,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.06141709359321e-05,
      "loss": 0.0207,
      "positive_teacher_pairs": 26835.2,
      "query_h_norm_ratio": 0.4168775141239166,
      "relation_loss": 0.002168475661892444,
      "selected_pairs": 26835.2,
      "step": 23980,
      "student_distance_mean": 23.055560493469237,
      "teacher_distance_mean": 0.8574924468994141,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.002168475661892444,
      "weighted_relation_loss": 0.002168475661892444
    },
    {
      "action_query_batch_std": 0.34806529581546786,
      "action_query_norm": 40.78258590698242,
      "action_relation_loss": 0.0020554266520775855,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.44153414405987,
      "flow_matching_loss": 0.016590792499482632,
      "grad_norm": 0.10853446274995804,
      "h_stock_valid_norm": 97.90133666992188,
      "invalid_tail_samples": 25.4,
      "learning_rate": 1.058024154052114e-05,
      "loss": 0.0186,
      "positive_teacher_pairs": 26475.9,
      "query_h_norm_ratio": 0.4165710985660553,
      "relation_loss": 0.0020554266520775855,
      "selected_pairs": 26475.9,
      "step": 23990,
      "student_distance_mean": 22.838426208496095,
      "teacher_distance_mean": 0.849645984172821,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0020554266520775855,
      "weighted_relation_loss": 0.0020554266520775855
    },
    {
      "action_query_batch_std": 0.34995185136795043,
      "action_query_norm": 40.83575439453125,
      "action_relation_loss": 0.002111117565073073,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.45088868101029,
      "flow_matching_loss": 0.01703068716451526,
      "grad_norm": 0.15551283955574036,
      "h_stock_valid_norm": 97.81092529296875,
      "invalid_tail_samples": 22.6,
      "learning_rate": 1.0546360043833747e-05,
      "loss": 0.0191,
      "positive_teacher_pairs": 27126.5,
      "query_h_norm_ratio": 0.4174982160329819,
      "relation_loss": 0.002111117565073073,
      "selected_pairs": 27126.5,
      "step": 24000,
      "student_distance_mean": 23.10806064605713,
      "teacher_distance_mean": 0.8591970384120942,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.002111117565073073,
      "weighted_relation_loss": 0.002111117565073073
    },
    {
      "action_query_batch_std": 0.34357411563396456,
      "action_query_norm": 40.76793441772461,
      "action_relation_loss": 0.002166377764660865,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.46024321796071,
      "flow_matching_loss": 0.0184944030828774,
      "grad_norm": 0.1053575649857521,
      "h_stock_valid_norm": 97.83793258666992,
      "invalid_tail_samples": 21.7,
      "learning_rate": 1.0512526487039138e-05,
      "loss": 0.0207,
      "positive_teacher_pairs": 27335.3,
      "query_h_norm_ratio": 0.41669322848320006,
      "relation_loss": 0.002166377764660865,
      "selected_pairs": 27335.3,
      "step": 24010,
      "student_distance_mean": 22.630657577514647,
      "teacher_distance_mean": 0.8427047371864319,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.002166377764660865,
      "weighted_relation_loss": 0.002166377764660865
    },
    {
      "action_query_batch_std": 0.3509309083223343,
      "action_query_norm": 40.798851013183594,
      "action_relation_loss": 0.002009846572764218,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.46959775491113,
      "flow_matching_loss": 0.0160703438334167,
      "grad_norm": 0.08967047929763794,
      "h_stock_valid_norm": 97.82081756591796,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.0478740911248258e-05,
      "loss": 0.0181,
      "positive_teacher_pairs": 27018.1,
      "query_h_norm_ratio": 0.4170789748430252,
      "relation_loss": 0.002009846572764218,
      "selected_pairs": 27018.1,
      "step": 24020,
      "student_distance_mean": 22.875905418395995,
      "teacher_distance_mean": 0.8529518127441407,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002009846572764218,
      "weighted_relation_loss": 0.002009846572764218
    },
    {
      "action_query_batch_std": 0.35007095634937285,
      "action_query_norm": 40.790442657470706,
      "action_relation_loss": 0.002178037608973682,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.478952291861553,
      "flow_matching_loss": 0.016530559025704862,
      "grad_norm": 0.10697557777166367,
      "h_stock_valid_norm": 97.80721817016601,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.0445003357513793e-05,
      "loss": 0.0187,
      "positive_teacher_pairs": 27104.7,
      "query_h_norm_ratio": 0.41705048084259033,
      "relation_loss": 0.002178037608973682,
      "selected_pairs": 27104.7,
      "step": 24030,
      "student_distance_mean": 22.944181632995605,
      "teacher_distance_mean": 0.8573908507823944,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.002178037608973682,
      "weighted_relation_loss": 0.002178037608973682
    },
    {
      "action_query_batch_std": 0.3486235707998276,
      "action_query_norm": 40.7749137878418,
      "action_relation_loss": 0.002009091363288462,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.488306828811975,
      "flow_matching_loss": 0.01700847102329135,
      "grad_norm": 0.09336473047733307,
      "h_stock_valid_norm": 97.81156539916992,
      "invalid_tail_samples": 22.4,
      "learning_rate": 1.0411313866830042e-05,
      "loss": 0.019,
      "positive_teacher_pairs": 27182.1,
      "query_h_norm_ratio": 0.4168752431869507,
      "relation_loss": 0.002009091363288462,
      "selected_pairs": 27182.1,
      "step": 24040,
      "student_distance_mean": 22.993414306640624,
      "teacher_distance_mean": 0.8568706572055816,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.002009091363288462,
      "weighted_relation_loss": 0.002009091363288462
    },
    {
      "action_query_batch_std": 0.348845911026001,
      "action_query_norm": 40.79842758178711,
      "action_relation_loss": 0.002148772485088557,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.497661365762394,
      "flow_matching_loss": 0.015731734689325094,
      "grad_norm": 0.09689901769161224,
      "h_stock_valid_norm": 97.80419235229492,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.0377672480132915e-05,
      "loss": 0.0179,
      "positive_teacher_pairs": 27081.9,
      "query_h_norm_ratio": 0.41714669167995455,
      "relation_loss": 0.002148772485088557,
      "selected_pairs": 27081.9,
      "step": 24050,
      "student_distance_mean": 22.84507808685303,
      "teacher_distance_mean": 0.8495422303676605,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.002148772485088557,
      "weighted_relation_loss": 0.002148772485088557
    },
    {
      "action_query_batch_std": 0.34957429468631745,
      "action_query_norm": 40.80439338684082,
      "action_relation_loss": 0.002120108553208411,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.507015902712816,
      "flow_matching_loss": 0.01643299050629139,
      "grad_norm": 0.12947343289852142,
      "h_stock_valid_norm": 97.73323287963868,
      "invalid_tail_samples": 20.9,
      "learning_rate": 1.0344079238299864e-05,
      "loss": 0.0186,
      "positive_teacher_pairs": 27527.0,
      "query_h_norm_ratio": 0.41751269102096555,
      "relation_loss": 0.002120108553208411,
      "selected_pairs": 27527.0,
      "step": 24060,
      "student_distance_mean": 22.848410606384277,
      "teacher_distance_mean": 0.8507800161838531,
      "valid_action_samples": 235.1,
      "weighted_auxiliary_loss": 0.002120108553208411,
      "weighted_relation_loss": 0.002120108553208411
    },
    {
      "action_query_batch_std": 0.3525917261838913,
      "action_query_norm": 40.809823989868164,
      "action_relation_loss": 0.002105812565423548,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.516370439663238,
      "flow_matching_loss": 0.01731391092762351,
      "grad_norm": 0.0879070833325386,
      "h_stock_valid_norm": 97.82227020263672,
      "invalid_tail_samples": 25.4,
      "learning_rate": 1.0310534182149833e-05,
      "loss": 0.0194,
      "positive_teacher_pairs": 26483.9,
      "query_h_norm_ratio": 0.41718330085277555,
      "relation_loss": 0.002105812565423548,
      "selected_pairs": 26483.9,
      "step": 24070,
      "student_distance_mean": 23.17246780395508,
      "teacher_distance_mean": 0.8612770140171051,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.002105812565423548,
      "weighted_relation_loss": 0.002105812565423548
    },
    {
      "action_query_batch_std": 0.3471011400222778,
      "action_query_norm": 40.77782936096192,
      "action_relation_loss": 0.0021514427033253016,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.525724976613656,
      "flow_matching_loss": 0.017151215020567177,
      "grad_norm": 0.0926484763622284,
      "h_stock_valid_norm": 97.65115661621094,
      "invalid_tail_samples": 24.8,
      "learning_rate": 1.0277037352443258e-05,
      "loss": 0.0193,
      "positive_teacher_pairs": 26625.2,
      "query_h_norm_ratio": 0.417587947845459,
      "relation_loss": 0.0021514427033253016,
      "selected_pairs": 26625.2,
      "step": 24080,
      "student_distance_mean": 22.835616493225096,
      "teacher_distance_mean": 0.850798225402832,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0021514427033253016,
      "weighted_relation_loss": 0.0021514427033253016
    },
    {
      "action_query_batch_std": 0.3479691594839096,
      "action_query_norm": 40.7947696685791,
      "action_relation_loss": 0.002157353749498725,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.535079513564078,
      "flow_matching_loss": 0.018015866354107856,
      "grad_norm": 0.09572894871234894,
      "h_stock_valid_norm": 97.6223991394043,
      "invalid_tail_samples": 21.6,
      "learning_rate": 1.024358878988193e-05,
      "loss": 0.0202,
      "positive_teacher_pairs": 27366.9,
      "query_h_norm_ratio": 0.41788381338119507,
      "relation_loss": 0.002157353749498725,
      "selected_pairs": 27366.9,
      "step": 24090,
      "student_distance_mean": 22.871367073059083,
      "teacher_distance_mean": 0.8527596354484558,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.002157353749498725,
      "weighted_relation_loss": 0.002157353749498725
    },
    {
      "action_query_batch_std": 0.3468227356672287,
      "action_query_norm": 40.78254432678223,
      "action_relation_loss": 0.0021073397132568062,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.5444340505145,
      "flow_matching_loss": 0.017986131086945533,
      "grad_norm": 0.09608770906925201,
      "h_stock_valid_norm": 97.77827453613281,
      "invalid_tail_samples": 26.1,
      "learning_rate": 1.0210188535108995e-05,
      "loss": 0.0201,
      "positive_teacher_pairs": 26328.0,
      "query_h_norm_ratio": 0.41709227561950685,
      "relation_loss": 0.0021073397132568062,
      "selected_pairs": 26328.0,
      "step": 24100,
      "student_distance_mean": 22.7656192779541,
      "teacher_distance_mean": 0.8485282361507416,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.0021073397132568062,
      "weighted_relation_loss": 0.0021073397132568062
    },
    {
      "action_query_batch_std": 0.35223848521709444,
      "action_query_norm": 40.82504386901856,
      "action_relation_loss": 0.0021083715837448835,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.553788587464922,
      "flow_matching_loss": 0.01581037463620305,
      "grad_norm": 0.09754971414804459,
      "h_stock_valid_norm": 97.96093673706055,
      "invalid_tail_samples": 22.6,
      "learning_rate": 1.0176836628708935e-05,
      "loss": 0.0179,
      "positive_teacher_pairs": 27130.6,
      "query_h_norm_ratio": 0.4167494922876358,
      "relation_loss": 0.0021083715837448835,
      "selected_pairs": 27130.6,
      "step": 24110,
      "student_distance_mean": 23.189783668518068,
      "teacher_distance_mean": 0.8670444846153259,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.0021083715837448835,
      "weighted_relation_loss": 0.0021083715837448835
    },
    {
      "action_query_batch_std": 0.3498631030321121,
      "action_query_norm": 40.792847061157225,
      "action_relation_loss": 0.0020279604359529912,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.56314312441534,
      "flow_matching_loss": 0.0180779074318707,
      "grad_norm": 0.09113913774490356,
      "h_stock_valid_norm": 97.99922561645508,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.0143533111207454e-05,
      "loss": 0.0201,
      "positive_teacher_pairs": 26878.9,
      "query_h_norm_ratio": 0.41625922322273257,
      "relation_loss": 0.0020279604359529912,
      "selected_pairs": 26878.9,
      "step": 24120,
      "student_distance_mean": 23.033404922485353,
      "teacher_distance_mean": 0.8619843065738678,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0020279604359529912,
      "weighted_relation_loss": 0.0020279604359529912
    },
    {
      "action_query_batch_std": 0.34789353609085083,
      "action_query_norm": 40.79242362976074,
      "action_relation_loss": 0.002142752450890839,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.572497661365762,
      "flow_matching_loss": 0.018716789595782757,
      "grad_norm": 0.0952097699046135,
      "h_stock_valid_norm": 98.04525833129883,
      "invalid_tail_samples": 23.4,
      "learning_rate": 1.0110278023071446e-05,
      "loss": 0.0209,
      "positive_teacher_pairs": 26956.0,
      "query_h_norm_ratio": 0.41605846881866454,
      "relation_loss": 0.002142752450890839,
      "selected_pairs": 26956.0,
      "step": 24130,
      "student_distance_mean": 22.86411762237549,
      "teacher_distance_mean": 0.8538169026374817,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002142752450890839,
      "weighted_relation_loss": 0.002142752450890839
    },
    {
      "action_query_batch_std": 0.34236363172531126,
      "action_query_norm": 40.738132858276366,
      "action_relation_loss": 0.00214854150544852,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.581852198316184,
      "flow_matching_loss": 0.01815407397225499,
      "grad_norm": 0.11968429386615753,
      "h_stock_valid_norm": 97.94654541015625,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.007707140470901e-05,
      "loss": 0.0203,
      "positive_teacher_pairs": 27023.3,
      "query_h_norm_ratio": 0.41592274606227875,
      "relation_loss": 0.00214854150544852,
      "selected_pairs": 27023.3,
      "step": 24140,
      "student_distance_mean": 22.553985595703125,
      "teacher_distance_mean": 0.8428349375724793,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.00214854150544852,
      "weighted_relation_loss": 0.00214854150544852
    },
    {
      "action_query_batch_std": 0.34621343910694125,
      "action_query_norm": 40.79181900024414,
      "action_relation_loss": 0.002030929399188608,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.591206735266603,
      "flow_matching_loss": 0.015902674570679665,
      "grad_norm": 0.11750461906194687,
      "h_stock_valid_norm": 97.70786514282227,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.0043913296469298e-05,
      "loss": 0.0179,
      "positive_teacher_pairs": 26734.7,
      "query_h_norm_ratio": 0.4174901694059372,
      "relation_loss": 0.002030929399188608,
      "selected_pairs": 26734.7,
      "step": 24150,
      "student_distance_mean": 22.82017230987549,
      "teacher_distance_mean": 0.8502480864524842,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002030929399188608,
      "weighted_relation_loss": 0.002030929399188608
    },
    {
      "action_query_batch_std": 0.34761241674423216,
      "action_query_norm": 40.78842391967773,
      "action_relation_loss": 0.0021302281995303927,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.600561272217025,
      "flow_matching_loss": 0.016754387971013786,
      "grad_norm": 0.14889897406101227,
      "h_stock_valid_norm": 97.84331588745117,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.001080373864255e-05,
      "loss": 0.0189,
      "positive_teacher_pairs": 26760.6,
      "query_h_norm_ratio": 0.41687813103199006,
      "relation_loss": 0.0021302281995303927,
      "selected_pairs": 26760.6,
      "step": 24160,
      "student_distance_mean": 22.855389976501463,
      "teacher_distance_mean": 0.852543169260025,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0021302281995303927,
      "weighted_relation_loss": 0.0021302281995303927
    },
    {
      "action_query_batch_std": 0.3476191520690918,
      "action_query_norm": 40.78401260375976,
      "action_relation_loss": 0.002187184733338654,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.609915809167447,
      "flow_matching_loss": 0.017301785293966532,
      "grad_norm": 0.10170643031597137,
      "h_stock_valid_norm": 98.04324188232422,
      "invalid_tail_samples": 24.5,
      "learning_rate": 9.977742771459991e-06,
      "loss": 0.0195,
      "positive_teacher_pairs": 26699.1,
      "query_h_norm_ratio": 0.415983447432518,
      "relation_loss": 0.002187184733338654,
      "selected_pairs": 26699.1,
      "step": 24170,
      "student_distance_mean": 22.799064064025877,
      "teacher_distance_mean": 0.8507328987121582,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.002187184733338654,
      "weighted_relation_loss": 0.002187184733338654
    },
    {
      "action_query_batch_std": 0.3467151880264282,
      "action_query_norm": 40.788478469848634,
      "action_relation_loss": 0.0020696508814580738,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.61927034611787,
      "flow_matching_loss": 0.018180291447788478,
      "grad_norm": 0.0828692764043808,
      "h_stock_valid_norm": 98.09579544067383,
      "invalid_tail_samples": 21.1,
      "learning_rate": 9.944730435093802e-06,
      "loss": 0.0202,
      "positive_teacher_pairs": 27487.1,
      "query_h_norm_ratio": 0.4158034652471542,
      "relation_loss": 0.0020696508814580738,
      "selected_pairs": 27487.1,
      "step": 24180,
      "student_distance_mean": 22.867459106445313,
      "teacher_distance_mean": 0.8534013748168945,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.0020696508814580738,
      "weighted_relation_loss": 0.0020696508814580738
    },
    {
      "action_query_batch_std": 0.34763323366641996,
      "action_query_norm": 40.79763984680176,
      "action_relation_loss": 0.0021710295812226834,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.628624883068287,
      "flow_matching_loss": 0.01752804294228554,
      "grad_norm": 0.08384914696216583,
      "h_stock_valid_norm": 97.97075119018555,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.911766769657115e-06,
      "loss": 0.0197,
      "positive_teacher_pairs": 26993.2,
      "query_h_norm_ratio": 0.41643090546131134,
      "relation_loss": 0.0021710295812226834,
      "selected_pairs": 26993.2,
      "step": 24190,
      "student_distance_mean": 22.961390876770018,
      "teacher_distance_mean": 0.8552118599414825,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0021710295812226834,
      "weighted_relation_loss": 0.0021710295812226834
    },
    {
      "action_query_batch_std": 0.3508993059396744,
      "action_query_norm": 40.79710426330566,
      "action_relation_loss": 0.002154997014440596,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.63797942001871,
      "flow_matching_loss": 0.017286648880690335,
      "grad_norm": 0.08270809054374695,
      "h_stock_valid_norm": 98.04245223999024,
      "invalid_tail_samples": 22.8,
      "learning_rate": 9.878851815203882e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 27096.7,
      "query_h_norm_ratio": 0.41611859798431394,
      "relation_loss": 0.002154997014440596,
      "selected_pairs": 27096.7,
      "step": 24200,
      "student_distance_mean": 23.046519660949706,
      "teacher_distance_mean": 0.8588520467281342,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.002154997014440596,
      "weighted_relation_loss": 0.002154997014440596
    },
    {
      "action_query_batch_std": 0.3483246326446533,
      "action_query_norm": 40.78819198608399,
      "action_relation_loss": 0.002147616143338382,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.64733395696913,
      "flow_matching_loss": 0.017265864368528127,
      "grad_norm": 0.08303944021463394,
      "h_stock_valid_norm": 97.87439041137695,
      "invalid_tail_samples": 24.8,
      "learning_rate": 9.845985611728864e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 26617.4,
      "query_h_norm_ratio": 0.4167467415332794,
      "relation_loss": 0.002147616143338382,
      "selected_pairs": 26617.4,
      "step": 24210,
      "student_distance_mean": 22.9161735534668,
      "teacher_distance_mean": 0.8565591633319855,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.002147616143338382,
      "weighted_relation_loss": 0.002147616143338382
    },
    {
      "action_query_batch_std": 0.3504564255475998,
      "action_query_norm": 40.78449478149414,
      "action_relation_loss": 0.0021651130286045373,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.65668849391955,
      "flow_matching_loss": 0.019254656601697208,
      "grad_norm": 0.1608385443687439,
      "h_stock_valid_norm": 97.73444061279297,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.813168199167606e-06,
      "loss": 0.0214,
      "positive_teacher_pairs": 26825.5,
      "query_h_norm_ratio": 0.41730065941810607,
      "relation_loss": 0.0021651130286045373,
      "selected_pairs": 26825.5,
      "step": 24220,
      "student_distance_mean": 22.980266189575197,
      "teacher_distance_mean": 0.8566284716129303,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0021651130286045373,
      "weighted_relation_loss": 0.0021651130286045373
    },
    {
      "action_query_batch_std": 0.34744673371315005,
      "action_query_norm": 40.7797248840332,
      "action_relation_loss": 0.002035243425052613,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.66604303086997,
      "flow_matching_loss": 0.01957924831658602,
      "grad_norm": 0.09654366225004196,
      "h_stock_valid_norm": 97.87623138427735,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.780399617396341e-06,
      "loss": 0.0216,
      "positive_teacher_pairs": 26780.5,
      "query_h_norm_ratio": 0.41664727628231046,
      "relation_loss": 0.002035243425052613,
      "selected_pairs": 26780.5,
      "step": 24230,
      "student_distance_mean": 22.803497314453125,
      "teacher_distance_mean": 0.8502298176288605,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.002035243425052613,
      "weighted_relation_loss": 0.002035243425052613
    },
    {
      "action_query_batch_std": 0.3493572026491165,
      "action_query_norm": 40.76748771667481,
      "action_relation_loss": 0.0020903711556456983,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.675397567820394,
      "flow_matching_loss": 0.017971752025187016,
      "grad_norm": 0.09016891568899155,
      "h_stock_valid_norm": 97.74524154663087,
      "invalid_tail_samples": 23.7,
      "learning_rate": 9.747679906232016e-06,
      "loss": 0.0201,
      "positive_teacher_pairs": 26877.4,
      "query_h_norm_ratio": 0.41708196997642516,
      "relation_loss": 0.0020903711556456983,
      "selected_pairs": 26877.4,
      "step": 24240,
      "student_distance_mean": 22.928942489624024,
      "teacher_distance_mean": 0.8577432453632354,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0020903711556456983,
      "weighted_relation_loss": 0.0020903711556456983
    },
    {
      "action_query_batch_std": 0.3455043226480484,
      "action_query_norm": 40.73721694946289,
      "action_relation_loss": 0.002107298863120377,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.684752104770816,
      "flow_matching_loss": 0.017579436115920544,
      "grad_norm": 0.07746658474206924,
      "h_stock_valid_norm": 97.9000358581543,
      "invalid_tail_samples": 21.2,
      "learning_rate": 9.715009105432138e-06,
      "loss": 0.0197,
      "positive_teacher_pairs": 27467.2,
      "query_h_norm_ratio": 0.4161131829023361,
      "relation_loss": 0.002107298863120377,
      "selected_pairs": 27467.2,
      "step": 24250,
      "student_distance_mean": 22.677493476867674,
      "teacher_distance_mean": 0.8477605104446411,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.002107298863120377,
      "weighted_relation_loss": 0.002107298863120377
    },
    {
      "action_query_batch_std": 0.3529743760824203,
      "action_query_norm": 40.80659217834473,
      "action_relation_loss": 0.0020424593007192017,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.694106641721234,
      "flow_matching_loss": 0.017885181587189436,
      "grad_norm": 0.10172972828149796,
      "h_stock_valid_norm": 97.9221206665039,
      "invalid_tail_samples": 25.4,
      "learning_rate": 9.682387254694835e-06,
      "loss": 0.0199,
      "positive_teacher_pairs": 26482.2,
      "query_h_norm_ratio": 0.416726416349411,
      "relation_loss": 0.0020424593007192017,
      "selected_pairs": 26482.2,
      "step": 24260,
      "student_distance_mean": 23.172096061706544,
      "teacher_distance_mean": 0.8660654067993164,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0020424593007192017,
      "weighted_relation_loss": 0.0020424593007192017
    },
    {
      "action_query_batch_std": 0.3485968381166458,
      "action_query_norm": 40.765890502929686,
      "action_relation_loss": 0.0020637293113395573,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.703461178671656,
      "flow_matching_loss": 0.017938202898949384,
      "grad_norm": 0.1107143759727478,
      "h_stock_valid_norm": 97.88024291992187,
      "invalid_tail_samples": 25.2,
      "learning_rate": 9.649814393658723e-06,
      "loss": 0.02,
      "positive_teacher_pairs": 26526.5,
      "query_h_norm_ratio": 0.4164906919002533,
      "relation_loss": 0.0020637293113395573,
      "selected_pairs": 26526.5,
      "step": 24270,
      "student_distance_mean": 22.842038345336913,
      "teacher_distance_mean": 0.852258950471878,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0020637293113395573,
      "weighted_relation_loss": 0.0020637293113395573
    },
    {
      "action_query_batch_std": 0.34945483803749083,
      "action_query_norm": 40.79711723327637,
      "action_relation_loss": 0.002123286062851548,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.712815715622078,
      "flow_matching_loss": 0.018646195344626904,
      "grad_norm": 0.15158464014530182,
      "h_stock_valid_norm": 98.04828491210938,
      "invalid_tail_samples": 22.2,
      "learning_rate": 9.617290561902881e-06,
      "loss": 0.0208,
      "positive_teacher_pairs": 27218.8,
      "query_h_norm_ratio": 0.4160931080579758,
      "relation_loss": 0.002123286062851548,
      "selected_pairs": 27218.8,
      "step": 24280,
      "student_distance_mean": 22.989536094665528,
      "teacher_distance_mean": 0.8579918205738067,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.002123286062851548,
      "weighted_relation_loss": 0.002123286062851548
    },
    {
      "action_query_batch_std": 0.3466006815433502,
      "action_query_norm": 40.75594863891602,
      "action_relation_loss": 0.0020904571516439318,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.722170252572496,
      "flow_matching_loss": 0.018580698966979982,
      "grad_norm": 0.12260454893112183,
      "h_stock_valid_norm": 98.01770095825195,
      "invalid_tail_samples": 23.2,
      "learning_rate": 9.584815798946861e-06,
      "loss": 0.0207,
      "positive_teacher_pairs": 26989.9,
      "query_h_norm_ratio": 0.41580512523651125,
      "relation_loss": 0.0020904571516439318,
      "selected_pairs": 26989.9,
      "step": 24290,
      "student_distance_mean": 22.815884017944335,
      "teacher_distance_mean": 0.8526827692985535,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0020904571516439318,
      "weighted_relation_loss": 0.0020904571516439318
    },
    {
      "action_query_batch_std": 0.3490023612976074,
      "action_query_norm": 40.783233642578125,
      "action_relation_loss": 0.002072194789070636,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.73152478952292,
      "flow_matching_loss": 0.01584047554060817,
      "grad_norm": 0.1275133192539215,
      "h_stock_valid_norm": 97.85112686157227,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.55239014425055e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 26968.6,
      "query_h_norm_ratio": 0.4167889267206192,
      "relation_loss": 0.002072194789070636,
      "selected_pairs": 26968.6,
      "step": 24300,
      "student_distance_mean": 22.932421493530274,
      "teacher_distance_mean": 0.8578994631767273,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.002072194789070636,
      "weighted_relation_loss": 0.002072194789070636
    },
    {
      "action_query_batch_std": 0.34606834650039675,
      "action_query_norm": 40.74465713500977,
      "action_relation_loss": 0.0020887105958536266,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.74087932647334,
      "flow_matching_loss": 0.019021268095821144,
      "grad_norm": 0.1189136728644371,
      "h_stock_valid_norm": 98.04926071166992,
      "invalid_tail_samples": 23.6,
      "learning_rate": 9.520013637214176e-06,
      "loss": 0.0211,
      "positive_teacher_pairs": 26895.7,
      "query_h_norm_ratio": 0.4155567139387131,
      "relation_loss": 0.0020887105958536266,
      "selected_pairs": 26895.7,
      "step": 24310,
      "student_distance_mean": 22.731558990478515,
      "teacher_distance_mean": 0.8506321966648102,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0020887105958536266,
      "weighted_relation_loss": 0.0020887105958536266
    },
    {
      "action_query_batch_std": 0.3478266835212708,
      "action_query_norm": 40.773141860961914,
      "action_relation_loss": 0.0020914795226417483,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.75023386342376,
      "flow_matching_loss": 0.01710228994488716,
      "grad_norm": 0.09496574848890305,
      "h_stock_valid_norm": 98.12352676391602,
      "invalid_tail_samples": 21.1,
      "learning_rate": 9.48768631717824e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 27479.5,
      "query_h_norm_ratio": 0.4155302196741104,
      "relation_loss": 0.0020914795226417483,
      "selected_pairs": 27479.5,
      "step": 24320,
      "student_distance_mean": 22.71428108215332,
      "teacher_distance_mean": 0.849378091096878,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.0020914795226417483,
      "weighted_relation_loss": 0.0020914795226417483
    },
    {
      "action_query_batch_std": 0.347366127371788,
      "action_query_norm": 40.743910217285155,
      "action_relation_loss": 0.0020701287081465124,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.75958840037418,
      "flow_matching_loss": 0.01727978279814124,
      "grad_norm": 0.11441058665513992,
      "h_stock_valid_norm": 98.0903923034668,
      "invalid_tail_samples": 27.3,
      "learning_rate": 9.455408223423496e-06,
      "loss": 0.0193,
      "positive_teacher_pairs": 26052.3,
      "query_h_norm_ratio": 0.415372297167778,
      "relation_loss": 0.0020701287081465124,
      "selected_pairs": 26052.3,
      "step": 24330,
      "student_distance_mean": 22.892733764648437,
      "teacher_distance_mean": 0.8559616327285766,
      "valid_action_samples": 228.7,
      "weighted_auxiliary_loss": 0.0020701287081465124,
      "weighted_relation_loss": 0.0020701287081465124
    },
    {
      "action_query_batch_std": 0.3466757953166962,
      "action_query_norm": 40.77121696472168,
      "action_relation_loss": 0.00206946162506938,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.768942937324603,
      "flow_matching_loss": 0.019098660722374917,
      "grad_norm": 0.114744633436203,
      "h_stock_valid_norm": 98.09696426391602,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.423179395170845e-06,
      "loss": 0.0212,
      "positive_teacher_pairs": 26851.1,
      "query_h_norm_ratio": 0.4156242787837982,
      "relation_loss": 0.00206946162506938,
      "selected_pairs": 26851.1,
      "step": 24340,
      "student_distance_mean": 22.789191055297852,
      "teacher_distance_mean": 0.8511699616909028,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.00206946162506938,
      "weighted_relation_loss": 0.00206946162506938
    },
    {
      "action_query_batch_std": 0.350870418548584,
      "action_query_norm": 40.806864166259764,
      "action_relation_loss": 0.0020714620361104606,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.778297474275025,
      "flow_matching_loss": 0.01777997612953186,
      "grad_norm": 0.11060896515846252,
      "h_stock_valid_norm": 97.9223731994629,
      "invalid_tail_samples": 19.2,
      "learning_rate": 9.390999871581391e-06,
      "loss": 0.0199,
      "positive_teacher_pairs": 27927.1,
      "query_h_norm_ratio": 0.41672763228416443,
      "relation_loss": 0.0020714620361104606,
      "selected_pairs": 27927.1,
      "step": 24350,
      "student_distance_mean": 23.01049976348877,
      "teacher_distance_mean": 0.8596964716911316,
      "valid_action_samples": 236.8,
      "weighted_auxiliary_loss": 0.0020714620361104606,
      "weighted_relation_loss": 0.0020714620361104606
    },
    {
      "action_query_batch_std": 0.352052703499794,
      "action_query_norm": 40.82632675170898,
      "action_relation_loss": 0.00212076521711424,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.787652011225443,
      "flow_matching_loss": 0.018546319101005793,
      "grad_norm": 0.08198712021112442,
      "h_stock_valid_norm": 97.92401885986328,
      "invalid_tail_samples": 24.4,
      "learning_rate": 9.358869691756273e-06,
      "loss": 0.0207,
      "positive_teacher_pairs": 26719.0,
      "query_h_norm_ratio": 0.41692063212394714,
      "relation_loss": 0.00212076521711424,
      "selected_pairs": 26719.0,
      "step": 24360,
      "student_distance_mean": 23.02833309173584,
      "teacher_distance_mean": 0.8596248149871826,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.00212076521711424,
      "weighted_relation_loss": 0.00212076521711424
    },
    {
      "action_query_batch_std": 0.3446770519018173,
      "action_query_norm": 40.766302490234374,
      "action_relation_loss": 0.0021305480040609837,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.797006548175865,
      "flow_matching_loss": 0.017267544846981765,
      "grad_norm": 0.09254445880651474,
      "h_stock_valid_norm": 97.96540603637695,
      "invalid_tail_samples": 26.1,
      "learning_rate": 9.326788894736688e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 26320.1,
      "query_h_norm_ratio": 0.4161306768655777,
      "relation_loss": 0.0021305480040609837,
      "selected_pairs": 26320.1,
      "step": 24370,
      "student_distance_mean": 22.730103302001954,
      "teacher_distance_mean": 0.8488696813583374,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.0021305480040609837,
      "weighted_relation_loss": 0.0021305480040609837
    },
    {
      "action_query_batch_std": 0.3466774135828018,
      "action_query_norm": 40.78260192871094,
      "action_relation_loss": 0.002082586265169084,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.806361085126287,
      "flow_matching_loss": 0.01729603037238121,
      "grad_norm": 0.12889400124549866,
      "h_stock_valid_norm": 97.8546356201172,
      "invalid_tail_samples": 23.8,
      "learning_rate": 9.294757519503811e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 26846.6,
      "query_h_norm_ratio": 0.4167681485414505,
      "relation_loss": 0.002082586265169084,
      "selected_pairs": 26846.6,
      "step": 24380,
      "student_distance_mean": 22.827331352233887,
      "teacher_distance_mean": 0.8503986895084381,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.002082586265169084,
      "weighted_relation_loss": 0.002082586265169084
    },
    {
      "action_query_batch_std": 0.3450722604990005,
      "action_query_norm": 40.74889450073242,
      "action_relation_loss": 0.0020830529392696917,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.815715622076706,
      "flow_matching_loss": 0.016937302984297277,
      "grad_norm": 0.08525311946868896,
      "h_stock_valid_norm": 98.0740364074707,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.262775604978819e-06,
      "loss": 0.019,
      "positive_teacher_pairs": 26919.4,
      "query_h_norm_ratio": 0.415493506193161,
      "relation_loss": 0.0020830529392696917,
      "selected_pairs": 26919.4,
      "step": 24390,
      "student_distance_mean": 22.70676441192627,
      "teacher_distance_mean": 0.8490934491157531,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0020830529392696917,
      "weighted_relation_loss": 0.0020830529392696917
    },
    {
      "action_query_batch_std": 0.3483558475971222,
      "action_query_norm": 40.793164825439455,
      "action_relation_loss": 0.0021122679696418344,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.825070159027128,
      "flow_matching_loss": 0.01865677097812295,
      "grad_norm": 0.1038319319486618,
      "h_stock_valid_norm": 98.03370742797851,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.230843190022725e-06,
      "loss": 0.0208,
      "positive_teacher_pairs": 26739.8,
      "query_h_norm_ratio": 0.41611576080322266,
      "relation_loss": 0.0021122679696418344,
      "selected_pairs": 26739.8,
      "step": 24400,
      "student_distance_mean": 22.9426118850708,
      "teacher_distance_mean": 0.8556721270084381,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0021122679696418344,
      "weighted_relation_loss": 0.0021122679696418344
    },
    {
      "action_query_batch_std": 0.34475726187229155,
      "action_query_norm": 40.75646820068359,
      "action_relation_loss": 0.0021178033668547867,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.83442469597755,
      "flow_matching_loss": 0.01712898975238204,
      "grad_norm": 0.1054869294166565,
      "h_stock_valid_norm": 98.08187408447266,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.198960313436445e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 26754.1,
      "query_h_norm_ratio": 0.41553749740123747,
      "relation_loss": 0.0021178033668547867,
      "selected_pairs": 26754.1,
      "step": 24410,
      "student_distance_mean": 22.70933303833008,
      "teacher_distance_mean": 0.8483922421932221,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0021178033668547867,
      "weighted_relation_loss": 0.0021178033668547867
    },
    {
      "action_query_batch_std": 0.3533098191022873,
      "action_query_norm": 40.80668144226074,
      "action_relation_loss": 0.001989171782042831,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.84377923292797,
      "flow_matching_loss": 0.017099518980830907,
      "grad_norm": 0.0990842953324318,
      "h_stock_valid_norm": 98.29472045898437,
      "invalid_tail_samples": 22.1,
      "learning_rate": 9.16712701396067e-06,
      "loss": 0.0191,
      "positive_teacher_pairs": 27244.9,
      "query_h_norm_ratio": 0.41514859199523924,
      "relation_loss": 0.001989171782042831,
      "selected_pairs": 27244.9,
      "step": 24420,
      "student_distance_mean": 23.09694061279297,
      "teacher_distance_mean": 0.8618566691875458,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.001989171782042831,
      "weighted_relation_loss": 0.001989171782042831
    },
    {
      "action_query_batch_std": 0.3483207285404205,
      "action_query_norm": 40.76109313964844,
      "action_relation_loss": 0.0022237664088606834,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.85313376987839,
      "flow_matching_loss": 0.015592989977449179,
      "grad_norm": 0.10030663013458252,
      "h_stock_valid_norm": 97.94803619384766,
      "invalid_tail_samples": 24.3,
      "learning_rate": 9.135343330275864e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 26730.6,
      "query_h_norm_ratio": 0.4161576867103577,
      "relation_loss": 0.0022237664088606834,
      "selected_pairs": 26730.6,
      "step": 24430,
      "student_distance_mean": 22.89478130340576,
      "teacher_distance_mean": 0.8547436356544494,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0022237664088606834,
      "weighted_relation_loss": 0.0022237664088606834
    },
    {
      "action_query_batch_std": 0.34947572350502015,
      "action_query_norm": 40.76898384094238,
      "action_relation_loss": 0.0020729853655211627,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.862488306828812,
      "flow_matching_loss": 0.015876058861613272,
      "grad_norm": 0.13105541467666626,
      "h_stock_valid_norm": 98.05869598388672,
      "invalid_tail_samples": 22.3,
      "learning_rate": 9.10360930100218e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 27199.3,
      "query_h_norm_ratio": 0.4157626301050186,
      "relation_loss": 0.0020729853655211627,
      "selected_pairs": 27199.3,
      "step": 24440,
      "student_distance_mean": 22.84840621948242,
      "teacher_distance_mean": 0.8535149216651916,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0020729853655211627,
      "weighted_relation_loss": 0.0020729853655211627
    },
    {
      "action_query_batch_std": 0.35328505635261537,
      "action_query_norm": 40.81799850463867,
      "action_relation_loss": 0.002016133302822709,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.871842843779234,
      "flow_matching_loss": 0.017191296629607677,
      "grad_norm": 0.09765328466892242,
      "h_stock_valid_norm": 98.01180572509766,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.07192496469949e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 26649.6,
      "query_h_norm_ratio": 0.41646764278411863,
      "relation_loss": 0.002016133302822709,
      "selected_pairs": 26649.6,
      "step": 24450,
      "student_distance_mean": 23.15149269104004,
      "teacher_distance_mean": 0.8668935060501098,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.002016133302822709,
      "weighted_relation_loss": 0.002016133302822709
    },
    {
      "action_query_batch_std": 0.35283600687980654,
      "action_query_norm": 40.76839256286621,
      "action_relation_loss": 0.0020629003527574243,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.881197380729652,
      "flow_matching_loss": 0.016366266552358864,
      "grad_norm": 0.09530330449342728,
      "h_stock_valid_norm": 97.99501495361328,
      "invalid_tail_samples": 23.5,
      "learning_rate": 9.040290359867231e-06,
      "loss": 0.0184,
      "positive_teacher_pairs": 26925.1,
      "query_h_norm_ratio": 0.41602663397789,
      "relation_loss": 0.0020629003527574243,
      "selected_pairs": 26925.1,
      "step": 24460,
      "student_distance_mean": 23.216433143615724,
      "teacher_distance_mean": 0.8674599707126618,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0020629003527574243,
      "weighted_relation_loss": 0.0020629003527574243
    },
    {
      "action_query_batch_std": 0.3477292060852051,
      "action_query_norm": 40.771643447875974,
      "action_relation_loss": 0.00204258079174906,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.890551917680074,
      "flow_matching_loss": 0.018095082510262726,
      "grad_norm": 0.08494090288877487,
      "h_stock_valid_norm": 98.11907043457032,
      "invalid_tail_samples": 22.4,
      "learning_rate": 9.008705524944438e-06,
      "loss": 0.0201,
      "positive_teacher_pairs": 27174.9,
      "query_h_norm_ratio": 0.4155353128910065,
      "relation_loss": 0.00204258079174906,
      "selected_pairs": 27174.9,
      "step": 24470,
      "student_distance_mean": 22.820771980285645,
      "teacher_distance_mean": 0.8546174168586731,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.00204258079174906,
      "weighted_relation_loss": 0.00204258079174906
    },
    {
      "action_query_batch_std": 0.34240533113479615,
      "action_query_norm": 40.72547950744629,
      "action_relation_loss": 0.0020828594220802187,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.899906454630496,
      "flow_matching_loss": 0.017485632747411727,
      "grad_norm": 0.10019098222255707,
      "h_stock_valid_norm": 98.18506088256837,
      "invalid_tail_samples": 24.2,
      "learning_rate": 8.977170498309651e-06,
      "loss": 0.0196,
      "positive_teacher_pairs": 26758.1,
      "query_h_norm_ratio": 0.4147870600223541,
      "relation_loss": 0.0020828594220802187,
      "selected_pairs": 26758.1,
      "step": 24480,
      "student_distance_mean": 22.528042030334472,
      "teacher_distance_mean": 0.8423701405525208,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0020828594220802187,
      "weighted_relation_loss": 0.0020828594220802187
    },
    {
      "action_query_batch_std": 0.34448841512203215,
      "action_query_norm": 40.74551429748535,
      "action_relation_loss": 0.0021043709828518333,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.90926099158092,
      "flow_matching_loss": 0.01702535282820463,
      "grad_norm": 0.11495797336101532,
      "h_stock_valid_norm": 98.24905853271484,
      "invalid_tail_samples": 23.4,
      "learning_rate": 8.945685318280916e-06,
      "loss": 0.0191,
      "positive_teacher_pairs": 26942.2,
      "query_h_norm_ratio": 0.41472010016441346,
      "relation_loss": 0.0021043709828518333,
      "selected_pairs": 26942.2,
      "step": 24490,
      "student_distance_mean": 22.60946617126465,
      "teacher_distance_mean": 0.8451205134391785,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0021043709828518333,
      "weighted_relation_loss": 0.0021043709828518333
    },
    {
      "action_query_batch_std": 0.34732393026351926,
      "action_query_norm": 40.76275100708008,
      "action_relation_loss": 0.0021748880739323796,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.918615528531337,
      "flow_matching_loss": 0.01771827954798937,
      "grad_norm": 0.10794107615947723,
      "h_stock_valid_norm": 97.94320831298828,
      "invalid_tail_samples": 21.9,
      "learning_rate": 8.914250023115672e-06,
      "loss": 0.0199,
      "positive_teacher_pairs": 27293.2,
      "query_h_norm_ratio": 0.416191092133522,
      "relation_loss": 0.0021748880739323796,
      "selected_pairs": 27293.2,
      "step": 24500,
      "student_distance_mean": 22.822027015686036,
      "teacher_distance_mean": 0.8535805821418763,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.0021748880739323796,
      "weighted_relation_loss": 0.0021748880739323796
    },
    {
      "action_query_batch_std": 0.3498919397592545,
      "action_query_norm": 40.77208290100098,
      "action_relation_loss": 0.0020532696973532437,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.92797006548176,
      "flow_matching_loss": 0.015765600837767124,
      "grad_norm": 0.10149592906236649,
      "h_stock_valid_norm": 98.1176368713379,
      "invalid_tail_samples": 24.9,
      "learning_rate": 8.882864651010798e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 26600.5,
      "query_h_norm_ratio": 0.4155478924512863,
      "relation_loss": 0.0020532696973532437,
      "selected_pairs": 26600.5,
      "step": 24510,
      "student_distance_mean": 23.0004976272583,
      "teacher_distance_mean": 0.8592985391616821,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0020532696973532437,
      "weighted_relation_loss": 0.0020532696973532437
    },
    {
      "action_query_batch_std": 0.3466715842485428,
      "action_query_norm": 40.76272735595703,
      "action_relation_loss": 0.0020903391530737283,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.93732460243218,
      "flow_matching_loss": 0.017486527655273677,
      "grad_norm": 0.10467404127120972,
      "h_stock_valid_norm": 98.03298110961914,
      "invalid_tail_samples": 25.1,
      "learning_rate": 8.851529240102462e-06,
      "loss": 0.0196,
      "positive_teacher_pairs": 26551.2,
      "query_h_norm_ratio": 0.4158078104257584,
      "relation_loss": 0.0020903391530737283,
      "selected_pairs": 26551.2,
      "step": 24520,
      "student_distance_mean": 22.75035228729248,
      "teacher_distance_mean": 0.850022429227829,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0020903391530737283,
      "weighted_relation_loss": 0.0020903391530737283
    },
    {
      "action_query_batch_std": 0.3460367053747177,
      "action_query_norm": 40.73759994506836,
      "action_relation_loss": 0.0020390406367368995,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.9466791393826,
      "flow_matching_loss": 0.018156478367745876,
      "grad_norm": 0.1159542053937912,
      "h_stock_valid_norm": 98.11792526245117,
      "invalid_tail_samples": 24.8,
      "learning_rate": 8.820243828466134e-06,
      "loss": 0.0202,
      "positive_teacher_pairs": 26626.4,
      "query_h_norm_ratio": 0.4151920258998871,
      "relation_loss": 0.0020390406367368995,
      "selected_pairs": 26626.4,
      "step": 24530,
      "student_distance_mean": 22.76610622406006,
      "teacher_distance_mean": 0.8505248427391052,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0020390406367368995,
      "weighted_relation_loss": 0.0020390406367368995
    },
    {
      "action_query_batch_std": 0.35041622519493104,
      "action_query_norm": 40.782532119750975,
      "action_relation_loss": 0.0021804188610985873,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.95603367633302,
      "flow_matching_loss": 0.017160100396722555,
      "grad_norm": 0.10268943756818771,
      "h_stock_valid_norm": 98.05061416625976,
      "invalid_tail_samples": 22.6,
      "learning_rate": 8.789008454116566e-06,
      "loss": 0.0193,
      "positive_teacher_pairs": 27127.0,
      "query_h_norm_ratio": 0.4159388929605484,
      "relation_loss": 0.0021804188610985873,
      "selected_pairs": 27127.0,
      "step": 24540,
      "student_distance_mean": 22.980444526672365,
      "teacher_distance_mean": 0.8588224649429321,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.0021804188610985873,
      "weighted_relation_loss": 0.0021804188610985873
    },
    {
      "action_query_batch_std": 0.3468331038951874,
      "action_query_norm": 40.75528144836426,
      "action_relation_loss": 0.002142968901898712,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.965388213283443,
      "flow_matching_loss": 0.01726660141721368,
      "grad_norm": 0.09703025966882706,
      "h_stock_valid_norm": 97.83344955444336,
      "invalid_tail_samples": 25.0,
      "learning_rate": 8.757823155007655e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 26574.6,
      "query_h_norm_ratio": 0.4165796786546707,
      "relation_loss": 0.002142968901898712,
      "selected_pairs": 26574.6,
      "step": 24550,
      "student_distance_mean": 22.781289100646973,
      "teacher_distance_mean": 0.8544678926467896,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.002142968901898712,
      "weighted_relation_loss": 0.002142968901898712
    },
    {
      "action_query_batch_std": 0.34486712217330934,
      "action_query_norm": 40.77389259338379,
      "action_relation_loss": 0.0020587034989148377,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.974742750233865,
      "flow_matching_loss": 0.015173956099897624,
      "grad_norm": 0.09739833325147629,
      "h_stock_valid_norm": 97.91430130004883,
      "invalid_tail_samples": 23.3,
      "learning_rate": 8.72668796903251e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 26961.0,
      "query_h_norm_ratio": 0.4164296269416809,
      "relation_loss": 0.0020587034989148377,
      "selected_pairs": 26961.0,
      "step": 24560,
      "student_distance_mean": 22.646388435363768,
      "teacher_distance_mean": 0.8486080408096314,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0020587034989148377,
      "weighted_relation_loss": 0.0020587034989148377
    },
    {
      "action_query_batch_std": 0.3452453941106796,
      "action_query_norm": 40.750105667114255,
      "action_relation_loss": 0.0021316677564755084,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.984097287184284,
      "flow_matching_loss": 0.017454800102859734,
      "grad_norm": 0.09415741264820099,
      "h_stock_valid_norm": 98.01689376831055,
      "invalid_tail_samples": 25.1,
      "learning_rate": 8.695602934023301e-06,
      "loss": 0.0196,
      "positive_teacher_pairs": 26544.5,
      "query_h_norm_ratio": 0.41574729681015016,
      "relation_loss": 0.0021316677564755084,
      "selected_pairs": 26544.5,
      "step": 24570,
      "student_distance_mean": 22.708436965942383,
      "teacher_distance_mean": 0.8498486697673797,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0021316677564755084,
      "weighted_relation_loss": 0.0021316677564755084
    },
    {
      "action_query_batch_std": 0.3494799882173538,
      "action_query_norm": 40.7651424407959,
      "action_relation_loss": 0.0021281092660501598,
      "auxiliary_loss_weight": 1.0,
      "epoch": 22.993451824134706,
      "flow_matching_loss": 0.01579985162243247,
      "grad_norm": 0.10585572570562363,
      "h_stock_valid_norm": 98.07760848999024,
      "invalid_tail_samples": 25.8,
      "learning_rate": 8.664568087751274e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 26387.5,
      "query_h_norm_ratio": 0.4156438231468201,
      "relation_loss": 0.0021281092660501598,
      "selected_pairs": 26387.5,
      "step": 24580,
      "student_distance_mean": 22.962254905700682,
      "teacher_distance_mean": 0.8596825420856475,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0021281092660501598,
      "weighted_relation_loss": 0.0021281092660501598
    },
    {
      "action_query_batch_std": 0.348647403717041,
      "action_query_norm": 40.7900749206543,
      "action_relation_loss": 0.0020171675249002876,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.002806361085128,
      "flow_matching_loss": 0.017588116135448216,
      "grad_norm": 0.09333271533250809,
      "h_stock_valid_norm": 98.07856674194336,
      "invalid_tail_samples": 21.4,
      "learning_rate": 8.633583467926697e-06,
      "loss": 0.0196,
      "positive_teacher_pairs": 24489.3,
      "query_h_norm_ratio": 0.4158943474292755,
      "relation_loss": 0.0020171675249002876,
      "selected_pairs": 24489.3,
      "step": 24590,
      "student_distance_mean": 23.106528663635252,
      "teacher_distance_mean": 0.8614144921302795,
      "valid_action_samples": 214.7,
      "weighted_auxiliary_loss": 0.0020171675249002876,
      "weighted_relation_loss": 0.0020171675249002876
    },
    {
      "action_query_batch_std": 0.34946068823337556,
      "action_query_norm": 40.77101211547851,
      "action_relation_loss": 0.002041423064656556,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.012160898035546,
      "flow_matching_loss": 0.01776952464133501,
      "grad_norm": 0.07504939287900925,
      "h_stock_valid_norm": 97.85028305053712,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.602649112198796e-06,
      "loss": 0.0198,
      "positive_teacher_pairs": 26872.9,
      "query_h_norm_ratio": 0.41666948795318604,
      "relation_loss": 0.002041423064656556,
      "selected_pairs": 26872.9,
      "step": 24600,
      "student_distance_mean": 22.893512153625487,
      "teacher_distance_mean": 0.8546447157859802,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.002041423064656556,
      "weighted_relation_loss": 0.002041423064656556
    },
    {
      "action_query_batch_std": 0.34871854782104494,
      "action_query_norm": 40.781058883666994,
      "action_relation_loss": 0.0020744840148836374,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.021515434985968,
      "flow_matching_loss": 0.01821792433038354,
      "grad_norm": 0.1410948932170868,
      "h_stock_valid_norm": 97.7973876953125,
      "invalid_tail_samples": 23.5,
      "learning_rate": 8.571765058155745e-06,
      "loss": 0.0203,
      "positive_teacher_pairs": 26931.1,
      "query_h_norm_ratio": 0.4169978141784668,
      "relation_loss": 0.0020744840148836374,
      "selected_pairs": 26931.1,
      "step": 24610,
      "student_distance_mean": 22.889016342163085,
      "teacher_distance_mean": 0.8547844529151917,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0020744840148836374,
      "weighted_relation_loss": 0.0020744840148836374
    },
    {
      "action_query_batch_std": 0.3441041558980942,
      "action_query_norm": 40.72617378234863,
      "action_relation_loss": 0.0020799717167392373,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.03086997193639,
      "flow_matching_loss": 0.016985157504677773,
      "grad_norm": 0.11679089069366455,
      "h_stock_valid_norm": 97.48934783935547,
      "invalid_tail_samples": 23.2,
      "learning_rate": 8.540931343324583e-06,
      "loss": 0.0191,
      "positive_teacher_pairs": 26990.8,
      "query_h_norm_ratio": 0.4177521139383316,
      "relation_loss": 0.0020799717167392373,
      "selected_pairs": 26990.8,
      "step": 24620,
      "student_distance_mean": 22.639284896850587,
      "teacher_distance_mean": 0.8456320583820343,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0020799717167392373,
      "weighted_relation_loss": 0.0020799717167392373
    },
    {
      "action_query_batch_std": 0.34869123697280885,
      "action_query_norm": 40.79165115356445,
      "action_relation_loss": 0.0020393873914144934,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.04022450888681,
      "flow_matching_loss": 0.01707688020542264,
      "grad_norm": 0.09566428512334824,
      "h_stock_valid_norm": 97.59707336425781,
      "invalid_tail_samples": 24.0,
      "learning_rate": 8.51014800517117e-06,
      "loss": 0.0191,
      "positive_teacher_pairs": 26803.9,
      "query_h_norm_ratio": 0.41796507239341735,
      "relation_loss": 0.0020393873914144934,
      "selected_pairs": 26803.9,
      "step": 24630,
      "student_distance_mean": 22.98780059814453,
      "teacher_distance_mean": 0.8587874352931977,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0020393873914144934,
      "weighted_relation_loss": 0.0020393873914144934
    },
    {
      "action_query_batch_std": 0.34896980226039886,
      "action_query_norm": 40.777880859375,
      "action_relation_loss": 0.002038756152614951,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.04957904583723,
      "flow_matching_loss": 0.01575220385566354,
      "grad_norm": 0.09679616242647171,
      "h_stock_valid_norm": 97.5096824645996,
      "invalid_tail_samples": 21.9,
      "learning_rate": 8.479415081100167e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 27299.5,
      "query_h_norm_ratio": 0.4181956976652145,
      "relation_loss": 0.002038756152614951,
      "selected_pairs": 27299.5,
      "step": 24640,
      "student_distance_mean": 22.917684936523436,
      "teacher_distance_mean": 0.8555600583553314,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.002038756152614951,
      "weighted_relation_loss": 0.002038756152614951
    },
    {
      "action_query_batch_std": 0.3485812693834305,
      "action_query_norm": 40.76556129455567,
      "action_relation_loss": 0.002082550199702382,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.058933582787652,
      "flow_matching_loss": 0.017819190677255393,
      "grad_norm": 0.13774143159389496,
      "h_stock_valid_norm": 97.47199783325195,
      "invalid_tail_samples": 22.6,
      "learning_rate": 8.448732608454967e-06,
      "loss": 0.0199,
      "positive_teacher_pairs": 27127.9,
      "query_h_norm_ratio": 0.41823089122772217,
      "relation_loss": 0.002082550199702382,
      "selected_pairs": 27127.9,
      "step": 24650,
      "student_distance_mean": 23.020295715332033,
      "teacher_distance_mean": 0.8605577051639557,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.002082550199702382,
      "weighted_relation_loss": 0.002082550199702382
    },
    {
      "action_query_batch_std": 0.34223687648773193,
      "action_query_norm": 40.697056198120116,
      "action_relation_loss": 0.0021121171885170043,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.068288119738074,
      "flow_matching_loss": 0.01710210870951414,
      "grad_norm": 0.12551350891590118,
      "h_stock_valid_norm": 97.70117797851563,
      "invalid_tail_samples": 24.7,
      "learning_rate": 8.418100624517688e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 26635.6,
      "query_h_norm_ratio": 0.4165483206510544,
      "relation_loss": 0.0021121171885170043,
      "selected_pairs": 26635.6,
      "step": 24660,
      "student_distance_mean": 22.456073760986328,
      "teacher_distance_mean": 0.838535475730896,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0021121171885170043,
      "weighted_relation_loss": 0.0021121171885170043
    },
    {
      "action_query_batch_std": 0.34558612704277036,
      "action_query_norm": 40.72804336547851,
      "action_relation_loss": 0.002117391733918339,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.077642656688493,
      "flow_matching_loss": 0.01763462955132127,
      "grad_norm": 0.10127682983875275,
      "h_stock_valid_norm": 97.82369995117188,
      "invalid_tail_samples": 24.0,
      "learning_rate": 8.387519166509062e-06,
      "loss": 0.0198,
      "positive_teacher_pairs": 26813.1,
      "query_h_norm_ratio": 0.4163457900285721,
      "relation_loss": 0.002117391733918339,
      "selected_pairs": 26813.1,
      "step": 24670,
      "student_distance_mean": 22.786998748779297,
      "teacher_distance_mean": 0.853293526172638,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.002117391733918339,
      "weighted_relation_loss": 0.002117391733918339
    },
    {
      "action_query_batch_std": 0.346370142698288,
      "action_query_norm": 40.77567596435547,
      "action_relation_loss": 0.0020492708310484884,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.086997193638915,
      "flow_matching_loss": 0.01610864754766226,
      "grad_norm": 0.1319095343351364,
      "h_stock_valid_norm": 97.9210220336914,
      "invalid_tail_samples": 22.2,
      "learning_rate": 8.356988271588445e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 27224.4,
      "query_h_norm_ratio": 0.41641653776168824,
      "relation_loss": 0.0020492708310484884,
      "selected_pairs": 27224.4,
      "step": 24680,
      "student_distance_mean": 22.754622650146484,
      "teacher_distance_mean": 0.850665706396103,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0020492708310484884,
      "weighted_relation_loss": 0.0020492708310484884
    },
    {
      "action_query_batch_std": 0.35110002756118774,
      "action_query_norm": 40.764462280273435,
      "action_relation_loss": 0.0019527132040821015,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.096351730589337,
      "flow_matching_loss": 0.017975630331784488,
      "grad_norm": 0.1013169065117836,
      "h_stock_valid_norm": 98.00078048706055,
      "invalid_tail_samples": 25.4,
      "learning_rate": 8.326507976853765e-06,
      "loss": 0.0199,
      "positive_teacher_pairs": 26491.7,
      "query_h_norm_ratio": 0.41596412658691406,
      "relation_loss": 0.0019527132040821015,
      "selected_pairs": 26491.7,
      "step": 24690,
      "student_distance_mean": 22.97237777709961,
      "teacher_distance_mean": 0.8608135402202606,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0019527132040821015,
      "weighted_relation_loss": 0.0019527132040821015
    },
    {
      "action_query_batch_std": 0.3474987894296646,
      "action_query_norm": 40.73275299072266,
      "action_relation_loss": 0.0020795765449292958,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.105706267539755,
      "flow_matching_loss": 0.016500148735940457,
      "grad_norm": 0.11660478264093399,
      "h_stock_valid_norm": 97.81790390014649,
      "invalid_tail_samples": 22.7,
      "learning_rate": 8.296078319341443e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 27105.7,
      "query_h_norm_ratio": 0.4164156913757324,
      "relation_loss": 0.0020795765449292958,
      "selected_pairs": 27105.7,
      "step": 24700,
      "student_distance_mean": 22.667393302917482,
      "teacher_distance_mean": 0.8480458498001099,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0020795765449292958,
      "weighted_relation_loss": 0.0020795765449292958
    },
    {
      "action_query_batch_std": 0.35180575847625734,
      "action_query_norm": 40.77725448608398,
      "action_relation_loss": 0.002011965401470661,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.115060804490177,
      "flow_matching_loss": 0.016140114422887562,
      "grad_norm": 0.08449000120162964,
      "h_stock_valid_norm": 97.82626190185547,
      "invalid_tail_samples": 23.3,
      "learning_rate": 8.265699336026383e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 26964.4,
      "query_h_norm_ratio": 0.4168358385562897,
      "relation_loss": 0.002011965401470661,
      "selected_pairs": 26964.4,
      "step": 24710,
      "student_distance_mean": 23.07126941680908,
      "teacher_distance_mean": 0.8621327579021454,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.002011965401470661,
      "weighted_relation_loss": 0.002011965401470661
    },
    {
      "action_query_batch_std": 0.34771236777305603,
      "action_query_norm": 40.7476619720459,
      "action_relation_loss": 0.0021140942350029945,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.1244153414406,
      "flow_matching_loss": 0.016877587139606475,
      "grad_norm": 0.15506285429000854,
      "h_stock_valid_norm": 97.82284698486328,
      "invalid_tail_samples": 23.5,
      "learning_rate": 8.235371063821923e-06,
      "loss": 0.019,
      "positive_teacher_pairs": 26925.7,
      "query_h_norm_ratio": 0.41654608845710756,
      "relation_loss": 0.0021140942350029945,
      "selected_pairs": 26925.7,
      "step": 24720,
      "student_distance_mean": 22.772105598449706,
      "teacher_distance_mean": 0.8507194697856904,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0021140942350029945,
      "weighted_relation_loss": 0.0021140942350029945
    },
    {
      "action_query_batch_std": 0.3468026667833328,
      "action_query_norm": 40.72882347106933,
      "action_relation_loss": 0.002062406961340457,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.13376987839102,
      "flow_matching_loss": 0.016762857884168626,
      "grad_norm": 0.08367954194545746,
      "h_stock_valid_norm": 97.94164276123047,
      "invalid_tail_samples": 22.8,
      "learning_rate": 8.205093539579767e-06,
      "loss": 0.0188,
      "positive_teacher_pairs": 27083.4,
      "query_h_norm_ratio": 0.4158536732196808,
      "relation_loss": 0.002062406961340457,
      "selected_pairs": 27083.4,
      "step": 24730,
      "student_distance_mean": 22.7744140625,
      "teacher_distance_mean": 0.8512308239936829,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.002062406961340457,
      "weighted_relation_loss": 0.002062406961340457
    },
    {
      "action_query_batch_std": 0.34847443699836733,
      "action_query_norm": 40.772453689575194,
      "action_relation_loss": 0.001985512860119343,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.14312441534144,
      "flow_matching_loss": 0.01800412777811289,
      "grad_norm": 0.09317531436681747,
      "h_stock_valid_norm": 97.90293045043946,
      "invalid_tail_samples": 22.6,
      "learning_rate": 8.174866800089963e-06,
      "loss": 0.02,
      "positive_teacher_pairs": 27128.5,
      "query_h_norm_ratio": 0.416462641954422,
      "relation_loss": 0.001985512860119343,
      "selected_pairs": 27128.5,
      "step": 24740,
      "student_distance_mean": 22.86556510925293,
      "teacher_distance_mean": 0.855720865726471,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.001985512860119343,
      "weighted_relation_loss": 0.001985512860119343
    },
    {
      "action_query_batch_std": 0.35014876127243044,
      "action_query_norm": 40.77800178527832,
      "action_relation_loss": 0.0020214877906255422,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.15247895229186,
      "flow_matching_loss": 0.01704214308410883,
      "grad_norm": 0.09782521426677704,
      "h_stock_valid_norm": 97.94093246459961,
      "invalid_tail_samples": 23.4,
      "learning_rate": 8.144690882080853e-06,
      "loss": 0.0191,
      "positive_teacher_pairs": 26945.2,
      "query_h_norm_ratio": 0.41635439693927767,
      "relation_loss": 0.0020214877906255422,
      "selected_pairs": 26945.2,
      "step": 24750,
      "student_distance_mean": 23.04148635864258,
      "teacher_distance_mean": 0.8614963114261627,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0020214877906255422,
      "weighted_relation_loss": 0.0020214877906255422
    },
    {
      "action_query_batch_std": 0.3478027552366257,
      "action_query_norm": 40.73755111694336,
      "action_relation_loss": 0.002005912933964282,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.161833489242284,
      "flow_matching_loss": 0.017171171866357325,
      "grad_norm": 0.09745035320520401,
      "h_stock_valid_norm": 97.97151336669921,
      "invalid_tail_samples": 25.5,
      "learning_rate": 8.114565822219006e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 26463.0,
      "query_h_norm_ratio": 0.41581220328807833,
      "relation_loss": 0.002005912933964282,
      "selected_pairs": 26463.0,
      "step": 24760,
      "student_distance_mean": 22.721972465515137,
      "teacher_distance_mean": 0.8527036786079407,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.002005912933964282,
      "weighted_relation_loss": 0.002005912933964282
    },
    {
      "action_query_batch_std": 0.34885447323322294,
      "action_query_norm": 40.752837371826175,
      "action_relation_loss": 0.0020286345272324978,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.171188026192702,
      "flow_matching_loss": 0.016557225864380597,
      "grad_norm": 0.12638716399669647,
      "h_stock_valid_norm": 97.79086685180664,
      "invalid_tail_samples": 23.7,
      "learning_rate": 8.084491657109233e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 26872.1,
      "query_h_norm_ratio": 0.4167383223772049,
      "relation_loss": 0.0020286345272324978,
      "selected_pairs": 26872.1,
      "step": 24770,
      "student_distance_mean": 22.900328826904296,
      "teacher_distance_mean": 0.8588464856147766,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0020286345272324978,
      "weighted_relation_loss": 0.0020286345272324978
    },
    {
      "action_query_batch_std": 0.343980610370636,
      "action_query_norm": 40.73369789123535,
      "action_relation_loss": 0.0020244539133273064,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.180542563143124,
      "flow_matching_loss": 0.016384244617074727,
      "grad_norm": 0.09734518080949783,
      "h_stock_valid_norm": 97.86199951171875,
      "invalid_tail_samples": 26.3,
      "learning_rate": 8.054468423294476e-06,
      "loss": 0.0184,
      "positive_teacher_pairs": 26274.3,
      "query_h_norm_ratio": 0.4162383019924164,
      "relation_loss": 0.0020244539133273064,
      "selected_pairs": 26274.3,
      "step": 24780,
      "student_distance_mean": 22.590379524230958,
      "teacher_distance_mean": 0.8449669897556304,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.0020244539133273064,
      "weighted_relation_loss": 0.0020244539133273064
    },
    {
      "action_query_batch_std": 0.3502886474132538,
      "action_query_norm": 40.735151290893555,
      "action_relation_loss": 0.0019815304200164973,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.189897100093546,
      "flow_matching_loss": 0.015329010784626007,
      "grad_norm": 0.10362957417964935,
      "h_stock_valid_norm": 97.94108963012695,
      "invalid_tail_samples": 24.6,
      "learning_rate": 8.024496157255785e-06,
      "loss": 0.0173,
      "positive_teacher_pairs": 26669.2,
      "query_h_norm_ratio": 0.41591823697090147,
      "relation_loss": 0.0019815304200164973,
      "selected_pairs": 26669.2,
      "step": 24790,
      "student_distance_mean": 22.994124603271484,
      "teacher_distance_mean": 0.8604507565498352,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0019815304200164973,
      "weighted_relation_loss": 0.0019815304200164973
    },
    {
      "action_query_batch_std": 0.34613523483276365,
      "action_query_norm": 40.73296966552734,
      "action_relation_loss": 0.002055450121406466,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.199251637043965,
      "flow_matching_loss": 0.01764195170253515,
      "grad_norm": 0.08876968920230865,
      "h_stock_valid_norm": 97.95620651245117,
      "invalid_tail_samples": 24.5,
      "learning_rate": 7.994574895412293e-06,
      "loss": 0.0197,
      "positive_teacher_pairs": 26687.5,
      "query_h_norm_ratio": 0.41582997143268585,
      "relation_loss": 0.002055450121406466,
      "selected_pairs": 26687.5,
      "step": 24800,
      "student_distance_mean": 22.800552558898925,
      "teacher_distance_mean": 0.8533102333545685,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.002055450121406466,
      "weighted_relation_loss": 0.002055450121406466
    },
    {
      "action_query_batch_std": 0.34782259464263915,
      "action_query_norm": 40.75630493164063,
      "action_relation_loss": 0.002027591783553362,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.208606173994387,
      "flow_matching_loss": 0.017509107198566198,
      "grad_norm": 0.10109513998031616,
      "h_stock_valid_norm": 98.03141784667969,
      "invalid_tail_samples": 24.0,
      "learning_rate": 7.964704674121149e-06,
      "loss": 0.0195,
      "positive_teacher_pairs": 26806.3,
      "query_h_norm_ratio": 0.415750190615654,
      "relation_loss": 0.002027591783553362,
      "selected_pairs": 26806.3,
      "step": 24810,
      "student_distance_mean": 22.93185329437256,
      "teacher_distance_mean": 0.8585955798625946,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.002027591783553362,
      "weighted_relation_loss": 0.002027591783553362
    },
    {
      "action_query_batch_std": 0.3517557144165039,
      "action_query_norm": 40.772509002685545,
      "action_relation_loss": 0.002007829525973648,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.21796071094481,
      "flow_matching_loss": 0.01611957075074315,
      "grad_norm": 0.082331083714962,
      "h_stock_valid_norm": 98.11890411376953,
      "invalid_tail_samples": 24.2,
      "learning_rate": 7.934885529677505e-06,
      "loss": 0.0181,
      "positive_teacher_pairs": 26762.7,
      "query_h_norm_ratio": 0.41554428040981295,
      "relation_loss": 0.002007829525973648,
      "selected_pairs": 26762.7,
      "step": 24820,
      "student_distance_mean": 23.120599365234376,
      "teacher_distance_mean": 0.8659135282039643,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.002007829525973648,
      "weighted_relation_loss": 0.002007829525973648
    },
    {
      "action_query_batch_std": 0.3461989969015121,
      "action_query_norm": 40.75299224853516,
      "action_relation_loss": 0.0020581895252689717,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.22731524789523,
      "flow_matching_loss": 0.01883517885580659,
      "grad_norm": 0.08120597153902054,
      "h_stock_valid_norm": 98.11047439575195,
      "invalid_tail_samples": 24.5,
      "learning_rate": 7.905117498314412e-06,
      "loss": 0.0209,
      "positive_teacher_pairs": 26683.5,
      "query_h_norm_ratio": 0.4153804838657379,
      "relation_loss": 0.0020581895252689717,
      "selected_pairs": 26683.5,
      "step": 24830,
      "student_distance_mean": 22.75551700592041,
      "teacher_distance_mean": 0.8521635830402374,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0020581895252689717,
      "weighted_relation_loss": 0.0020581895252689717
    },
    {
      "action_query_batch_std": 0.3476424187421799,
      "action_query_norm": 40.76216201782226,
      "action_relation_loss": 0.0020283070276491343,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.23666978484565,
      "flow_matching_loss": 0.01689100619405508,
      "grad_norm": 0.14037416875362396,
      "h_stock_valid_norm": 98.16100845336913,
      "invalid_tail_samples": 22.2,
      "learning_rate": 7.87540061620286e-06,
      "loss": 0.0189,
      "positive_teacher_pairs": 27226.3,
      "query_h_norm_ratio": 0.4152607411146164,
      "relation_loss": 0.0020283070276491343,
      "selected_pairs": 27226.3,
      "step": 24840,
      "student_distance_mean": 22.803909683227538,
      "teacher_distance_mean": 0.8538430154323577,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0020283070276491343,
      "weighted_relation_loss": 0.0020283070276491343
    },
    {
      "action_query_batch_std": 0.3472156047821045,
      "action_query_norm": 40.75777816772461,
      "action_relation_loss": 0.0020439678337424994,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.24602432179607,
      "flow_matching_loss": 0.017643998842686415,
      "grad_norm": 0.09651478379964828,
      "h_stock_valid_norm": 98.11901473999023,
      "invalid_tail_samples": 24.5,
      "learning_rate": 7.845734919451647e-06,
      "loss": 0.0197,
      "positive_teacher_pairs": 26690.9,
      "query_h_norm_ratio": 0.41539360880851744,
      "relation_loss": 0.0020439678337424994,
      "selected_pairs": 26690.9,
      "step": 24850,
      "student_distance_mean": 22.830424308776855,
      "teacher_distance_mean": 0.8543524026870728,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0020439678337424994,
      "weighted_relation_loss": 0.0020439678337424994
    },
    {
      "action_query_batch_std": 0.3472936302423477,
      "action_query_norm": 40.73690681457519,
      "action_relation_loss": 0.002155410498380661,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.255378858746493,
      "flow_matching_loss": 0.016024164296686648,
      "grad_norm": 0.11114592105150223,
      "h_stock_valid_norm": 97.94485855102539,
      "invalid_tail_samples": 25.1,
      "learning_rate": 7.816120444107383e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 26548.6,
      "query_h_norm_ratio": 0.41592179238796234,
      "relation_loss": 0.002155410498380661,
      "selected_pairs": 26548.6,
      "step": 24860,
      "student_distance_mean": 22.92107219696045,
      "teacher_distance_mean": 0.8571955502033234,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.002155410498380661,
      "weighted_relation_loss": 0.002155410498380661
    },
    {
      "action_query_batch_std": 0.34687656760215757,
      "action_query_norm": 40.740461349487305,
      "action_relation_loss": 0.0020379848778247834,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.26473339569691,
      "flow_matching_loss": 0.01724417684599757,
      "grad_norm": 0.09939008206129074,
      "h_stock_valid_norm": 98.02946243286132,
      "invalid_tail_samples": 24.1,
      "learning_rate": 7.786557226154473e-06,
      "loss": 0.0193,
      "positive_teacher_pairs": 26778.5,
      "query_h_norm_ratio": 0.41559688150882723,
      "relation_loss": 0.0020379848778247834,
      "selected_pairs": 26778.5,
      "step": 24870,
      "student_distance_mean": 22.844051361083984,
      "teacher_distance_mean": 0.8541529655456543,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0020379848778247834,
      "weighted_relation_loss": 0.0020379848778247834
    },
    {
      "action_query_batch_std": 0.34512803256511687,
      "action_query_norm": 40.73731460571289,
      "action_relation_loss": 0.0020845553488470616,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.274087932647333,
      "flow_matching_loss": 0.018589724879711868,
      "grad_norm": 0.1051168143749237,
      "h_stock_valid_norm": 97.83503036499023,
      "invalid_tail_samples": 26.4,
      "learning_rate": 7.757045301514998e-06,
      "loss": 0.0207,
      "positive_teacher_pairs": 26249.9,
      "query_h_norm_ratio": 0.41638941764831544,
      "relation_loss": 0.0020845553488470616,
      "selected_pairs": 26249.9,
      "step": 24880,
      "student_distance_mean": 22.630630874633788,
      "teacher_distance_mean": 0.8475408613681793,
      "valid_action_samples": 229.6,
      "weighted_auxiliary_loss": 0.0020845553488470616,
      "weighted_relation_loss": 0.0020845553488470616
    },
    {
      "action_query_batch_std": 0.34633457064628603,
      "action_query_norm": 40.732791137695315,
      "action_relation_loss": 0.0020776102202944456,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.283442469597755,
      "flow_matching_loss": 0.0167292264290154,
      "grad_norm": 0.1756746768951416,
      "h_stock_valid_norm": 98.02664947509766,
      "invalid_tail_samples": 23.8,
      "learning_rate": 7.727584706048735e-06,
      "loss": 0.0188,
      "positive_teacher_pairs": 26849.3,
      "query_h_norm_ratio": 0.415532311797142,
      "relation_loss": 0.0020776102202944456,
      "selected_pairs": 26849.3,
      "step": 24890,
      "student_distance_mean": 22.787127876281737,
      "teacher_distance_mean": 0.8540769279003143,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0020776102202944456,
      "weighted_relation_loss": 0.0020776102202944456
    },
    {
      "action_query_batch_std": 0.34551896452903746,
      "action_query_norm": 40.7384635925293,
      "action_relation_loss": 0.0020078653236851097,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.292797006548177,
      "flow_matching_loss": 0.017597032990306617,
      "grad_norm": 0.1370079517364502,
      "h_stock_valid_norm": 97.77407455444336,
      "invalid_tail_samples": 24.1,
      "learning_rate": 7.698175475553077e-06,
      "loss": 0.0196,
      "positive_teacher_pairs": 26786.7,
      "query_h_norm_ratio": 0.41666240692138673,
      "relation_loss": 0.0020078653236851097,
      "selected_pairs": 26786.7,
      "step": 24900,
      "student_distance_mean": 22.718737411499024,
      "teacher_distance_mean": 0.854044783115387,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0020078653236851097,
      "weighted_relation_loss": 0.0020078653236851097
    },
    {
      "action_query_batch_std": 0.35079931020736693,
      "action_query_norm": 40.79915657043457,
      "action_relation_loss": 0.0020266053499653934,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.302151543498596,
      "flow_matching_loss": 0.018419349659234287,
      "grad_norm": 0.09546174108982086,
      "h_stock_valid_norm": 97.95435943603516,
      "invalid_tail_samples": 22.3,
      "learning_rate": 7.668817645763021e-06,
      "loss": 0.0204,
      "positive_teacher_pairs": 27193.4,
      "query_h_norm_ratio": 0.41651511490345,
      "relation_loss": 0.0020266053499653934,
      "selected_pairs": 27193.4,
      "step": 24910,
      "student_distance_mean": 23.010663986206055,
      "teacher_distance_mean": 0.8632689654827118,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0020266053499653934,
      "weighted_relation_loss": 0.0020266053499653934
    },
    {
      "action_query_batch_std": 0.3473410904407501,
      "action_query_norm": 40.752133178710935,
      "action_relation_loss": 0.0020683595910668375,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.311506080449018,
      "flow_matching_loss": 0.017906177695840596,
      "grad_norm": 0.09005887061357498,
      "h_stock_valid_norm": 97.85909423828124,
      "invalid_tail_samples": 24.0,
      "learning_rate": 7.639511252351088e-06,
      "loss": 0.02,
      "positive_teacher_pairs": 26798.3,
      "query_h_norm_ratio": 0.41643805503845216,
      "relation_loss": 0.0020683595910668375,
      "selected_pairs": 26798.3,
      "step": 24920,
      "student_distance_mean": 22.82270202636719,
      "teacher_distance_mean": 0.8541198790073394,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0020683595910668375,
      "weighted_relation_loss": 0.0020683595910668375
    },
    {
      "action_query_batch_std": 0.34884867668151853,
      "action_query_norm": 40.74888038635254,
      "action_relation_loss": 0.00207964078290388,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.32086061739944,
      "flow_matching_loss": 0.01803843742236495,
      "grad_norm": 0.10810409486293793,
      "h_stock_valid_norm": 98.00203247070313,
      "invalid_tail_samples": 24.5,
      "learning_rate": 7.610256330927324e-06,
      "loss": 0.0201,
      "positive_teacher_pairs": 26691.1,
      "query_h_norm_ratio": 0.41579863131046296,
      "relation_loss": 0.00207964078290388,
      "selected_pairs": 26691.1,
      "step": 24930,
      "student_distance_mean": 22.93487377166748,
      "teacher_distance_mean": 0.8558902740478516,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.00207964078290388,
      "weighted_relation_loss": 0.00207964078290388
    },
    {
      "action_query_batch_std": 0.3394696950912476,
      "action_query_norm": 40.690133285522464,
      "action_relation_loss": 0.002182164532132447,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.330215154349858,
      "flow_matching_loss": 0.015867122169584035,
      "grad_norm": 0.0816485658288002,
      "h_stock_valid_norm": 98.14118576049805,
      "invalid_tail_samples": 25.9,
      "learning_rate": 7.581052917039211e-06,
      "loss": 0.018,
      "positive_teacher_pairs": 26368.4,
      "query_h_norm_ratio": 0.4146092802286148,
      "relation_loss": 0.002182164532132447,
      "selected_pairs": 26368.4,
      "step": 24940,
      "student_distance_mean": 22.359049797058105,
      "teacher_distance_mean": 0.8364940047264099,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.002182164532132447,
      "weighted_relation_loss": 0.002182164532132447
    },
    {
      "action_query_batch_std": 0.3485654056072235,
      "action_query_norm": 40.749569702148435,
      "action_relation_loss": 0.0019698293064720927,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.33956969130028,
      "flow_matching_loss": 0.01710126306861639,
      "grad_norm": 0.09375561028718948,
      "h_stock_valid_norm": 97.85088119506835,
      "invalid_tail_samples": 23.5,
      "learning_rate": 7.551901046171645e-06,
      "loss": 0.0191,
      "positive_teacher_pairs": 26924.3,
      "query_h_norm_ratio": 0.4164476454257965,
      "relation_loss": 0.0019698293064720927,
      "selected_pairs": 26924.3,
      "step": 24950,
      "student_distance_mean": 22.88029899597168,
      "teacher_distance_mean": 0.8571976661682129,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0019698293064720927,
      "weighted_relation_loss": 0.0019698293064720927
    },
    {
      "action_query_batch_std": 0.3462609976530075,
      "action_query_norm": 40.72504844665527,
      "action_relation_loss": 0.0020821419777348636,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.348924228250702,
      "flow_matching_loss": 0.016073268093168735,
      "grad_norm": 0.08714274317026138,
      "h_stock_valid_norm": 97.72581481933594,
      "invalid_tail_samples": 25.6,
      "learning_rate": 7.522800753746895e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 26442.2,
      "query_h_norm_ratio": 0.41672970950603483,
      "relation_loss": 0.0020821419777348636,
      "selected_pairs": 26442.2,
      "step": 24960,
      "student_distance_mean": 22.812699127197266,
      "teacher_distance_mean": 0.8561252772808075,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0020821419777348636,
      "weighted_relation_loss": 0.0020821419777348636
    },
    {
      "action_query_batch_std": 0.3486126929521561,
      "action_query_norm": 40.74413223266602,
      "action_relation_loss": 0.002108931285329163,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.358278765201124,
      "flow_matching_loss": 0.01738578137010336,
      "grad_norm": 0.08968445658683777,
      "h_stock_valid_norm": 97.87494049072265,
      "invalid_tail_samples": 24.3,
      "learning_rate": 7.493752075124577e-06,
      "loss": 0.0195,
      "positive_teacher_pairs": 26739.0,
      "query_h_norm_ratio": 0.41628915071487427,
      "relation_loss": 0.002108931285329163,
      "selected_pairs": 26739.0,
      "step": 24970,
      "student_distance_mean": 22.96895236968994,
      "teacher_distance_mean": 0.8601784944534302,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.002108931285329163,
      "weighted_relation_loss": 0.002108931285329163
    },
    {
      "action_query_batch_std": 0.34200324714183805,
      "action_query_norm": 40.68895950317383,
      "action_relation_loss": 0.002003942185547203,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.367633302151543,
      "flow_matching_loss": 0.016449785511940717,
      "grad_norm": 0.09030092507600784,
      "h_stock_valid_norm": 97.9266860961914,
      "invalid_tail_samples": 25.4,
      "learning_rate": 7.464755045601557e-06,
      "loss": 0.0185,
      "positive_teacher_pairs": 26484.2,
      "query_h_norm_ratio": 0.41550560891628263,
      "relation_loss": 0.002003942185547203,
      "selected_pairs": 26484.2,
      "step": 24980,
      "student_distance_mean": 22.528100776672364,
      "teacher_distance_mean": 0.8436359584331512,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.002003942185547203,
      "weighted_relation_loss": 0.002003942185547203
    },
    {
      "action_query_batch_std": 0.3486866295337677,
      "action_query_norm": 40.76504020690918,
      "action_relation_loss": 0.0019853984471410514,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.376987839101965,
      "flow_matching_loss": 0.015619882941246032,
      "grad_norm": 0.07973547279834747,
      "h_stock_valid_norm": 98.14310455322266,
      "invalid_tail_samples": 23.7,
      "learning_rate": 7.435809700411972e-06,
      "loss": 0.0176,
      "positive_teacher_pairs": 26875.9,
      "query_h_norm_ratio": 0.4153659850358963,
      "relation_loss": 0.0019853984471410514,
      "selected_pairs": 26875.9,
      "step": 24990,
      "student_distance_mean": 22.86029167175293,
      "teacher_distance_mean": 0.8558652102947235,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0019853984471410514,
      "weighted_relation_loss": 0.0019853984471410514
    },
    {
      "action_query_batch_std": 0.3515302002429962,
      "action_query_norm": 40.761832046508786,
      "action_relation_loss": 0.0019940051482990385,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.386342376052387,
      "flow_matching_loss": 0.016647026129066943,
      "grad_norm": 0.1071041151881218,
      "h_stock_valid_norm": 98.02650604248046,
      "invalid_tail_samples": 21.8,
      "learning_rate": 7.40691607472715e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 27314.7,
      "query_h_norm_ratio": 0.41583068370819093,
      "relation_loss": 0.0019940051482990385,
      "selected_pairs": 27314.7,
      "step": 25000,
      "student_distance_mean": 23.04846935272217,
      "teacher_distance_mean": 0.8619038224220276,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.0019940051482990385,
      "weighted_relation_loss": 0.0019940051482990385
    },
    {
      "action_query_batch_std": 0.3409247934818268,
      "action_query_norm": 40.6980697631836,
      "action_relation_loss": 0.0021389634581282734,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.395696913002805,
      "flow_matching_loss": 0.01559969699010253,
      "grad_norm": 0.08028081804513931,
      "h_stock_valid_norm": 98.05153274536133,
      "invalid_tail_samples": 23.1,
      "learning_rate": 7.3780742036555606e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 27018.6,
      "query_h_norm_ratio": 0.41506982743740084,
      "relation_loss": 0.0021389634581282734,
      "selected_pairs": 27018.6,
      "step": 25010,
      "student_distance_mean": 22.39442958831787,
      "teacher_distance_mean": 0.837670648097992,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0021389634581282734,
      "weighted_relation_loss": 0.0021389634581282734
    },
    {
      "action_query_batch_std": 0.34763861298561094,
      "action_query_norm": 40.75246620178223,
      "action_relation_loss": 0.002064408641308546,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.405051449953227,
      "flow_matching_loss": 0.017193533945828676,
      "grad_norm": 0.14122284948825836,
      "h_stock_valid_norm": 98.04494171142578,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.349284122242783e-06,
      "loss": 0.0193,
      "positive_teacher_pairs": 27034.2,
      "query_h_norm_ratio": 0.4156548023223877,
      "relation_loss": 0.002064408641308546,
      "selected_pairs": 27034.2,
      "step": 25020,
      "student_distance_mean": 22.73447380065918,
      "teacher_distance_mean": 0.8518600404262543,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.002064408641308546,
      "weighted_relation_loss": 0.002064408641308546
    },
    {
      "action_query_batch_std": 0.3445045590400696,
      "action_query_norm": 40.72468719482422,
      "action_relation_loss": 0.00213906456483528,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.41440598690365,
      "flow_matching_loss": 0.01626672763377428,
      "grad_norm": 0.20869964361190796,
      "h_stock_valid_norm": 97.74055404663086,
      "invalid_tail_samples": 26.2,
      "learning_rate": 7.320545865471512e-06,
      "loss": 0.0184,
      "positive_teacher_pairs": 26298.4,
      "query_h_norm_ratio": 0.41666334569454194,
      "relation_loss": 0.00213906456483528,
      "selected_pairs": 26298.4,
      "step": 25030,
      "student_distance_mean": 22.654017066955568,
      "teacher_distance_mean": 0.8490740776062011,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.00213906456483528,
      "weighted_relation_loss": 0.00213906456483528
    },
    {
      "action_query_batch_std": 0.3449371725320816,
      "action_query_norm": 40.747208786010745,
      "action_relation_loss": 0.0019530956516973675,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.42376052385407,
      "flow_matching_loss": 0.016327657271176577,
      "grad_norm": 0.08609282970428467,
      "h_stock_valid_norm": 98.07883224487304,
      "invalid_tail_samples": 22.6,
      "learning_rate": 7.291859468261425e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 27138.2,
      "query_h_norm_ratio": 0.41545735895633695,
      "relation_loss": 0.0019530956516973675,
      "selected_pairs": 27138.2,
      "step": 25040,
      "student_distance_mean": 22.657612800598145,
      "teacher_distance_mean": 0.8482806086540222,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.0019530956516973675,
      "weighted_relation_loss": 0.0019530956516973675
    },
    {
      "action_query_batch_std": 0.3475876361131668,
      "action_query_norm": 40.746258926391604,
      "action_relation_loss": 0.00208037244156003,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.43311506080449,
      "flow_matching_loss": 0.015662139095366002,
      "grad_norm": 0.1364711970090866,
      "h_stock_valid_norm": 97.8647834777832,
      "invalid_tail_samples": 24.4,
      "learning_rate": 7.263224965469195e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 26711.3,
      "query_h_norm_ratio": 0.4163556009531021,
      "relation_loss": 0.00208037244156003,
      "selected_pairs": 26711.3,
      "step": 25050,
      "student_distance_mean": 22.76005268096924,
      "teacher_distance_mean": 0.8531992137432098,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.00208037244156003,
      "weighted_relation_loss": 0.00208037244156003
    },
    {
      "action_query_batch_std": 0.3476477324962616,
      "action_query_norm": 40.738485717773436,
      "action_relation_loss": 0.0020452418248169124,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.44246959775491,
      "flow_matching_loss": 0.017520982772111893,
      "grad_norm": 0.2387775480747223,
      "h_stock_valid_norm": 97.76121139526367,
      "invalid_tail_samples": 24.8,
      "learning_rate": 7.234642391888446e-06,
      "loss": 0.0196,
      "positive_teacher_pairs": 26623.0,
      "query_h_norm_ratio": 0.41671771109104155,
      "relation_loss": 0.0020452418248169124,
      "selected_pairs": 26623.0,
      "step": 25060,
      "student_distance_mean": 22.948941040039063,
      "teacher_distance_mean": 0.8610044360160828,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0020452418248169124,
      "weighted_relation_loss": 0.0020452418248169124
    },
    {
      "action_query_batch_std": 0.34648035168647767,
      "action_query_norm": 40.72793807983398,
      "action_relation_loss": 0.0020869842963293197,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.451824134705333,
      "flow_matching_loss": 0.017317048367112876,
      "grad_norm": 0.09936577081680298,
      "h_stock_valid_norm": 97.5571533203125,
      "invalid_tail_samples": 24.0,
      "learning_rate": 7.206111782249697e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 26804.4,
      "query_h_norm_ratio": 0.41748283207416537,
      "relation_loss": 0.0020869842963293197,
      "selected_pairs": 26804.4,
      "step": 25070,
      "student_distance_mean": 22.67321891784668,
      "teacher_distance_mean": 0.8511164486408234,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0020869842963293197,
      "weighted_relation_loss": 0.0020869842963293197
    },
    {
      "action_query_batch_std": 0.34974526762962344,
      "action_query_norm": 40.72610015869141,
      "action_relation_loss": 0.0020697711152024567,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.46117867165575,
      "flow_matching_loss": 0.017178981471806763,
      "grad_norm": 0.09482438862323761,
      "h_stock_valid_norm": 97.84447326660157,
      "invalid_tail_samples": 23.2,
      "learning_rate": 7.1776331712203385e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 26987.8,
      "query_h_norm_ratio": 0.41623567044734955,
      "relation_loss": 0.0020697711152024567,
      "selected_pairs": 26987.8,
      "step": 25080,
      "student_distance_mean": 22.946511459350585,
      "teacher_distance_mean": 0.8602640450000762,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0020697711152024567,
      "weighted_relation_loss": 0.0020697711152024567
    },
    {
      "action_query_batch_std": 0.34950862526893617,
      "action_query_norm": 40.7448429107666,
      "action_relation_loss": 0.0019383959239348768,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.470533208606174,
      "flow_matching_loss": 0.01662822309881449,
      "grad_norm": 0.08026940375566483,
      "h_stock_valid_norm": 97.77490768432617,
      "invalid_tail_samples": 23.7,
      "learning_rate": 7.149206593404561e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 26871.2,
      "query_h_norm_ratio": 0.41672379076480864,
      "relation_loss": 0.0019383959239348768,
      "selected_pairs": 26871.2,
      "step": 25090,
      "student_distance_mean": 22.97083396911621,
      "teacher_distance_mean": 0.862037068605423,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0019383959239348768,
      "weighted_relation_loss": 0.0019383959239348768
    },
    {
      "action_query_batch_std": 0.34771862626075745,
      "action_query_norm": 40.714700317382814,
      "action_relation_loss": 0.0019876194768585266,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.479887745556596,
      "flow_matching_loss": 0.016526438202708958,
      "grad_norm": 0.2076907604932785,
      "h_stock_valid_norm": 97.93125915527344,
      "invalid_tail_samples": 24.7,
      "learning_rate": 7.120832083343337e-06,
      "loss": 0.0185,
      "positive_teacher_pairs": 26639.2,
      "query_h_norm_ratio": 0.41575060188770296,
      "relation_loss": 0.0019876194768585266,
      "selected_pairs": 26639.2,
      "step": 25100,
      "student_distance_mean": 22.76803207397461,
      "teacher_distance_mean": 0.8525038421154022,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0019876194768585266,
      "weighted_relation_loss": 0.0019876194768585266
    },
    {
      "action_query_batch_std": 0.3468040645122528,
      "action_query_norm": 40.75459938049316,
      "action_relation_loss": 0.001974848494865,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.489242282507014,
      "flow_matching_loss": 0.014823612384498119,
      "grad_norm": 0.08157746493816376,
      "h_stock_valid_norm": 97.90364532470703,
      "invalid_tail_samples": 22.9,
      "learning_rate": 7.092509675514369e-06,
      "loss": 0.0168,
      "positive_teacher_pairs": 27055.6,
      "query_h_norm_ratio": 0.4162737250328064,
      "relation_loss": 0.001974848494865,
      "selected_pairs": 27055.6,
      "step": 25110,
      "student_distance_mean": 22.903787040710448,
      "teacher_distance_mean": 0.8569399237632751,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.001974848494865,
      "weighted_relation_loss": 0.001974848494865
    },
    {
      "action_query_batch_std": 0.3469160169363022,
      "action_query_norm": 40.74464111328125,
      "action_relation_loss": 0.002042776218149811,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.498596819457436,
      "flow_matching_loss": 0.016220353543758392,
      "grad_norm": 0.09940143674612045,
      "h_stock_valid_norm": 98.02597122192383,
      "invalid_tail_samples": 23.1,
      "learning_rate": 7.064239404332063e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 27012.7,
      "query_h_norm_ratio": 0.4156536072492599,
      "relation_loss": 0.002042776218149811,
      "selected_pairs": 27012.7,
      "step": 25120,
      "student_distance_mean": 22.770227432250977,
      "teacher_distance_mean": 0.8545400619506835,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002042776218149811,
      "weighted_relation_loss": 0.002042776218149811
    },
    {
      "action_query_batch_std": 0.34739882349967954,
      "action_query_norm": 40.75100135803223,
      "action_relation_loss": 0.0020835028379224243,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.507951356407858,
      "flow_matching_loss": 0.017431695014238358,
      "grad_norm": 0.11416081339120865,
      "h_stock_valid_norm": 98.02338409423828,
      "invalid_tail_samples": 22.2,
      "learning_rate": 7.03602130414745e-06,
      "loss": 0.0195,
      "positive_teacher_pairs": 27225.8,
      "query_h_norm_ratio": 0.4157279014587402,
      "relation_loss": 0.0020835028379224243,
      "selected_pairs": 27225.8,
      "step": 25130,
      "student_distance_mean": 22.857093620300294,
      "teacher_distance_mean": 0.8570019364356994,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0020835028379224243,
      "weighted_relation_loss": 0.0020835028379224243
    },
    {
      "action_query_batch_std": 0.3484572798013687,
      "action_query_norm": 40.73262062072754,
      "action_relation_loss": 0.002052266034297645,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.51730589335828,
      "flow_matching_loss": 0.0176374688744545,
      "grad_norm": 0.11554910242557526,
      "h_stock_valid_norm": 98.04007949829102,
      "invalid_tail_samples": 23.1,
      "learning_rate": 7.007855409248198e-06,
      "loss": 0.0197,
      "positive_teacher_pairs": 27010.0,
      "query_h_norm_ratio": 0.41547088623046874,
      "relation_loss": 0.002052266034297645,
      "selected_pairs": 27010.0,
      "step": 25140,
      "student_distance_mean": 22.771446418762206,
      "teacher_distance_mean": 0.8534028589725494,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002052266034297645,
      "weighted_relation_loss": 0.002052266034297645
    },
    {
      "action_query_batch_std": 0.3487879395484924,
      "action_query_norm": 40.7857666015625,
      "action_relation_loss": 0.001983218884561211,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.5266604303087,
      "flow_matching_loss": 0.015698001254349947,
      "grad_norm": 0.10259365290403366,
      "h_stock_valid_norm": 98.13163681030274,
      "invalid_tail_samples": 22.7,
      "learning_rate": 6.979741753858521e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 27117.5,
      "query_h_norm_ratio": 0.41562536358833313,
      "relation_loss": 0.001983218884561211,
      "selected_pairs": 27117.5,
      "step": 25150,
      "student_distance_mean": 22.96022758483887,
      "teacher_distance_mean": 0.8620621621608734,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.001983218884561211,
      "weighted_relation_loss": 0.001983218884561211
    },
    {
      "action_query_batch_std": 0.34654973447322845,
      "action_query_norm": 40.74836196899414,
      "action_relation_loss": 0.0019794676103629173,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.53601496725912,
      "flow_matching_loss": 0.01743858549743891,
      "grad_norm": 0.09213615208864212,
      "h_stock_valid_norm": 98.08988571166992,
      "invalid_tail_samples": 24.4,
      "learning_rate": 6.951680372139158e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 26711.1,
      "query_h_norm_ratio": 0.41542240381240847,
      "relation_loss": 0.0019794676103629173,
      "selected_pairs": 26711.1,
      "step": 25160,
      "student_distance_mean": 22.752892875671385,
      "teacher_distance_mean": 0.852127081155777,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0019794676103629173,
      "weighted_relation_loss": 0.0019794676103629173
    },
    {
      "action_query_batch_std": 0.34347043633461,
      "action_query_norm": 40.716762924194335,
      "action_relation_loss": 0.0020291159395128488,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.545369504209543,
      "flow_matching_loss": 0.017901107668876648,
      "grad_norm": 0.10247954726219177,
      "h_stock_valid_norm": 97.94337615966796,
      "invalid_tail_samples": 23.2,
      "learning_rate": 6.923671298187334e-06,
      "loss": 0.0199,
      "positive_teacher_pairs": 26988.1,
      "query_h_norm_ratio": 0.41572031676769255,
      "relation_loss": 0.0020291159395128488,
      "selected_pairs": 26988.1,
      "step": 25170,
      "student_distance_mean": 22.505489921569826,
      "teacher_distance_mean": 0.8429591476917266,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0020291159395128488,
      "weighted_relation_loss": 0.0020291159395128488
    },
    {
      "action_query_batch_std": 0.3476086616516113,
      "action_query_norm": 40.76179084777832,
      "action_relation_loss": 0.002032387291546911,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.55472404115996,
      "flow_matching_loss": 0.01588670127093792,
      "grad_norm": 0.08829838782548904,
      "h_stock_valid_norm": 98.07935409545898,
      "invalid_tail_samples": 23.5,
      "learning_rate": 6.895714566036704e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 26915.5,
      "query_h_norm_ratio": 0.41560650169849395,
      "relation_loss": 0.002032387291546911,
      "selected_pairs": 26915.5,
      "step": 25180,
      "student_distance_mean": 22.72064838409424,
      "teacher_distance_mean": 0.8523014485836029,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.002032387291546911,
      "weighted_relation_loss": 0.002032387291546911
    },
    {
      "action_query_batch_std": 0.3502443045377731,
      "action_query_norm": 40.756425857543945,
      "action_relation_loss": 0.002029734989628196,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.564078578110383,
      "flow_matching_loss": 0.01609508404508233,
      "grad_norm": 0.15413956344127655,
      "h_stock_valid_norm": 97.87354583740235,
      "invalid_tail_samples": 22.3,
      "learning_rate": 6.86781020965736e-06,
      "loss": 0.0181,
      "positive_teacher_pairs": 27198.2,
      "query_h_norm_ratio": 0.4164247751235962,
      "relation_loss": 0.002029734989628196,
      "selected_pairs": 27198.2,
      "step": 25190,
      "student_distance_mean": 22.925461769104004,
      "teacher_distance_mean": 0.8594076454639434,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.002029734989628196,
      "weighted_relation_loss": 0.002029734989628196
    },
    {
      "action_query_batch_std": 0.3476431846618652,
      "action_query_norm": 40.7352596282959,
      "action_relation_loss": 0.0020174438948743047,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.573433115060805,
      "flow_matching_loss": 0.01611078353598714,
      "grad_norm": 0.09838441014289856,
      "h_stock_valid_norm": 98.10590438842773,
      "invalid_tail_samples": 24.2,
      "learning_rate": 6.839958262955709e-06,
      "loss": 0.0181,
      "positive_teacher_pairs": 26762.5,
      "query_h_norm_ratio": 0.41521849036216735,
      "relation_loss": 0.0020174438948743047,
      "selected_pairs": 26762.5,
      "step": 25200,
      "student_distance_mean": 22.828954696655273,
      "teacher_distance_mean": 0.8574173092842102,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0020174438948743047,
      "weighted_relation_loss": 0.0020174438948743047
    },
    {
      "action_query_batch_std": 0.349499836564064,
      "action_query_norm": 40.76260604858398,
      "action_relation_loss": 0.0019456233480013908,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.582787652011227,
      "flow_matching_loss": 0.016313616745173933,
      "grad_norm": 0.13494010269641876,
      "h_stock_valid_norm": 98.03893356323242,
      "invalid_tail_samples": 21.6,
      "learning_rate": 6.8121587597744894e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 27359.0,
      "query_h_norm_ratio": 0.41578373610973357,
      "relation_loss": 0.0019456233480013908,
      "selected_pairs": 27359.0,
      "step": 25210,
      "student_distance_mean": 22.96843147277832,
      "teacher_distance_mean": 0.8614673972129822,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.0019456233480013908,
      "weighted_relation_loss": 0.0019456233480013908
    },
    {
      "action_query_batch_std": 0.3484904170036316,
      "action_query_norm": 40.73589897155762,
      "action_relation_loss": 0.002016292931511998,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.592142188961645,
      "flow_matching_loss": 0.016638618893921375,
      "grad_norm": 0.08781719952821732,
      "h_stock_valid_norm": 97.89434204101562,
      "invalid_tail_samples": 26.0,
      "learning_rate": 6.784411733892732e-06,
      "loss": 0.0187,
      "positive_teacher_pairs": 26344.1,
      "query_h_norm_ratio": 0.41612487733364106,
      "relation_loss": 0.002016292931511998,
      "selected_pairs": 26344.1,
      "step": 25220,
      "student_distance_mean": 22.84037971496582,
      "teacher_distance_mean": 0.8575632810592652,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.002016292931511998,
      "weighted_relation_loss": 0.002016292931511998
    },
    {
      "action_query_batch_std": 0.3481846243143082,
      "action_query_norm": 40.73599510192871,
      "action_relation_loss": 0.002052739984355867,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.601496725912067,
      "flow_matching_loss": 0.015518965292721986,
      "grad_norm": 0.11239901930093765,
      "h_stock_valid_norm": 97.79420700073243,
      "invalid_tail_samples": 22.0,
      "learning_rate": 6.756717219025666e-06,
      "loss": 0.0176,
      "positive_teacher_pairs": 27274.5,
      "query_h_norm_ratio": 0.4165508896112442,
      "relation_loss": 0.002052739984355867,
      "selected_pairs": 27274.5,
      "step": 25230,
      "student_distance_mean": 22.825635528564455,
      "teacher_distance_mean": 0.8550836324691773,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.002052739984355867,
      "weighted_relation_loss": 0.002052739984355867
    },
    {
      "action_query_batch_std": 0.34444091022014617,
      "action_query_norm": 40.705312728881836,
      "action_relation_loss": 0.0020393623970448973,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.61085126286249,
      "flow_matching_loss": 0.016620440408587456,
      "grad_norm": 0.12446478009223938,
      "h_stock_valid_norm": 97.8706672668457,
      "invalid_tail_samples": 21.6,
      "learning_rate": 6.729075248824762e-06,
      "loss": 0.0187,
      "positive_teacher_pairs": 27363.7,
      "query_h_norm_ratio": 0.4159132570028305,
      "relation_loss": 0.0020393623970448973,
      "selected_pairs": 27363.7,
      "step": 25240,
      "student_distance_mean": 22.64878101348877,
      "teacher_distance_mean": 0.8501187860965729,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.0020393623970448973,
      "weighted_relation_loss": 0.0020393623970448973
    },
    {
      "action_query_batch_std": 0.345394229888916,
      "action_query_norm": 40.71989822387695,
      "action_relation_loss": 0.00199774953071028,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.620205799812908,
      "flow_matching_loss": 0.016794197168201207,
      "grad_norm": 0.1169453114271164,
      "h_stock_valid_norm": 97.82650985717774,
      "invalid_tail_samples": 22.9,
      "learning_rate": 6.701485856877615e-06,
      "loss": 0.0188,
      "positive_teacher_pairs": 27063.2,
      "query_h_norm_ratio": 0.41624786555767057,
      "relation_loss": 0.00199774953071028,
      "selected_pairs": 27063.2,
      "step": 25250,
      "student_distance_mean": 22.76842098236084,
      "teacher_distance_mean": 0.8541735887527466,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.00199774953071028,
      "weighted_relation_loss": 0.00199774953071028
    },
    {
      "action_query_batch_std": 0.3477183490991592,
      "action_query_norm": 40.73935546875,
      "action_relation_loss": 0.0019427810912020504,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.62956033676333,
      "flow_matching_loss": 0.017032453510910272,
      "grad_norm": 0.08551908284425735,
      "h_stock_valid_norm": 97.99978179931641,
      "invalid_tail_samples": 26.2,
      "learning_rate": 6.673949076707925e-06,
      "loss": 0.019,
      "positive_teacher_pairs": 26298.0,
      "query_h_norm_ratio": 0.4157110840082169,
      "relation_loss": 0.0019427810912020504,
      "selected_pairs": 26298.0,
      "step": 25260,
      "student_distance_mean": 22.778371810913086,
      "teacher_distance_mean": 0.8533273279666901,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.0019427810912020504,
      "weighted_relation_loss": 0.0019427810912020504
    },
    {
      "action_query_batch_std": 0.35004229545593263,
      "action_query_norm": 40.745222091674805,
      "action_relation_loss": 0.001950160099659115,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.63891487371375,
      "flow_matching_loss": 0.01680471245199442,
      "grad_norm": 0.1357143521308899,
      "h_stock_valid_norm": 97.9077049255371,
      "invalid_tail_samples": 25.2,
      "learning_rate": 6.646464941775499e-06,
      "loss": 0.0188,
      "positive_teacher_pairs": 26523.9,
      "query_h_norm_ratio": 0.4161617785692215,
      "relation_loss": 0.001950160099659115,
      "selected_pairs": 26523.9,
      "step": 25270,
      "student_distance_mean": 23.04728603363037,
      "teacher_distance_mean": 0.8642933785915374,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.001950160099659115,
      "weighted_relation_loss": 0.001950160099659115
    },
    {
      "action_query_batch_std": 0.3454556792974472,
      "action_query_norm": 40.70995025634765,
      "action_relation_loss": 0.0019971991190686823,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.648269410664174,
      "flow_matching_loss": 0.015059915184974671,
      "grad_norm": 0.08569449931383133,
      "h_stock_valid_norm": 97.82760772705078,
      "invalid_tail_samples": 23.9,
      "learning_rate": 6.619033485476128e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 26828.6,
      "query_h_norm_ratio": 0.4161423474550247,
      "relation_loss": 0.0019971991190686823,
      "selected_pairs": 26828.6,
      "step": 25280,
      "student_distance_mean": 22.68051414489746,
      "teacher_distance_mean": 0.8521145522594452,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0019971991190686823,
      "weighted_relation_loss": 0.0019971991190686823
    },
    {
      "action_query_batch_std": 0.3459338039159775,
      "action_query_norm": 40.714324188232425,
      "action_relation_loss": 0.002116319735068828,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.657623947614592,
      "flow_matching_loss": 0.01709431866183877,
      "grad_norm": 0.08505735546350479,
      "h_stock_valid_norm": 97.71660690307617,
      "invalid_tail_samples": 23.1,
      "learning_rate": 6.59165474114164e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 27009.3,
      "query_h_norm_ratio": 0.4166595250368118,
      "relation_loss": 0.002116319735068828,
      "selected_pairs": 27009.3,
      "step": 25290,
      "student_distance_mean": 22.81599826812744,
      "teacher_distance_mean": 0.8544628262519837,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002116319735068828,
      "weighted_relation_loss": 0.002116319735068828
    },
    {
      "action_query_batch_std": 0.3487987250089645,
      "action_query_norm": 40.7347297668457,
      "action_relation_loss": 0.00196103883208707,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.666978484565014,
      "flow_matching_loss": 0.01602438846603036,
      "grad_norm": 0.09324026852846146,
      "h_stock_valid_norm": 97.88962326049804,
      "invalid_tail_samples": 25.3,
      "learning_rate": 6.564328742039783e-06,
      "loss": 0.018,
      "positive_teacher_pairs": 26507.6,
      "query_h_norm_ratio": 0.4161305069923401,
      "relation_loss": 0.00196103883208707,
      "selected_pairs": 26507.6,
      "step": 25300,
      "student_distance_mean": 22.91395511627197,
      "teacher_distance_mean": 0.8608613610267639,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.00196103883208707,
      "weighted_relation_loss": 0.00196103883208707
    },
    {
      "action_query_batch_std": 0.34637720286846163,
      "action_query_norm": 40.718849182128906,
      "action_relation_loss": 0.0020155386184342204,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.676333021515436,
      "flow_matching_loss": 0.016540302243083716,
      "grad_norm": 0.10933537781238556,
      "h_stock_valid_norm": 97.73377914428711,
      "invalid_tail_samples": 24.6,
      "learning_rate": 6.537055521374219e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 26662.2,
      "query_h_norm_ratio": 0.41663173139095305,
      "relation_loss": 0.0020155386184342204,
      "selected_pairs": 26662.2,
      "step": 25310,
      "student_distance_mean": 22.780628776550294,
      "teacher_distance_mean": 0.8551479160785675,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0020155386184342204,
      "weighted_relation_loss": 0.0020155386184342204
    },
    {
      "action_query_batch_std": 0.34766672253608705,
      "action_query_norm": 40.70105323791504,
      "action_relation_loss": 0.0020346434554085135,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.685687558465855,
      "flow_matching_loss": 0.017848546989262103,
      "grad_norm": 0.10494983196258545,
      "h_stock_valid_norm": 97.85922012329101,
      "invalid_tail_samples": 25.7,
      "learning_rate": 6.509835112284485e-06,
      "loss": 0.0199,
      "positive_teacher_pairs": 26413.3,
      "query_h_norm_ratio": 0.41591706573963166,
      "relation_loss": 0.0020346434554085135,
      "selected_pairs": 26413.3,
      "step": 25320,
      "student_distance_mean": 22.819813346862794,
      "teacher_distance_mean": 0.8562312662601471,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0020346434554085135,
      "weighted_relation_loss": 0.0020346434554085135
    },
    {
      "action_query_batch_std": 0.34235765039920807,
      "action_query_norm": 40.6927848815918,
      "action_relation_loss": 0.00205215752357617,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.695042095416277,
      "flow_matching_loss": 0.016176778823137283,
      "grad_norm": 0.08763092011213303,
      "h_stock_valid_norm": 97.64910278320312,
      "invalid_tail_samples": 23.5,
      "learning_rate": 6.482667547845944e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 26928.2,
      "query_h_norm_ratio": 0.41672863364219664,
      "relation_loss": 0.00205215752357617,
      "selected_pairs": 26928.2,
      "step": 25330,
      "student_distance_mean": 22.451692008972167,
      "teacher_distance_mean": 0.8429404199123383,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.00205215752357617,
      "weighted_relation_loss": 0.00205215752357617
    },
    {
      "action_query_batch_std": 0.342371928691864,
      "action_query_norm": 40.682373428344725,
      "action_relation_loss": 0.0019787222845479846,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.7043966323667,
      "flow_matching_loss": 0.01619078582152724,
      "grad_norm": 0.12628796696662903,
      "h_stock_valid_norm": 97.78047332763671,
      "invalid_tail_samples": 24.9,
      "learning_rate": 6.455552861069736e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 26608.2,
      "query_h_norm_ratio": 0.41606042981147767,
      "relation_loss": 0.0019787222845479846,
      "selected_pairs": 26608.2,
      "step": 25340,
      "student_distance_mean": 22.626897239685057,
      "teacher_distance_mean": 0.8509035408496857,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0019787222845479846,
      "weighted_relation_loss": 0.0019787222845479846
    },
    {
      "action_query_batch_std": 0.34805082678794863,
      "action_query_norm": 40.730097961425784,
      "action_relation_loss": 0.0019592809956520798,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.71375116931712,
      "flow_matching_loss": 0.01746669951826334,
      "grad_norm": 0.11724373698234558,
      "h_stock_valid_norm": 97.91788864135742,
      "invalid_tail_samples": 24.2,
      "learning_rate": 6.428491084902788e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 26758.3,
      "query_h_norm_ratio": 0.41596410870552064,
      "relation_loss": 0.0019592809956520798,
      "selected_pairs": 26758.3,
      "step": 25350,
      "student_distance_mean": 22.910112190246583,
      "teacher_distance_mean": 0.8587798416614533,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0019592809956520798,
      "weighted_relation_loss": 0.0019592809956520798
    },
    {
      "action_query_batch_std": 0.3450307101011276,
      "action_query_norm": 40.731056213378906,
      "action_relation_loss": 0.00205424897139892,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.72310570626754,
      "flow_matching_loss": 0.015671515744179486,
      "grad_norm": 0.09408263117074966,
      "h_stock_valid_norm": 97.73210296630859,
      "invalid_tail_samples": 22.9,
      "learning_rate": 6.401482252227698e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 27056.5,
      "query_h_norm_ratio": 0.41676445603370665,
      "relation_loss": 0.00205424897139892,
      "selected_pairs": 27056.5,
      "step": 25360,
      "student_distance_mean": 22.675310134887695,
      "teacher_distance_mean": 0.8500539839267731,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.00205424897139892,
      "weighted_relation_loss": 0.00205424897139892
    },
    {
      "action_query_batch_std": 0.3482148140668869,
      "action_query_norm": 40.73749961853027,
      "action_relation_loss": 0.0020030262414366005,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.73246024321796,
      "flow_matching_loss": 0.014839706756174565,
      "grad_norm": 0.07245764136314392,
      "h_stock_valid_norm": 97.89058380126953,
      "invalid_tail_samples": 23.5,
      "learning_rate": 6.37452639586274e-06,
      "loss": 0.0168,
      "positive_teacher_pairs": 26917.3,
      "query_h_norm_ratio": 0.41615564525127413,
      "relation_loss": 0.0020030262414366005,
      "selected_pairs": 26917.3,
      "step": 25370,
      "student_distance_mean": 22.801546478271483,
      "teacher_distance_mean": 0.8543863475322724,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0020030262414366005,
      "weighted_relation_loss": 0.0020030262414366005
    },
    {
      "action_query_batch_std": 0.34853472411632536,
      "action_query_norm": 40.75855560302735,
      "action_relation_loss": 0.0020222926279529927,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.741814780168383,
      "flow_matching_loss": 0.016028139833360912,
      "grad_norm": 0.13099409639835358,
      "h_stock_valid_norm": 97.95098495483398,
      "invalid_tail_samples": 23.5,
      "learning_rate": 6.347623548561826e-06,
      "loss": 0.0181,
      "positive_teacher_pairs": 26919.8,
      "query_h_norm_ratio": 0.41611321568489074,
      "relation_loss": 0.0020222926279529927,
      "selected_pairs": 26919.8,
      "step": 25380,
      "student_distance_mean": 22.852530670166015,
      "teacher_distance_mean": 0.8581567645072937,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0020222926279529927,
      "weighted_relation_loss": 0.0020222926279529927
    },
    {
      "action_query_batch_std": 0.3489438593387604,
      "action_query_norm": 40.74990348815918,
      "action_relation_loss": 0.0019666382460854946,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.7511693171188,
      "flow_matching_loss": 0.01701699560508132,
      "grad_norm": 0.07207988202571869,
      "h_stock_valid_norm": 98.09116821289062,
      "invalid_tail_samples": 23.4,
      "learning_rate": 6.320773743014441e-06,
      "loss": 0.019,
      "positive_teacher_pairs": 26946.2,
      "query_h_norm_ratio": 0.4154298037290573,
      "relation_loss": 0.0019666382460854946,
      "selected_pairs": 26946.2,
      "step": 25390,
      "student_distance_mean": 22.87278232574463,
      "teacher_distance_mean": 0.8573069274425507,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0019666382460854946,
      "weighted_relation_loss": 0.0019666382460854946
    },
    {
      "action_query_batch_std": 0.3494224339723587,
      "action_query_norm": 40.74763984680176,
      "action_relation_loss": 0.0019280819571577013,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.760523854069223,
      "flow_matching_loss": 0.01627680119127035,
      "grad_norm": 0.08350599557161331,
      "h_stock_valid_norm": 97.99827117919922,
      "invalid_tail_samples": 23.5,
      "learning_rate": 6.293977011845648e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 26923.0,
      "query_h_norm_ratio": 0.41580261290073395,
      "relation_loss": 0.0019280819571577013,
      "selected_pairs": 26923.0,
      "step": 25400,
      "student_distance_mean": 23.03394603729248,
      "teacher_distance_mean": 0.8643244743347168,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0019280819571577013,
      "weighted_relation_loss": 0.0019280819571577013
    },
    {
      "action_query_batch_std": 0.34528208076953887,
      "action_query_norm": 40.712771606445315,
      "action_relation_loss": 0.002005394222214818,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.769878391019645,
      "flow_matching_loss": 0.01569254705682397,
      "grad_norm": 0.12096519023180008,
      "h_stock_valid_norm": 98.00024871826172,
      "invalid_tail_samples": 25.2,
      "learning_rate": 6.2672333876159835e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 26533.1,
      "query_h_norm_ratio": 0.41543759107589723,
      "relation_loss": 0.002005394222214818,
      "selected_pairs": 26533.1,
      "step": 25410,
      "student_distance_mean": 22.62521457672119,
      "teacher_distance_mean": 0.8491647601127624,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.002005394222214818,
      "weighted_relation_loss": 0.002005394222214818
    },
    {
      "action_query_batch_std": 0.3471266657114029,
      "action_query_norm": 40.720573806762694,
      "action_relation_loss": 0.002055643079802394,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.779232927970064,
      "flow_matching_loss": 0.017181696835905313,
      "grad_norm": 0.09436918050050735,
      "h_stock_valid_norm": 97.97545852661133,
      "invalid_tail_samples": 22.5,
      "learning_rate": 6.2405429028215e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 27150.1,
      "query_h_norm_ratio": 0.41562274694442747,
      "relation_loss": 0.002055643079802394,
      "selected_pairs": 27150.1,
      "step": 25420,
      "student_distance_mean": 22.854799461364745,
      "teacher_distance_mean": 0.8573266923427582,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.002055643079802394,
      "weighted_relation_loss": 0.002055643079802394
    },
    {
      "action_query_batch_std": 0.34472190141677855,
      "action_query_norm": 40.69859580993652,
      "action_relation_loss": 0.00192225300706923,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.788587464920486,
      "flow_matching_loss": 0.016320824064314364,
      "grad_norm": 0.09569409489631653,
      "h_stock_valid_norm": 97.96891937255859,
      "invalid_tail_samples": 25.5,
      "learning_rate": 6.213905589893631e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 26458.2,
      "query_h_norm_ratio": 0.4154249161481857,
      "relation_loss": 0.00192225300706923,
      "selected_pairs": 26458.2,
      "step": 25430,
      "student_distance_mean": 22.715534591674803,
      "teacher_distance_mean": 0.8517082035541534,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.00192225300706923,
      "weighted_relation_loss": 0.00192225300706923
    },
    {
      "action_query_batch_std": 0.35018416345119474,
      "action_query_norm": 40.75208778381348,
      "action_relation_loss": 0.002005868311971426,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.797942001870908,
      "flow_matching_loss": 0.015808261279016734,
      "grad_norm": 0.14964516460895538,
      "h_stock_valid_norm": 97.8653694152832,
      "invalid_tail_samples": 23.3,
      "learning_rate": 6.187321481199221e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 26967.1,
      "query_h_norm_ratio": 0.41641426980495455,
      "relation_loss": 0.002005868311971426,
      "selected_pairs": 26967.1,
      "step": 25440,
      "student_distance_mean": 23.04481906890869,
      "teacher_distance_mean": 0.8642821550369263,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.002005868311971426,
      "weighted_relation_loss": 0.002005868311971426
    },
    {
      "action_query_batch_std": 0.3484767436981201,
      "action_query_norm": 40.74485549926758,
      "action_relation_loss": 0.0020017250208184123,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.80729653882133,
      "flow_matching_loss": 0.018921134062111378,
      "grad_norm": 0.10588591545820236,
      "h_stock_valid_norm": 97.87754440307617,
      "invalid_tail_samples": 23.9,
      "learning_rate": 6.1607906090404895e-06,
      "loss": 0.0209,
      "positive_teacher_pairs": 26824.3,
      "query_h_norm_ratio": 0.41628819704055786,
      "relation_loss": 0.0020017250208184123,
      "selected_pairs": 26824.3,
      "step": 25450,
      "student_distance_mean": 22.892047500610353,
      "teacher_distance_mean": 0.8573077023029327,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0020017250208184123,
      "weighted_relation_loss": 0.0020017250208184123
    },
    {
      "action_query_batch_std": 0.3457764536142349,
      "action_query_norm": 40.70555458068848,
      "action_relation_loss": 0.0019114892813377083,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.81665107577175,
      "flow_matching_loss": 0.01749465465545654,
      "grad_norm": 0.09878968447446823,
      "h_stock_valid_norm": 97.70949020385743,
      "invalid_tail_samples": 23.2,
      "learning_rate": 6.134313005654929e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 27002.4,
      "query_h_norm_ratio": 0.41660297811031344,
      "relation_loss": 0.0019114892813377083,
      "selected_pairs": 27002.4,
      "step": 25460,
      "student_distance_mean": 22.787814140319824,
      "teacher_distance_mean": 0.8550746977329254,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0019114892813377083,
      "weighted_relation_loss": 0.0019114892813377083
    },
    {
      "action_query_batch_std": 0.34536112248897555,
      "action_query_norm": 40.704508209228514,
      "action_relation_loss": 0.002035416790749878,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.82600561272217,
      "flow_matching_loss": 0.016706031281501055,
      "grad_norm": 0.09791398048400879,
      "h_stock_valid_norm": 97.75278778076172,
      "invalid_tail_samples": 24.9,
      "learning_rate": 6.107888703215337e-06,
      "loss": 0.0187,
      "positive_teacher_pairs": 26602.2,
      "query_h_norm_ratio": 0.4164045363664627,
      "relation_loss": 0.002035416790749878,
      "selected_pairs": 26602.2,
      "step": 25470,
      "student_distance_mean": 22.568789100646974,
      "teacher_distance_mean": 0.8469401121139526,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.002035416790749878,
      "weighted_relation_loss": 0.002035416790749878
    },
    {
      "action_query_batch_std": 0.3444299757480621,
      "action_query_norm": 40.69476470947266,
      "action_relation_loss": 0.0021263332455419004,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.835360149672592,
      "flow_matching_loss": 0.01659996621310711,
      "grad_norm": 0.10334820300340652,
      "h_stock_valid_norm": 97.78198852539063,
      "invalid_tail_samples": 25.1,
      "learning_rate": 6.081517733829722e-06,
      "loss": 0.0187,
      "positive_teacher_pairs": 26549.8,
      "query_h_norm_ratio": 0.4161795198917389,
      "relation_loss": 0.0021263332455419004,
      "selected_pairs": 26549.8,
      "step": 25480,
      "student_distance_mean": 22.68378791809082,
      "teacher_distance_mean": 0.8493653535842896,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0021263332455419004,
      "weighted_relation_loss": 0.0021263332455419004
    },
    {
      "action_query_batch_std": 0.3478917211294174,
      "action_query_norm": 40.721093368530276,
      "action_relation_loss": 0.0020497477846220134,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.84471468662301,
      "flow_matching_loss": 0.016891910042613745,
      "grad_norm": 0.08555004000663757,
      "h_stock_valid_norm": 97.87212524414062,
      "invalid_tail_samples": 20.8,
      "learning_rate": 6.055200129541294e-06,
      "loss": 0.0189,
      "positive_teacher_pairs": 27551.4,
      "query_h_norm_ratio": 0.41606738567352297,
      "relation_loss": 0.0020497477846220134,
      "selected_pairs": 27551.4,
      "step": 25490,
      "student_distance_mean": 22.799495697021484,
      "teacher_distance_mean": 0.8546085596084595,
      "valid_action_samples": 235.2,
      "weighted_auxiliary_loss": 0.0020497477846220134,
      "weighted_relation_loss": 0.0020497477846220134
    },
    {
      "action_query_batch_std": 0.3438145309686661,
      "action_query_norm": 40.701052856445315,
      "action_relation_loss": 0.0019755305256694555,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.854069223573433,
      "flow_matching_loss": 0.016575330309569834,
      "grad_norm": 0.09487071633338928,
      "h_stock_valid_norm": 97.6781509399414,
      "invalid_tail_samples": 24.8,
      "learning_rate": 6.0289359223284434e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 26616.7,
      "query_h_norm_ratio": 0.4166867107152939,
      "relation_loss": 0.0019755305256694555,
      "selected_pairs": 26616.7,
      "step": 25500,
      "student_distance_mean": 22.651385879516603,
      "teacher_distance_mean": 0.848243910074234,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0019755305256694555,
      "weighted_relation_loss": 0.0019755305256694555
    },
    {
      "action_query_batch_std": 0.35188809037208557,
      "action_query_norm": 40.75667839050293,
      "action_relation_loss": 0.0020200424361974,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.863423760523855,
      "flow_matching_loss": 0.01639723777770996,
      "grad_norm": 0.10658877342939377,
      "h_stock_valid_norm": 97.7925193786621,
      "invalid_tail_samples": 22.1,
      "learning_rate": 6.002725144104648e-06,
      "loss": 0.0184,
      "positive_teacher_pairs": 27244.8,
      "query_h_norm_ratio": 0.41677382588386536,
      "relation_loss": 0.0020200424361974,
      "selected_pairs": 27244.8,
      "step": 25510,
      "student_distance_mean": 23.01175174713135,
      "teacher_distance_mean": 0.8631963133811951,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0020200424361974,
      "weighted_relation_loss": 0.0020200424361974
    },
    {
      "action_query_batch_std": 0.34535232484340667,
      "action_query_norm": 40.70101089477539,
      "action_relation_loss": 0.002037016814574599,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.872778297474277,
      "flow_matching_loss": 0.017503343056887388,
      "grad_norm": 0.1305437684059143,
      "h_stock_valid_norm": 97.7208740234375,
      "invalid_tail_samples": 24.0,
      "learning_rate": 5.976567826718477e-06,
      "loss": 0.0195,
      "positive_teacher_pairs": 26802.6,
      "query_h_norm_ratio": 0.4165066808462143,
      "relation_loss": 0.002037016814574599,
      "selected_pairs": 26802.6,
      "step": 25520,
      "student_distance_mean": 22.60199394226074,
      "teacher_distance_mean": 0.8483396530151367,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.002037016814574599,
      "weighted_relation_loss": 0.002037016814574599
    },
    {
      "action_query_batch_std": 0.3497002154588699,
      "action_query_norm": 40.73396797180176,
      "action_relation_loss": 0.0019444869249127805,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.882132834424695,
      "flow_matching_loss": 0.01760655762627721,
      "grad_norm": 0.10572923719882965,
      "h_stock_valid_norm": 97.58830032348632,
      "invalid_tail_samples": 23.1,
      "learning_rate": 5.9504640019535315e-06,
      "loss": 0.0196,
      "positive_teacher_pairs": 27007.7,
      "query_h_norm_ratio": 0.41740704476833346,
      "relation_loss": 0.0019444869249127805,
      "selected_pairs": 27007.7,
      "step": 25530,
      "student_distance_mean": 22.985490798950195,
      "teacher_distance_mean": 0.8622898459434509,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0019444869249127805,
      "weighted_relation_loss": 0.0019444869249127805
    },
    {
      "action_query_batch_std": 0.3468210458755493,
      "action_query_norm": 40.72671279907227,
      "action_relation_loss": 0.0021165146376006306,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.891487371375117,
      "flow_matching_loss": 0.016225191298872232,
      "grad_norm": 0.10818019509315491,
      "h_stock_valid_norm": 97.49924926757812,
      "invalid_tail_samples": 23.4,
      "learning_rate": 5.924413701528414e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 26937.4,
      "query_h_norm_ratio": 0.4177144765853882,
      "relation_loss": 0.0021165146376006306,
      "selected_pairs": 26937.4,
      "step": 25540,
      "student_distance_mean": 22.723182106018065,
      "teacher_distance_mean": 0.8544721662998199,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0021165146376006306,
      "weighted_relation_loss": 0.0021165146376006306
    },
    {
      "action_query_batch_std": 0.34587887227535247,
      "action_query_norm": 40.717094802856444,
      "action_relation_loss": 0.0019995384267531336,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.90084190832554,
      "flow_matching_loss": 0.01709451749920845,
      "grad_norm": 0.10202275216579437,
      "h_stock_valid_norm": 97.43702621459961,
      "invalid_tail_samples": 24.4,
      "learning_rate": 5.898416957096703e-06,
      "loss": 0.0191,
      "positive_teacher_pairs": 26711.8,
      "query_h_norm_ratio": 0.4178875356912613,
      "relation_loss": 0.0019995384267531336,
      "selected_pairs": 26711.8,
      "step": 25550,
      "student_distance_mean": 22.693419647216796,
      "teacher_distance_mean": 0.8529676735401154,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0019995384267531336,
      "weighted_relation_loss": 0.0019995384267531336
    },
    {
      "action_query_batch_std": 0.34647700786590574,
      "action_query_norm": 40.70559730529785,
      "action_relation_loss": 0.0019832113641314207,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.910196445275957,
      "flow_matching_loss": 0.015337708406150341,
      "grad_norm": 0.08590581268072128,
      "h_stock_valid_norm": 97.53593597412109,
      "invalid_tail_samples": 22.7,
      "learning_rate": 5.872473800246913e-06,
      "loss": 0.0173,
      "positive_teacher_pairs": 27112.2,
      "query_h_norm_ratio": 0.41734439432621,
      "relation_loss": 0.0019832113641314207,
      "selected_pairs": 27112.2,
      "step": 25560,
      "student_distance_mean": 22.786660385131835,
      "teacher_distance_mean": 0.855935400724411,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0019832113641314207,
      "weighted_relation_loss": 0.0019832113641314207
    },
    {
      "action_query_batch_std": 0.3449841022491455,
      "action_query_norm": 40.68352432250977,
      "action_relation_loss": 0.0020728703239001334,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.91955098222638,
      "flow_matching_loss": 0.015326428692787886,
      "grad_norm": 0.14282429218292236,
      "h_stock_valid_norm": 97.53402709960938,
      "invalid_tail_samples": 26.3,
      "learning_rate": 5.846584262502402e-06,
      "loss": 0.0174,
      "positive_teacher_pairs": 26278.7,
      "query_h_norm_ratio": 0.41712253987789155,
      "relation_loss": 0.0020728703239001334,
      "selected_pairs": 26278.7,
      "step": 25570,
      "student_distance_mean": 22.586478424072265,
      "teacher_distance_mean": 0.846331512928009,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.0020728703239001334,
      "weighted_relation_loss": 0.0020728703239001334
    },
    {
      "action_query_batch_std": 0.34577019810676574,
      "action_query_norm": 40.73562240600586,
      "action_relation_loss": 0.001970016793347895,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.9289055191768,
      "flow_matching_loss": 0.015436748042702675,
      "grad_norm": 0.10915224999189377,
      "h_stock_valid_norm": 97.67436981201172,
      "invalid_tail_samples": 23.0,
      "learning_rate": 5.820748375321411e-06,
      "loss": 0.0174,
      "positive_teacher_pairs": 27037.7,
      "query_h_norm_ratio": 0.4170562207698822,
      "relation_loss": 0.001970016793347895,
      "selected_pairs": 27037.7,
      "step": 25580,
      "student_distance_mean": 22.64460868835449,
      "teacher_distance_mean": 0.8504914283752442,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.001970016793347895,
      "weighted_relation_loss": 0.001970016793347895
    },
    {
      "action_query_batch_std": 0.34595268666744233,
      "action_query_norm": 40.71394653320313,
      "action_relation_loss": 0.0019737593131139873,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.938260056127223,
      "flow_matching_loss": 0.01620680596679449,
      "grad_norm": 0.10742640495300293,
      "h_stock_valid_norm": 97.7217010498047,
      "invalid_tail_samples": 22.8,
      "learning_rate": 5.794966170096977e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 27087.0,
      "query_h_norm_ratio": 0.41663223803043364,
      "relation_loss": 0.0019737593131139873,
      "selected_pairs": 27087.0,
      "step": 25590,
      "student_distance_mean": 22.624727249145508,
      "teacher_distance_mean": 0.8511089026927948,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0019737593131139873,
      "weighted_relation_loss": 0.0019737593131139873
    },
    {
      "action_query_batch_std": 0.3516565412282944,
      "action_query_norm": 40.74902458190918,
      "action_relation_loss": 0.0020201702252961697,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.947614593077642,
      "flow_matching_loss": 0.016257894411683082,
      "grad_norm": 0.07915398478507996,
      "h_stock_valid_norm": 97.71090316772461,
      "invalid_tail_samples": 22.9,
      "learning_rate": 5.769237678156897e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 27060.3,
      "query_h_norm_ratio": 0.41703979671001434,
      "relation_loss": 0.0020201702252961697,
      "selected_pairs": 27060.3,
      "step": 25600,
      "student_distance_mean": 23.03507423400879,
      "teacher_distance_mean": 0.8650133550167084,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0020201702252961697,
      "weighted_relation_loss": 0.0020201702252961697
    },
    {
      "action_query_batch_std": 0.3492622792720795,
      "action_query_norm": 40.77143516540527,
      "action_relation_loss": 0.00192562056472525,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.956969130028064,
      "flow_matching_loss": 0.015620946325361729,
      "grad_norm": 0.10739689320325851,
      "h_stock_valid_norm": 97.86555557250976,
      "invalid_tail_samples": 21.8,
      "learning_rate": 5.743562930763735e-06,
      "loss": 0.0175,
      "positive_teacher_pairs": 27312.3,
      "query_h_norm_ratio": 0.41660805940628054,
      "relation_loss": 0.00192562056472525,
      "selected_pairs": 27312.3,
      "step": 25610,
      "student_distance_mean": 22.938637351989748,
      "teacher_distance_mean": 0.8625989258289337,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.00192562056472525,
      "weighted_relation_loss": 0.00192562056472525
    },
    {
      "action_query_batch_std": 0.3478145897388458,
      "action_query_norm": 40.70330123901367,
      "action_relation_loss": 0.0019515263382345438,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.966323666978486,
      "flow_matching_loss": 0.01667455621063709,
      "grad_norm": 0.09567393362522125,
      "h_stock_valid_norm": 97.91506729125976,
      "invalid_tail_samples": 23.3,
      "learning_rate": 5.717941959114726e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 26976.4,
      "query_h_norm_ratio": 0.41570245027542113,
      "relation_loss": 0.0019515263382345438,
      "selected_pairs": 26976.4,
      "step": 25620,
      "student_distance_mean": 22.79834117889404,
      "teacher_distance_mean": 0.8557830631732941,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0019515263382345438,
      "weighted_relation_loss": 0.0019515263382345438
    },
    {
      "action_query_batch_std": 0.34473452568054197,
      "action_query_norm": 40.70646286010742,
      "action_relation_loss": 0.0019551914534531533,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.975678203928904,
      "flow_matching_loss": 0.016045672167092562,
      "grad_norm": 0.12905119359493256,
      "h_stock_valid_norm": 97.86812820434571,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.69237479434176e-06,
      "loss": 0.018,
      "positive_teacher_pairs": 26990.2,
      "query_h_norm_ratio": 0.41593667566776277,
      "relation_loss": 0.0019551914534531533,
      "selected_pairs": 26990.2,
      "step": 25630,
      "student_distance_mean": 22.7333044052124,
      "teacher_distance_mean": 0.8535255134105683,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0019551914534531533,
      "weighted_relation_loss": 0.0019551914534531533
    },
    {
      "action_query_batch_std": 0.35136068761348727,
      "action_query_norm": 40.7225528717041,
      "action_relation_loss": 0.002021171746309847,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.985032740879326,
      "flow_matching_loss": 0.019478092435747385,
      "grad_norm": 0.1098906472325325,
      "h_stock_valid_norm": 97.80858535766602,
      "invalid_tail_samples": 22.0,
      "learning_rate": 5.666861467511353e-06,
      "loss": 0.0215,
      "positive_teacher_pairs": 27268.8,
      "query_h_norm_ratio": 0.41635343730449675,
      "relation_loss": 0.002021171746309847,
      "selected_pairs": 27268.8,
      "step": 25640,
      "student_distance_mean": 23.028049278259278,
      "teacher_distance_mean": 0.8645212650299072,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.002021171746309847,
      "weighted_relation_loss": 0.002021171746309847
    },
    {
      "action_query_batch_std": 0.34623396396636963,
      "action_query_norm": 40.71686019897461,
      "action_relation_loss": 0.001997164427302778,
      "auxiliary_loss_weight": 1.0,
      "epoch": 23.99438727782975,
      "flow_matching_loss": 0.01776417689397931,
      "grad_norm": 0.12627212703227997,
      "h_stock_valid_norm": 97.45456085205078,
      "invalid_tail_samples": 24.4,
      "learning_rate": 5.641402009624591e-06,
      "loss": 0.0198,
      "positive_teacher_pairs": 26721.6,
      "query_h_norm_ratio": 0.4178058922290802,
      "relation_loss": 0.001997164427302778,
      "selected_pairs": 26721.6,
      "step": 25650,
      "student_distance_mean": 22.727458953857422,
      "teacher_distance_mean": 0.8540994584560394,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.001997164427302778,
      "weighted_relation_loss": 0.001997164427302778
    },
    {
      "action_query_batch_std": 0.3471417874097824,
      "action_query_norm": 40.71430625915527,
      "action_relation_loss": 0.0019335932796821,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.003741814780167,
      "flow_matching_loss": 0.017708856705576183,
      "grad_norm": 0.09659254550933838,
      "h_stock_valid_norm": 97.63247985839844,
      "invalid_tail_samples": 22.6,
      "learning_rate": 5.615996451617145e-06,
      "loss": 0.0196,
      "positive_teacher_pairs": 24154.5,
      "query_h_norm_ratio": 0.4170175164937973,
      "relation_loss": 0.0019335932796821,
      "selected_pairs": 24154.5,
      "step": 25660,
      "student_distance_mean": 22.870930480957032,
      "teacher_distance_mean": 0.8605303823947906,
      "valid_action_samples": 213.5,
      "weighted_auxiliary_loss": 0.0019335932796821,
      "weighted_relation_loss": 0.0019335932796821
    },
    {
      "action_query_batch_std": 0.34915971159935,
      "action_query_norm": 40.73521347045899,
      "action_relation_loss": 0.001993447553832084,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.01309635173059,
      "flow_matching_loss": 0.017183942720294,
      "grad_norm": 0.10152506083250046,
      "h_stock_valid_norm": 97.6190689086914,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.590644824359148e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 26990.7,
      "query_h_norm_ratio": 0.4172880232334137,
      "relation_loss": 0.001993447553832084,
      "selected_pairs": 26990.7,
      "step": 25670,
      "student_distance_mean": 22.986635780334474,
      "teacher_distance_mean": 0.8648000299930573,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.001993447553832084,
      "weighted_relation_loss": 0.001993447553832084
    },
    {
      "action_query_batch_std": 0.34089582562446596,
      "action_query_norm": 40.67399024963379,
      "action_relation_loss": 0.001972186774946749,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.02245088868101,
      "flow_matching_loss": 0.0169203688390553,
      "grad_norm": 0.10619154572486877,
      "h_stock_valid_norm": 97.62919464111329,
      "invalid_tail_samples": 23.4,
      "learning_rate": 5.56534715865521e-06,
      "loss": 0.0189,
      "positive_teacher_pairs": 26950.9,
      "query_h_norm_ratio": 0.41661970019340516,
      "relation_loss": 0.001972186774946749,
      "selected_pairs": 26950.9,
      "step": 25680,
      "student_distance_mean": 22.44938144683838,
      "teacher_distance_mean": 0.8440729796886444,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.001972186774946749,
      "weighted_relation_loss": 0.001972186774946749
    },
    {
      "action_query_batch_std": 0.34433724284172057,
      "action_query_norm": 40.71342163085937,
      "action_relation_loss": 0.002019184653181583,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.031805425631433,
      "flow_matching_loss": 0.01640786975622177,
      "grad_norm": 0.08012671023607254,
      "h_stock_valid_norm": 97.70212936401367,
      "invalid_tail_samples": 23.7,
      "learning_rate": 5.5401034852443965e-06,
      "loss": 0.0184,
      "positive_teacher_pairs": 26873.9,
      "query_h_norm_ratio": 0.4167117357254028,
      "relation_loss": 0.002019184653181583,
      "selected_pairs": 26873.9,
      "step": 25690,
      "student_distance_mean": 22.61585521697998,
      "teacher_distance_mean": 0.8506939768791199,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.002019184653181583,
      "weighted_relation_loss": 0.002019184653181583
    },
    {
      "action_query_batch_std": 0.349028667807579,
      "action_query_norm": 40.73986701965332,
      "action_relation_loss": 0.0019713807851076126,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.04115996258185,
      "flow_matching_loss": 0.01582814184948802,
      "grad_norm": 0.08994711935520172,
      "h_stock_valid_norm": 97.86076965332032,
      "invalid_tail_samples": 26.3,
      "learning_rate": 5.514913834800134e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 26285.0,
      "query_h_norm_ratio": 0.4163107484579086,
      "relation_loss": 0.0019713807851076126,
      "selected_pairs": 26285.0,
      "step": 25700,
      "student_distance_mean": 22.952449607849122,
      "teacher_distance_mean": 0.8630145847797394,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.0019713807851076126,
      "weighted_relation_loss": 0.0019713807851076126
    },
    {
      "action_query_batch_std": 0.35002942383289337,
      "action_query_norm": 40.74540824890137,
      "action_relation_loss": 0.0019422117271460592,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.050514499532273,
      "flow_matching_loss": 0.0177392085082829,
      "grad_norm": 0.09202460199594498,
      "h_stock_valid_norm": 97.77011108398438,
      "invalid_tail_samples": 23.1,
      "learning_rate": 5.489778237930238e-06,
      "loss": 0.0197,
      "positive_teacher_pairs": 27012.2,
      "query_h_norm_ratio": 0.41674855053424836,
      "relation_loss": 0.0019422117271460592,
      "selected_pairs": 27012.2,
      "step": 25710,
      "student_distance_mean": 23.02626838684082,
      "teacher_distance_mean": 0.8656353533267975,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0019422117271460592,
      "weighted_relation_loss": 0.0019422117271460592
    },
    {
      "action_query_batch_std": 0.346309033036232,
      "action_query_norm": 40.700704193115236,
      "action_relation_loss": 0.00195218255976215,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.059869036482695,
      "flow_matching_loss": 0.016333051957190037,
      "grad_norm": 0.0929916724562645,
      "h_stock_valid_norm": 97.7328872680664,
      "invalid_tail_samples": 25.6,
      "learning_rate": 5.464696725176827e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 26459.1,
      "query_h_norm_ratio": 0.41644851863384247,
      "relation_loss": 0.00195218255976215,
      "selected_pairs": 26459.1,
      "step": 25720,
      "student_distance_mean": 22.820880699157716,
      "teacher_distance_mean": 0.8549070119857788,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.00195218255976215,
      "weighted_relation_loss": 0.00195218255976215
    },
    {
      "action_query_batch_std": 0.35008967816829684,
      "action_query_norm": 40.76154098510742,
      "action_relation_loss": 0.002014702802989632,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.069223573433113,
      "flow_matching_loss": 0.014757737889885903,
      "grad_norm": 0.12217307835817337,
      "h_stock_valid_norm": 97.8601287841797,
      "invalid_tail_samples": 21.2,
      "learning_rate": 5.439669327016294e-06,
      "loss": 0.0168,
      "positive_teacher_pairs": 27454.3,
      "query_h_norm_ratio": 0.41653292775154116,
      "relation_loss": 0.002014702802989632,
      "selected_pairs": 27454.3,
      "step": 25730,
      "student_distance_mean": 23.013442993164062,
      "teacher_distance_mean": 0.8614897072315216,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.002014702802989632,
      "weighted_relation_loss": 0.002014702802989632
    },
    {
      "action_query_batch_std": 0.34431433379650117,
      "action_query_norm": 40.70149269104004,
      "action_relation_loss": 0.0020135967060923576,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.078578110383535,
      "flow_matching_loss": 0.016312819626182317,
      "grad_norm": 0.08834051340818405,
      "h_stock_valid_norm": 97.71073837280274,
      "invalid_tail_samples": 22.8,
      "learning_rate": 5.4146960738592985e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 27090.9,
      "query_h_norm_ratio": 0.41655333936214445,
      "relation_loss": 0.0020135967060923576,
      "selected_pairs": 27090.9,
      "step": 25740,
      "student_distance_mean": 22.632842445373534,
      "teacher_distance_mean": 0.8463819861412049,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0020135967060923576,
      "weighted_relation_loss": 0.0020135967060923576
    },
    {
      "action_query_batch_std": 0.3505248546600342,
      "action_query_norm": 40.76031227111817,
      "action_relation_loss": 0.001892353326547891,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.087932647333957,
      "flow_matching_loss": 0.015921211149543524,
      "grad_norm": 0.09076981991529465,
      "h_stock_valid_norm": 97.91356506347657,
      "invalid_tail_samples": 25.9,
      "learning_rate": 5.389776996050694e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 26375.8,
      "query_h_norm_ratio": 0.4162963002920151,
      "relation_loss": 0.001892353326547891,
      "selected_pairs": 26375.8,
      "step": 25750,
      "student_distance_mean": 23.13095703125,
      "teacher_distance_mean": 0.8673052668571473,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.001892353326547891,
      "weighted_relation_loss": 0.001892353326547891
    },
    {
      "action_query_batch_std": 0.3471780240535736,
      "action_query_norm": 40.707213973999025,
      "action_relation_loss": 0.002014081121888012,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.09728718428438,
      "flow_matching_loss": 0.016268499288707973,
      "grad_norm": 0.07307583838701248,
      "h_stock_valid_norm": 97.69912643432617,
      "invalid_tail_samples": 24.4,
      "learning_rate": 5.364912123869492e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 26707.1,
      "query_h_norm_ratio": 0.4166618317365646,
      "relation_loss": 0.002014081121888012,
      "selected_pairs": 26707.1,
      "step": 25760,
      "student_distance_mean": 22.728616142272948,
      "teacher_distance_mean": 0.8522765278816223,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.002014081121888012,
      "weighted_relation_loss": 0.002014081121888012
    },
    {
      "action_query_batch_std": 0.34111957252025604,
      "action_query_norm": 40.65851974487305,
      "action_relation_loss": 0.002071499393787235,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.106641721234798,
      "flow_matching_loss": 0.015187492314726114,
      "grad_norm": 0.06500968337059021,
      "h_stock_valid_norm": 97.77426452636719,
      "invalid_tail_samples": 25.3,
      "learning_rate": 5.3401014875288864e-06,
      "loss": 0.0173,
      "positive_teacher_pairs": 26504.7,
      "query_h_norm_ratio": 0.41584297716617585,
      "relation_loss": 0.002071499393787235,
      "selected_pairs": 26504.7,
      "step": 25770,
      "student_distance_mean": 22.419452095031737,
      "teacher_distance_mean": 0.8424722671508789,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.002071499393787235,
      "weighted_relation_loss": 0.002071499393787235
    },
    {
      "action_query_batch_std": 0.3469307482242584,
      "action_query_norm": 40.7313175201416,
      "action_relation_loss": 0.002028310520108789,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.11599625818522,
      "flow_matching_loss": 0.018222783971577883,
      "grad_norm": 0.1699662059545517,
      "h_stock_valid_norm": 97.8963996887207,
      "invalid_tail_samples": 21.0,
      "learning_rate": 5.315345117176129e-06,
      "loss": 0.0203,
      "positive_teacher_pairs": 27503.5,
      "query_h_norm_ratio": 0.41606766879558565,
      "relation_loss": 0.002028310520108789,
      "selected_pairs": 27503.5,
      "step": 25780,
      "student_distance_mean": 22.65946235656738,
      "teacher_distance_mean": 0.8513547897338867,
      "valid_action_samples": 235.0,
      "weighted_auxiliary_loss": 0.002028310520108789,
      "weighted_relation_loss": 0.002028310520108789
    },
    {
      "action_query_batch_std": 0.34675362408161164,
      "action_query_norm": 40.735086059570314,
      "action_relation_loss": 0.0019140461809001863,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.125350795135642,
      "flow_matching_loss": 0.015601447131484748,
      "grad_norm": 0.08746389299631119,
      "h_stock_valid_norm": 97.93358459472657,
      "invalid_tail_samples": 25.6,
      "learning_rate": 5.2906430428925405e-06,
      "loss": 0.0175,
      "positive_teacher_pairs": 26430.9,
      "query_h_norm_ratio": 0.4159483343362808,
      "relation_loss": 0.0019140461809001863,
      "selected_pairs": 26430.9,
      "step": 25790,
      "student_distance_mean": 22.819921493530273,
      "teacher_distance_mean": 0.8587824761867523,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0019140461809001863,
      "weighted_relation_loss": 0.0019140461809001863
    },
    {
      "action_query_batch_std": 0.3432715445756912,
      "action_query_norm": 40.6989070892334,
      "action_relation_loss": 0.0020112254773266614,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.13470533208606,
      "flow_matching_loss": 0.016549435164779424,
      "grad_norm": 0.10275005549192429,
      "h_stock_valid_norm": 98.07846221923828,
      "invalid_tail_samples": 23.1,
      "learning_rate": 5.265995294693471e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 27016.5,
      "query_h_norm_ratio": 0.41496547758579255,
      "relation_loss": 0.0020112254773266614,
      "selected_pairs": 27016.5,
      "step": 25800,
      "student_distance_mean": 22.626646995544434,
      "teacher_distance_mean": 0.848939198255539,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0020112254773266614,
      "weighted_relation_loss": 0.0020112254773266614
    },
    {
      "action_query_batch_std": 0.3449310749769211,
      "action_query_norm": 40.68340072631836,
      "action_relation_loss": 0.001987033779732883,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.144059869036482,
      "flow_matching_loss": 0.014683582354336977,
      "grad_norm": 0.15339034795761108,
      "h_stock_valid_norm": 98.03788528442382,
      "invalid_tail_samples": 25.9,
      "learning_rate": 5.241401902528254e-06,
      "loss": 0.0167,
      "positive_teacher_pairs": 26362.7,
      "query_h_norm_ratio": 0.41497922241687774,
      "relation_loss": 0.001987033779732883,
      "selected_pairs": 26362.7,
      "step": 25810,
      "student_distance_mean": 22.636153602600096,
      "teacher_distance_mean": 0.8507691502571106,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.001987033779732883,
      "weighted_relation_loss": 0.001987033779732883
    },
    {
      "action_query_batch_std": 0.3498484790325165,
      "action_query_norm": 40.73987731933594,
      "action_relation_loss": 0.0019625841290690006,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.153414405986904,
      "flow_matching_loss": 0.01692827856168151,
      "grad_norm": 0.16647562384605408,
      "h_stock_valid_norm": 97.8500144958496,
      "invalid_tail_samples": 23.8,
      "learning_rate": 5.216862896280189e-06,
      "loss": 0.0189,
      "positive_teacher_pairs": 26848.0,
      "query_h_norm_ratio": 0.4163508892059326,
      "relation_loss": 0.0019625841290690006,
      "selected_pairs": 26848.0,
      "step": 25820,
      "student_distance_mean": 22.998246002197266,
      "teacher_distance_mean": 0.8640395522117614,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0019625841290690006,
      "weighted_relation_loss": 0.0019625841290690006
    },
    {
      "action_query_batch_std": 0.34485094249248505,
      "action_query_norm": 40.70074195861817,
      "action_relation_loss": 0.0019585160771384835,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.162768942937326,
      "flow_matching_loss": 0.016454536374658345,
      "grad_norm": 0.11128968000411987,
      "h_stock_valid_norm": 97.90480194091796,
      "invalid_tail_samples": 24.3,
      "learning_rate": 5.192378305766471e-06,
      "loss": 0.0184,
      "positive_teacher_pairs": 26735.3,
      "query_h_norm_ratio": 0.4157203435897827,
      "relation_loss": 0.0019585160771384835,
      "selected_pairs": 26735.3,
      "step": 25830,
      "student_distance_mean": 22.67492332458496,
      "teacher_distance_mean": 0.8521020889282227,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0019585160771384835,
      "weighted_relation_loss": 0.0019585160771384835
    },
    {
      "action_query_batch_std": 0.344538813829422,
      "action_query_norm": 40.71064910888672,
      "action_relation_loss": 0.0019466530880890786,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.172123479887745,
      "flow_matching_loss": 0.01741872075945139,
      "grad_norm": 0.1402021199464798,
      "h_stock_valid_norm": 97.99424667358399,
      "invalid_tail_samples": 24.8,
      "learning_rate": 5.1679481607382065e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 26625.9,
      "query_h_norm_ratio": 0.41544193029403687,
      "relation_loss": 0.0019466530880890786,
      "selected_pairs": 26625.9,
      "step": 25840,
      "student_distance_mean": 22.687520217895507,
      "teacher_distance_mean": 0.8503529310226441,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0019466530880890786,
      "weighted_relation_loss": 0.0019466530880890786
    },
    {
      "action_query_batch_std": 0.34359546303749083,
      "action_query_norm": 40.71679420471192,
      "action_relation_loss": 0.0019782942486926912,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.181478016838167,
      "flow_matching_loss": 0.014504733216017484,
      "grad_norm": 0.0942864939570427,
      "h_stock_valid_norm": 97.83588485717773,
      "invalid_tail_samples": 25.3,
      "learning_rate": 5.1435724908803105e-06,
      "loss": 0.0165,
      "positive_teacher_pairs": 26507.2,
      "query_h_norm_ratio": 0.41617511510849,
      "relation_loss": 0.0019782942486926912,
      "selected_pairs": 26507.2,
      "step": 25850,
      "student_distance_mean": 22.73740291595459,
      "teacher_distance_mean": 0.853841507434845,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0019782942486926912,
      "weighted_relation_loss": 0.0019782942486926912
    },
    {
      "action_query_batch_std": 0.3430435746908188,
      "action_query_norm": 40.68876838684082,
      "action_relation_loss": 0.0020033957320265473,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.19083255378859,
      "flow_matching_loss": 0.017040968127548694,
      "grad_norm": 0.1121753677725792,
      "h_stock_valid_norm": 97.98733673095703,
      "invalid_tail_samples": 25.5,
      "learning_rate": 5.119251325811514e-06,
      "loss": 0.019,
      "positive_teacher_pairs": 26467.7,
      "query_h_norm_ratio": 0.4152481287717819,
      "relation_loss": 0.0020033957320265473,
      "selected_pairs": 26467.7,
      "step": 25860,
      "student_distance_mean": 22.530857849121094,
      "teacher_distance_mean": 0.8454391241073609,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0020033957320265473,
      "weighted_relation_loss": 0.0020033957320265473
    },
    {
      "action_query_batch_std": 0.3515368402004242,
      "action_query_norm": 40.74988059997558,
      "action_relation_loss": 0.0019404142163693904,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.200187090739007,
      "flow_matching_loss": 0.015482317935675383,
      "grad_norm": 0.12056808918714523,
      "h_stock_valid_norm": 97.64740676879883,
      "invalid_tail_samples": 22.1,
      "learning_rate": 5.094984695084348e-06,
      "loss": 0.0174,
      "positive_teacher_pairs": 27246.6,
      "query_h_norm_ratio": 0.41731933653354647,
      "relation_loss": 0.0019404142163693904,
      "selected_pairs": 27246.6,
      "step": 25870,
      "student_distance_mean": 23.01334743499756,
      "teacher_distance_mean": 0.8655564248561859,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0019404142163693904,
      "weighted_relation_loss": 0.0019404142163693904
    },
    {
      "action_query_batch_std": 0.3505537658929825,
      "action_query_norm": 40.734422302246095,
      "action_relation_loss": 0.0019447624799795449,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.20954162768943,
      "flow_matching_loss": 0.017292108107358217,
      "grad_norm": 0.128386452794075,
      "h_stock_valid_norm": 97.67241821289062,
      "invalid_tail_samples": 25.2,
      "learning_rate": 5.070772628185039e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 26524.4,
      "query_h_norm_ratio": 0.4170531749725342,
      "relation_loss": 0.0019447624799795449,
      "selected_pairs": 26524.4,
      "step": 25880,
      "student_distance_mean": 22.864990043640137,
      "teacher_distance_mean": 0.8591930329799652,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0019447624799795449,
      "weighted_relation_loss": 0.0019447624799795449
    },
    {
      "action_query_batch_std": 0.34945446252822876,
      "action_query_norm": 40.752587509155276,
      "action_relation_loss": 0.001901188085321337,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.21889616463985,
      "flow_matching_loss": 0.01628040401265025,
      "grad_norm": 0.07819628715515137,
      "h_stock_valid_norm": 97.81086883544921,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.046615154533535e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 26999.5,
      "query_h_norm_ratio": 0.41664821207523345,
      "relation_loss": 0.001901188085321337,
      "selected_pairs": 26999.5,
      "step": 25890,
      "student_distance_mean": 22.920051193237306,
      "teacher_distance_mean": 0.8622937560081482,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.001901188085321337,
      "weighted_relation_loss": 0.001901188085321337
    },
    {
      "action_query_batch_std": 0.34915741086006163,
      "action_query_norm": 40.740840148925784,
      "action_relation_loss": 0.001965860277414322,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.22825070159027,
      "flow_matching_loss": 0.01616215715184808,
      "grad_norm": 0.0906200036406517,
      "h_stock_valid_norm": 97.62332077026367,
      "invalid_tail_samples": 25.4,
      "learning_rate": 5.022512303483451e-06,
      "loss": 0.0181,
      "positive_teacher_pairs": 26480.8,
      "query_h_norm_ratio": 0.41732709407806395,
      "relation_loss": 0.001965860277414322,
      "selected_pairs": 26480.8,
      "step": 25900,
      "student_distance_mean": 22.832813453674316,
      "teacher_distance_mean": 0.8589461863040924,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.001965860277414322,
      "weighted_relation_loss": 0.001965860277414322
    },
    {
      "action_query_batch_std": 0.344023260474205,
      "action_query_norm": 40.66886825561524,
      "action_relation_loss": 0.0019650321686640383,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.23760523854069,
      "flow_matching_loss": 0.017529718484729527,
      "grad_norm": 0.09821547567844391,
      "h_stock_valid_norm": 97.60243606567383,
      "invalid_tail_samples": 24.2,
      "learning_rate": 4.998464104322015e-06,
      "loss": 0.0195,
      "positive_teacher_pairs": 26760.8,
      "query_h_norm_ratio": 0.4166809320449829,
      "relation_loss": 0.0019650321686640383,
      "selected_pairs": 26760.8,
      "step": 25910,
      "student_distance_mean": 22.533359336853028,
      "teacher_distance_mean": 0.8461904644966125,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0019650321686640383,
      "weighted_relation_loss": 0.0019650321686640383
    },
    {
      "action_query_batch_std": 0.3471950054168701,
      "action_query_norm": 40.7264461517334,
      "action_relation_loss": 0.001981324318330735,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.246959775491113,
      "flow_matching_loss": 0.01567436596378684,
      "grad_norm": 0.09350208938121796,
      "h_stock_valid_norm": 97.61920471191407,
      "invalid_tail_samples": 24.6,
      "learning_rate": 4.974470586270047e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 26667.5,
      "query_h_norm_ratio": 0.41719890832901,
      "relation_loss": 0.001981324318330735,
      "selected_pairs": 26667.5,
      "step": 25920,
      "student_distance_mean": 22.777193069458008,
      "teacher_distance_mean": 0.8562030851840973,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.001981324318330735,
      "weighted_relation_loss": 0.001981324318330735
    },
    {
      "action_query_batch_std": 0.34864521622657774,
      "action_query_norm": 40.710225677490236,
      "action_relation_loss": 0.0019243414513766765,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.256314312441535,
      "flow_matching_loss": 0.01563436249271035,
      "grad_norm": 0.0863773450255394,
      "h_stock_valid_norm": 97.74206924438477,
      "invalid_tail_samples": 21.7,
      "learning_rate": 4.950531778481965e-06,
      "loss": 0.0176,
      "positive_teacher_pairs": 27333.7,
      "query_h_norm_ratio": 0.4165085583925247,
      "relation_loss": 0.0019243414513766765,
      "selected_pairs": 27333.7,
      "step": 25930,
      "student_distance_mean": 22.888291931152345,
      "teacher_distance_mean": 0.8590570867061615,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0019243414513766765,
      "weighted_relation_loss": 0.0019243414513766765
    },
    {
      "action_query_batch_std": 0.345610174536705,
      "action_query_norm": 40.71420860290527,
      "action_relation_loss": 0.002011643152218312,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.265668849391954,
      "flow_matching_loss": 0.01637981813400984,
      "grad_norm": 0.09635227173566818,
      "h_stock_valid_norm": 97.6932975769043,
      "invalid_tail_samples": 23.2,
      "learning_rate": 4.926647710045651e-06,
      "loss": 0.0184,
      "positive_teacher_pairs": 27007.0,
      "query_h_norm_ratio": 0.4167572855949402,
      "relation_loss": 0.002011643152218312,
      "selected_pairs": 27007.0,
      "step": 25940,
      "student_distance_mean": 22.71100902557373,
      "teacher_distance_mean": 0.8521972894668579,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.002011643152218312,
      "weighted_relation_loss": 0.002011643152218312
    },
    {
      "action_query_batch_std": 0.34620143175125123,
      "action_query_norm": 40.680212783813474,
      "action_relation_loss": 0.0019508983823470771,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.275023386342376,
      "flow_matching_loss": 0.015987034328281878,
      "grad_norm": 0.10448918491601944,
      "h_stock_valid_norm": 97.77350616455078,
      "invalid_tail_samples": 25.1,
      "learning_rate": 4.9028184099825125e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 26547.5,
      "query_h_norm_ratio": 0.4160673290491104,
      "relation_loss": 0.0019508983823470771,
      "selected_pairs": 26547.5,
      "step": 25950,
      "student_distance_mean": 22.733317184448243,
      "teacher_distance_mean": 0.8535877704620362,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0019508983823470771,
      "weighted_relation_loss": 0.0019508983823470771
    },
    {
      "action_query_batch_std": 0.3442567765712738,
      "action_query_norm": 40.68384056091308,
      "action_relation_loss": 0.001967011217493564,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.284377923292798,
      "flow_matching_loss": 0.015889668092131614,
      "grad_norm": 0.10403163731098175,
      "h_stock_valid_norm": 97.72660064697266,
      "invalid_tail_samples": 23.8,
      "learning_rate": 4.879043907247382e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 26857.4,
      "query_h_norm_ratio": 0.41630509197711946,
      "relation_loss": 0.001967011217493564,
      "selected_pairs": 26857.4,
      "step": 25960,
      "student_distance_mean": 22.593253326416015,
      "teacher_distance_mean": 0.8483340620994568,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.001967011217493564,
      "weighted_relation_loss": 0.001967011217493564
    },
    {
      "action_query_batch_std": 0.3472143203020096,
      "action_query_norm": 40.7003849029541,
      "action_relation_loss": 0.0019553775317035617,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.293732460243216,
      "flow_matching_loss": 0.017201354075223208,
      "grad_norm": 0.09984355419874191,
      "h_stock_valid_norm": 97.88954467773438,
      "invalid_tail_samples": 23.2,
      "learning_rate": 4.8553242307285414e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 26987.7,
      "query_h_norm_ratio": 0.4157818168401718,
      "relation_loss": 0.0019553775317035617,
      "selected_pairs": 26987.7,
      "step": 25970,
      "student_distance_mean": 22.764182472229002,
      "teacher_distance_mean": 0.8549225747585296,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0019553775317035617,
      "weighted_relation_loss": 0.0019553775317035617
    },
    {
      "action_query_batch_std": 0.3466362476348877,
      "action_query_norm": 40.704851913452146,
      "action_relation_loss": 0.002032322040759027,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.30308699719364,
      "flow_matching_loss": 0.01551086874678731,
      "grad_norm": 0.07280497997999191,
      "h_stock_valid_norm": 97.74346313476562,
      "invalid_tail_samples": 25.7,
      "learning_rate": 4.8316594092476185e-06,
      "loss": 0.0175,
      "positive_teacher_pairs": 26414.4,
      "query_h_norm_ratio": 0.4164490312337875,
      "relation_loss": 0.002032322040759027,
      "selected_pairs": 26414.4,
      "step": 25980,
      "student_distance_mean": 22.803600120544434,
      "teacher_distance_mean": 0.8590709149837494,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.002032322040759027,
      "weighted_relation_loss": 0.002032322040759027
    },
    {
      "action_query_batch_std": 0.3480400323867798,
      "action_query_norm": 40.7234504699707,
      "action_relation_loss": 0.0018812108668498696,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.31244153414406,
      "flow_matching_loss": 0.016731508169323207,
      "grad_norm": 0.08858615905046463,
      "h_stock_valid_norm": 97.8927734375,
      "invalid_tail_samples": 22.3,
      "learning_rate": 4.808049471559617e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 27194.8,
      "query_h_norm_ratio": 0.4160026103258133,
      "relation_loss": 0.0018812108668498696,
      "selected_pairs": 27194.8,
      "step": 25990,
      "student_distance_mean": 22.78453254699707,
      "teacher_distance_mean": 0.8564530551433563,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0018812108668498696,
      "weighted_relation_loss": 0.0018812108668498696
    },
    {
      "action_query_batch_std": 0.34265189468860624,
      "action_query_norm": 40.67395286560058,
      "action_relation_loss": 0.0021226508310064674,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.321796071094482,
      "flow_matching_loss": 0.016473021917045118,
      "grad_norm": 0.09409723430871964,
      "h_stock_valid_norm": 97.87540664672852,
      "invalid_tail_samples": 21.9,
      "learning_rate": 4.784494446352833e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 27298.0,
      "query_h_norm_ratio": 0.41557046175003054,
      "relation_loss": 0.0021226508310064674,
      "selected_pairs": 27298.0,
      "step": 26000,
      "student_distance_mean": 22.410684394836426,
      "teacher_distance_mean": 0.8424726307392121,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.0021226508310064674,
      "weighted_relation_loss": 0.0021226508310064674
    },
    {
      "action_query_batch_std": 0.3435136079788208,
      "action_query_norm": 40.70029182434082,
      "action_relation_loss": 0.001954959274735302,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.3311506080449,
      "flow_matching_loss": 0.01650980208069086,
      "grad_norm": 0.1942884773015976,
      "h_stock_valid_norm": 97.95198440551758,
      "invalid_tail_samples": 25.5,
      "learning_rate": 4.760994362248833e-06,
      "loss": 0.0185,
      "positive_teacher_pairs": 26464.2,
      "query_h_norm_ratio": 0.41551493704319,
      "relation_loss": 0.001954959274735302,
      "selected_pairs": 26464.2,
      "step": 26010,
      "student_distance_mean": 22.52806854248047,
      "teacher_distance_mean": 0.8459527432918549,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.001954959274735302,
      "weighted_relation_loss": 0.001954959274735302
    },
    {
      "action_query_batch_std": 0.3462091952562332,
      "action_query_norm": 40.6985668182373,
      "action_relation_loss": 0.0019259684486314654,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.340505144995323,
      "flow_matching_loss": 0.015092043019831181,
      "grad_norm": 0.09341267496347427,
      "h_stock_valid_norm": 97.90926818847656,
      "invalid_tail_samples": 22.8,
      "learning_rate": 4.737549247802442e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 27081.7,
      "query_h_norm_ratio": 0.41567806899547577,
      "relation_loss": 0.0019259684486314654,
      "selected_pairs": 27081.7,
      "step": 26020,
      "student_distance_mean": 22.72015495300293,
      "teacher_distance_mean": 0.8539652526378632,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0019259684486314654,
      "weighted_relation_loss": 0.0019259684486314654
    },
    {
      "action_query_batch_std": 0.35169881880283355,
      "action_query_norm": 40.714224624633786,
      "action_relation_loss": 0.0019657825818285347,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.349859681945745,
      "flow_matching_loss": 0.01570819616317749,
      "grad_norm": 0.10083052515983582,
      "h_stock_valid_norm": 97.93196868896484,
      "invalid_tail_samples": 25.8,
      "learning_rate": 4.714159131501689e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 26393.5,
      "query_h_norm_ratio": 0.4157432347536087,
      "relation_loss": 0.0019657825818285347,
      "selected_pairs": 26393.5,
      "step": 26030,
      "student_distance_mean": 23.187365531921387,
      "teacher_distance_mean": 0.8706495463848114,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0019657825818285347,
      "weighted_relation_loss": 0.0019657825818285347
    },
    {
      "action_query_batch_std": 0.34823534190654754,
      "action_query_norm": 40.72103042602539,
      "action_relation_loss": 0.0019479844602756202,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.359214218896163,
      "flow_matching_loss": 0.014921520464122295,
      "grad_norm": 0.0936526283621788,
      "h_stock_valid_norm": 97.93578567504883,
      "invalid_tail_samples": 24.0,
      "learning_rate": 4.690824041767766e-06,
      "loss": 0.0169,
      "positive_teacher_pairs": 26807.1,
      "query_h_norm_ratio": 0.4157951533794403,
      "relation_loss": 0.0019479844602756202,
      "selected_pairs": 26807.1,
      "step": 26040,
      "student_distance_mean": 22.935769271850585,
      "teacher_distance_mean": 0.8628329813480378,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0019479844602756202,
      "weighted_relation_loss": 0.0019479844602756202
    },
    {
      "action_query_batch_std": 0.3495925575494766,
      "action_query_norm": 40.715749740600586,
      "action_relation_loss": 0.0019820838468149306,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.368568755846585,
      "flow_matching_loss": 0.015104941744357347,
      "grad_norm": 0.10580452531576157,
      "h_stock_valid_norm": 98.05875396728516,
      "invalid_tail_samples": 22.9,
      "learning_rate": 4.6675440069550105e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 27058.9,
      "query_h_norm_ratio": 0.4152214348316193,
      "relation_loss": 0.0019820838468149306,
      "selected_pairs": 27058.9,
      "step": 26050,
      "student_distance_mean": 23.007563591003418,
      "teacher_distance_mean": 0.8653549134731293,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0019820838468149306,
      "weighted_relation_loss": 0.0019820838468149306
    },
    {
      "action_query_batch_std": 0.34522143602371214,
      "action_query_norm": 40.717987823486325,
      "action_relation_loss": 0.001903063023928553,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.377923292797007,
      "flow_matching_loss": 0.016122500970959665,
      "grad_norm": 0.12376508861780167,
      "h_stock_valid_norm": 97.98771057128906,
      "invalid_tail_samples": 23.0,
      "learning_rate": 4.64431905535086e-06,
      "loss": 0.018,
      "positive_teacher_pairs": 27034.8,
      "query_h_norm_ratio": 0.415543732047081,
      "relation_loss": 0.001903063023928553,
      "selected_pairs": 27034.8,
      "step": 26060,
      "student_distance_mean": 22.622614669799805,
      "teacher_distance_mean": 0.852598637342453,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.001903063023928553,
      "weighted_relation_loss": 0.001903063023928553
    },
    {
      "action_query_batch_std": 0.3472137749195099,
      "action_query_norm": 40.687811279296874,
      "action_relation_loss": 0.0019996139802969993,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.38727782974743,
      "flow_matching_loss": 0.016737152077257633,
      "grad_norm": 0.11582843214273453,
      "h_stock_valid_norm": 98.03446197509766,
      "invalid_tail_samples": 22.3,
      "learning_rate": 4.621149215175808e-06,
      "loss": 0.0187,
      "positive_teacher_pairs": 27197.5,
      "query_h_norm_ratio": 0.41503959596157075,
      "relation_loss": 0.0019996139802969993,
      "selected_pairs": 27197.5,
      "step": 26070,
      "student_distance_mean": 22.84398365020752,
      "teacher_distance_mean": 0.8595598101615906,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0019996139802969993,
      "weighted_relation_loss": 0.0019996139802969993
    },
    {
      "action_query_batch_std": 0.3485974818468094,
      "action_query_norm": 40.71489143371582,
      "action_relation_loss": 0.0019369442830793559,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.396632366697848,
      "flow_matching_loss": 0.015494906157255173,
      "grad_norm": 0.09915784001350403,
      "h_stock_valid_norm": 98.28551330566407,
      "invalid_tail_samples": 21.6,
      "learning_rate": 4.5980345145834154e-06,
      "loss": 0.0174,
      "positive_teacher_pairs": 27361.6,
      "query_h_norm_ratio": 0.41425434350967405,
      "relation_loss": 0.0019369442830793559,
      "selected_pairs": 27361.6,
      "step": 26080,
      "student_distance_mean": 22.88420925140381,
      "teacher_distance_mean": 0.8594091951847076,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.0019369442830793559,
      "weighted_relation_loss": 0.0019369442830793559
    },
    {
      "action_query_batch_std": 0.3434859424829483,
      "action_query_norm": 40.68959007263184,
      "action_relation_loss": 0.0019674943294376137,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.40598690364827,
      "flow_matching_loss": 0.016319410316646098,
      "grad_norm": 0.08859653770923615,
      "h_stock_valid_norm": 98.36853942871093,
      "invalid_tail_samples": 23.3,
      "learning_rate": 4.574974981660213e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 26965.4,
      "query_h_norm_ratio": 0.4136494368314743,
      "relation_loss": 0.0019674943294376137,
      "selected_pairs": 26965.4,
      "step": 26090,
      "student_distance_mean": 22.549509048461914,
      "teacher_distance_mean": 0.8457883358001709,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0019674943294376137,
      "weighted_relation_loss": 0.0019674943294376137
    },
    {
      "action_query_batch_std": 0.34629112482070923,
      "action_query_norm": 40.708417129516604,
      "action_relation_loss": 0.0019404031569138168,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.41534144059869,
      "flow_matching_loss": 0.015966577362269162,
      "grad_norm": 0.09852728247642517,
      "h_stock_valid_norm": 98.35862884521484,
      "invalid_tail_samples": 22.3,
      "learning_rate": 4.551970644425707e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 27200.8,
      "query_h_norm_ratio": 0.41388037204742434,
      "relation_loss": 0.0019404031569138168,
      "selected_pairs": 27200.8,
      "step": 26100,
      "student_distance_mean": 22.7287935256958,
      "teacher_distance_mean": 0.8549024999141693,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0019404031569138168,
      "weighted_relation_loss": 0.0019404031569138168
    },
    {
      "action_query_batch_std": 0.3480671882629395,
      "action_query_norm": 40.69691772460938,
      "action_relation_loss": 0.0019633354735560716,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.42469597754911,
      "flow_matching_loss": 0.015062033198773861,
      "grad_norm": 0.09462246298789978,
      "h_stock_valid_norm": 98.2724983215332,
      "invalid_tail_samples": 21.9,
      "learning_rate": 4.529021530832328e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 27288.5,
      "query_h_norm_ratio": 0.414127054810524,
      "relation_loss": 0.0019633354735560716,
      "selected_pairs": 27288.5,
      "step": 26110,
      "student_distance_mean": 22.709775161743163,
      "teacher_distance_mean": 0.8534591555595398,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.0019633354735560716,
      "weighted_relation_loss": 0.0019633354735560716
    },
    {
      "action_query_batch_std": 0.3500743299722672,
      "action_query_norm": 40.73619003295899,
      "action_relation_loss": 0.001991976529825479,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.434050514499532,
      "flow_matching_loss": 0.017683285381644966,
      "grad_norm": 0.12089208513498306,
      "h_stock_valid_norm": 98.26843795776367,
      "invalid_tail_samples": 25.5,
      "learning_rate": 4.5061276687654306e-06,
      "loss": 0.0197,
      "positive_teacher_pairs": 26460.5,
      "query_h_norm_ratio": 0.41454215049743653,
      "relation_loss": 0.001991976529825479,
      "selected_pairs": 26460.5,
      "step": 26120,
      "student_distance_mean": 23.028341865539552,
      "teacher_distance_mean": 0.864560866355896,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.001991976529825479,
      "weighted_relation_loss": 0.001991976529825479
    },
    {
      "action_query_batch_std": 0.3423299193382263,
      "action_query_norm": 40.68173408508301,
      "action_relation_loss": 0.0019663236220367254,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.443405051449954,
      "flow_matching_loss": 0.015497940964996815,
      "grad_norm": 0.09566304832696915,
      "h_stock_valid_norm": 98.11632537841797,
      "invalid_tail_samples": 26.0,
      "learning_rate": 4.4832890860431965e-06,
      "loss": 0.0175,
      "positive_teacher_pairs": 26351.5,
      "query_h_norm_ratio": 0.41462917923927306,
      "relation_loss": 0.0019663236220367254,
      "selected_pairs": 26351.5,
      "step": 26130,
      "student_distance_mean": 22.494115829467773,
      "teacher_distance_mean": 0.8457066357135773,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.0019663236220367254,
      "weighted_relation_loss": 0.0019663236220367254
    },
    {
      "action_query_batch_std": 0.3415446370840073,
      "action_query_norm": 40.65990943908692,
      "action_relation_loss": 0.001997803570702672,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.452759588400376,
      "flow_matching_loss": 0.015087474416941405,
      "grad_norm": 0.09106363356113434,
      "h_stock_valid_norm": 98.08243179321289,
      "invalid_tail_samples": 23.8,
      "learning_rate": 4.460505810416682e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 26860.1,
      "query_h_norm_ratio": 0.41454980373382566,
      "relation_loss": 0.001997803570702672,
      "selected_pairs": 26860.1,
      "step": 26140,
      "student_distance_mean": 22.385251426696776,
      "teacher_distance_mean": 0.8414897561073303,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.001997803570702672,
      "weighted_relation_loss": 0.001997803570702672
    },
    {
      "action_query_batch_std": 0.34996587932109835,
      "action_query_norm": 40.707685470581055,
      "action_relation_loss": 0.0019300899817608296,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.462114125350794,
      "flow_matching_loss": 0.016273791156709193,
      "grad_norm": 0.1457020491361618,
      "h_stock_valid_norm": 97.8409408569336,
      "invalid_tail_samples": 21.4,
      "learning_rate": 4.4377778695696984e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 27406.5,
      "query_h_norm_ratio": 0.41606088578701017,
      "relation_loss": 0.0019300899817608296,
      "selected_pairs": 27406.5,
      "step": 26150,
      "student_distance_mean": 22.86258602142334,
      "teacher_distance_mean": 0.861279284954071,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.0019300899817608296,
      "weighted_relation_loss": 0.0019300899817608296
    },
    {
      "action_query_batch_std": 0.34937328696250913,
      "action_query_norm": 40.724813079833986,
      "action_relation_loss": 0.0018605261691845953,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.471468662301216,
      "flow_matching_loss": 0.016085118148475887,
      "grad_norm": 0.09771998226642609,
      "h_stock_valid_norm": 98.03037033081054,
      "invalid_tail_samples": 21.6,
      "learning_rate": 4.415105291118843e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 27360.8,
      "query_h_norm_ratio": 0.41543270349502565,
      "relation_loss": 0.0018605261691845953,
      "selected_pairs": 27360.8,
      "step": 26160,
      "student_distance_mean": 22.933075523376466,
      "teacher_distance_mean": 0.8622393429279327,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.0018605261691845953,
      "weighted_relation_loss": 0.0018605261691845953
    },
    {
      "action_query_batch_std": 0.34608738124370575,
      "action_query_norm": 40.68628578186035,
      "action_relation_loss": 0.0019658836885355414,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.48082319925164,
      "flow_matching_loss": 0.014964199531823397,
      "grad_norm": 0.09657455235719681,
      "h_stock_valid_norm": 97.88439865112305,
      "invalid_tail_samples": 21.5,
      "learning_rate": 4.392488102613435e-06,
      "loss": 0.0169,
      "positive_teacher_pairs": 27388.4,
      "query_h_norm_ratio": 0.4156586408615112,
      "relation_loss": 0.0019658836885355414,
      "selected_pairs": 27388.4,
      "step": 26170,
      "student_distance_mean": 22.6428430557251,
      "teacher_distance_mean": 0.8507429182529449,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.0019658836885355414,
      "weighted_relation_loss": 0.0019658836885355414
    },
    {
      "action_query_batch_std": 0.3482853263616562,
      "action_query_norm": 40.709634399414064,
      "action_relation_loss": 0.0019313187687657774,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.490177736202057,
      "flow_matching_loss": 0.014214618597179651,
      "grad_norm": 0.13864564895629883,
      "h_stock_valid_norm": 97.99451446533203,
      "invalid_tail_samples": 24.6,
      "learning_rate": 4.369926331535473e-06,
      "loss": 0.0161,
      "positive_teacher_pairs": 26666.3,
      "query_h_norm_ratio": 0.4154288679361343,
      "relation_loss": 0.0019313187687657774,
      "selected_pairs": 26666.3,
      "step": 26180,
      "student_distance_mean": 22.91625232696533,
      "teacher_distance_mean": 0.8616195976734161,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0019313187687657774,
      "weighted_relation_loss": 0.0019313187687657774
    },
    {
      "action_query_batch_std": 0.345378777384758,
      "action_query_norm": 40.67867851257324,
      "action_relation_loss": 0.0020114992279559375,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.49953227315248,
      "flow_matching_loss": 0.016173185780644417,
      "grad_norm": 0.09422285109758377,
      "h_stock_valid_norm": 98.13294296264648,
      "invalid_tail_samples": 23.0,
      "learning_rate": 4.347420005299668e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 27031.0,
      "query_h_norm_ratio": 0.4145287811756134,
      "relation_loss": 0.0020114992279559375,
      "selected_pairs": 27031.0,
      "step": 26190,
      "student_distance_mean": 22.63258171081543,
      "teacher_distance_mean": 0.852333790063858,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0020114992279559375,
      "weighted_relation_loss": 0.0020114992279559375
    },
    {
      "action_query_batch_std": 0.3481723815202713,
      "action_query_norm": 40.70675849914551,
      "action_relation_loss": 0.0019924533553421497,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.5088868101029,
      "flow_matching_loss": 0.017584202345460653,
      "grad_norm": 0.11715111881494522,
      "h_stock_valid_norm": 97.92015380859375,
      "invalid_tail_samples": 22.7,
      "learning_rate": 4.324969151253317e-06,
      "loss": 0.0196,
      "positive_teacher_pairs": 27108.0,
      "query_h_norm_ratio": 0.41571774482727053,
      "relation_loss": 0.0019924533553421497,
      "selected_pairs": 27108.0,
      "step": 26200,
      "student_distance_mean": 22.854326438903808,
      "teacher_distance_mean": 0.8619442999362945,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0019924533553421497,
      "weighted_relation_loss": 0.0019924533553421497
    },
    {
      "action_query_batch_std": 0.34472562968730924,
      "action_query_norm": 40.65942726135254,
      "action_relation_loss": 0.0020151128759607674,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.51824134705332,
      "flow_matching_loss": 0.016987452749162912,
      "grad_norm": 0.0849326029419899,
      "h_stock_valid_norm": 97.85604476928711,
      "invalid_tail_samples": 25.1,
      "learning_rate": 4.302573796676312e-06,
      "loss": 0.019,
      "positive_teacher_pairs": 26549.1,
      "query_h_norm_ratio": 0.41550564765930176,
      "relation_loss": 0.0020151128759607674,
      "selected_pairs": 26549.1,
      "step": 26210,
      "student_distance_mean": 22.609195518493653,
      "teacher_distance_mean": 0.85024134516716,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0020151128759607674,
      "weighted_relation_loss": 0.0020151128759607674
    },
    {
      "action_query_batch_std": 0.34457465410232546,
      "action_query_norm": 40.69706153869629,
      "action_relation_loss": 0.0019619671860709787,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.52759588400374,
      "flow_matching_loss": 0.015205666236579419,
      "grad_norm": 0.07967142015695572,
      "h_stock_valid_norm": 97.76651229858399,
      "invalid_tail_samples": 23.1,
      "learning_rate": 4.2802339687811385e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 27013.8,
      "query_h_norm_ratio": 0.4162705779075623,
      "relation_loss": 0.0019619671860709787,
      "selected_pairs": 27013.8,
      "step": 26220,
      "student_distance_mean": 22.668764305114745,
      "teacher_distance_mean": 0.8533636152744293,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0019619671860709787,
      "weighted_relation_loss": 0.0019619671860709787
    },
    {
      "action_query_batch_std": 0.34568984508514405,
      "action_query_norm": 40.67297210693359,
      "action_relation_loss": 0.0020222312421537937,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.536950420954163,
      "flow_matching_loss": 0.015450872853398323,
      "grad_norm": 0.11240563541650772,
      "h_stock_valid_norm": 97.87520446777344,
      "invalid_tail_samples": 25.8,
      "learning_rate": 4.257949694712771e-06,
      "loss": 0.0175,
      "positive_teacher_pairs": 26383.0,
      "query_h_norm_ratio": 0.4155616223812103,
      "relation_loss": 0.0020222312421537937,
      "selected_pairs": 26383.0,
      "step": 26230,
      "student_distance_mean": 22.690288543701172,
      "teacher_distance_mean": 0.8545526087284088,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0020222312421537937,
      "weighted_relation_loss": 0.0020222312421537937
    },
    {
      "action_query_batch_std": 0.34719103276729585,
      "action_query_norm": 40.69769401550293,
      "action_relation_loss": 0.002063684759195894,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.546304957904585,
      "flow_matching_loss": 0.017104511614888906,
      "grad_norm": 0.1015988290309906,
      "h_stock_valid_norm": 97.82771072387695,
      "invalid_tail_samples": 24.6,
      "learning_rate": 4.235721001548726e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 26667.5,
      "query_h_norm_ratio": 0.4160179316997528,
      "relation_loss": 0.002063684759195894,
      "selected_pairs": 26667.5,
      "step": 26240,
      "student_distance_mean": 22.68313217163086,
      "teacher_distance_mean": 0.8539424955844879,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.002063684759195894,
      "weighted_relation_loss": 0.002063684759195894
    },
    {
      "action_query_batch_std": 0.3469379752874374,
      "action_query_norm": 40.71034126281738,
      "action_relation_loss": 0.0019201438175514341,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.555659494855004,
      "flow_matching_loss": 0.016739954892545938,
      "grad_norm": 0.09335900098085403,
      "h_stock_valid_norm": 97.81738510131837,
      "invalid_tail_samples": 21.5,
      "learning_rate": 4.21354791629896e-06,
      "loss": 0.0187,
      "positive_teacher_pairs": 27389.7,
      "query_h_norm_ratio": 0.4161910951137543,
      "relation_loss": 0.0019201438175514341,
      "selected_pairs": 27389.7,
      "step": 26250,
      "student_distance_mean": 22.637127113342284,
      "teacher_distance_mean": 0.8522831857204437,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.0019201438175514341,
      "weighted_relation_loss": 0.0019201438175514341
    },
    {
      "action_query_batch_std": 0.339715787768364,
      "action_query_norm": 40.61813812255859,
      "action_relation_loss": 0.001965216547250748,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.565014031805426,
      "flow_matching_loss": 0.015472012665122747,
      "grad_norm": 0.10567410290241241,
      "h_stock_valid_norm": 98.00642395019531,
      "invalid_tail_samples": 24.0,
      "learning_rate": 4.191430465905843e-06,
      "loss": 0.0174,
      "positive_teacher_pairs": 26804.6,
      "query_h_norm_ratio": 0.4144494831562042,
      "relation_loss": 0.001965216547250748,
      "selected_pairs": 26804.6,
      "step": 26260,
      "student_distance_mean": 22.25915660858154,
      "teacher_distance_mean": 0.837384831905365,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.001965216547250748,
      "weighted_relation_loss": 0.001965216547250748
    },
    {
      "action_query_batch_std": 0.34422069489955903,
      "action_query_norm": 40.673279190063475,
      "action_relation_loss": 0.002020677307154983,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.574368568755848,
      "flow_matching_loss": 0.017867149971425533,
      "grad_norm": 0.09491481631994247,
      "h_stock_valid_norm": 98.0659049987793,
      "invalid_tail_samples": 25.5,
      "learning_rate": 4.169368677244184e-06,
      "loss": 0.0199,
      "positive_teacher_pairs": 26458.5,
      "query_h_norm_ratio": 0.4147576779127121,
      "relation_loss": 0.002020677307154983,
      "selected_pairs": 26458.5,
      "step": 26270,
      "student_distance_mean": 22.460037612915038,
      "teacher_distance_mean": 0.8442013382911682,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.002020677307154983,
      "weighted_relation_loss": 0.002020677307154983
    },
    {
      "action_query_batch_std": 0.35011328756809235,
      "action_query_norm": 40.71681518554688,
      "action_relation_loss": 0.001910493418108672,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.583723105706266,
      "flow_matching_loss": 0.014996388927102089,
      "grad_norm": 0.09990518540143967,
      "h_stock_valid_norm": 97.9389419555664,
      "invalid_tail_samples": 24.1,
      "learning_rate": 4.147362577121122e-06,
      "loss": 0.0169,
      "positive_teacher_pairs": 26782.5,
      "query_h_norm_ratio": 0.4157368868589401,
      "relation_loss": 0.001910493418108672,
      "selected_pairs": 26782.5,
      "step": 26280,
      "student_distance_mean": 23.054411125183105,
      "teacher_distance_mean": 0.8668675124645233,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.001910493418108672,
      "weighted_relation_loss": 0.001910493418108672
    },
    {
      "action_query_batch_std": 0.3453492820262909,
      "action_query_norm": 40.67926940917969,
      "action_relation_loss": 0.0019482666626572608,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.593077642656688,
      "flow_matching_loss": 0.014898242894560099,
      "grad_norm": 0.16605909168720245,
      "h_stock_valid_norm": 97.92928466796874,
      "invalid_tail_samples": 24.4,
      "learning_rate": 4.125412192276157e-06,
      "loss": 0.0168,
      "positive_teacher_pairs": 26708.4,
      "query_h_norm_ratio": 0.41539977490901947,
      "relation_loss": 0.0019482666626572608,
      "selected_pairs": 26708.4,
      "step": 26290,
      "student_distance_mean": 22.681930541992188,
      "teacher_distance_mean": 0.8519282639026642,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0019482666626572608,
      "weighted_relation_loss": 0.0019482666626572608
    },
    {
      "action_query_batch_std": 0.3507969856262207,
      "action_query_norm": 40.71549186706543,
      "action_relation_loss": 0.001992459991015494,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.60243217960711,
      "flow_matching_loss": 0.01710844235494733,
      "grad_norm": 0.10348804295063019,
      "h_stock_valid_norm": 97.82526321411133,
      "invalid_tail_samples": 22.6,
      "learning_rate": 4.10351754938107e-06,
      "loss": 0.0191,
      "positive_teacher_pairs": 27131.7,
      "query_h_norm_ratio": 0.4162087976932526,
      "relation_loss": 0.001992459991015494,
      "selected_pairs": 27131.7,
      "step": 26300,
      "student_distance_mean": 22.919045639038085,
      "teacher_distance_mean": 0.8624108612537384,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.001992459991015494,
      "weighted_relation_loss": 0.001992459991015494
    },
    {
      "action_query_batch_std": 0.34301899969577787,
      "action_query_norm": 40.68040046691895,
      "action_relation_loss": 0.002052655804436654,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.611786716557532,
      "flow_matching_loss": 0.014957759249955415,
      "grad_norm": 0.08574594557285309,
      "h_stock_valid_norm": 98.17883605957032,
      "invalid_tail_samples": 22.1,
      "learning_rate": 4.081678675039913e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 27249.3,
      "query_h_norm_ratio": 0.4143510043621063,
      "relation_loss": 0.002052655804436654,
      "selected_pairs": 27249.3,
      "step": 26310,
      "student_distance_mean": 22.375490951538087,
      "teacher_distance_mean": 0.8432863116264343,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.002052655804436654,
      "weighted_relation_loss": 0.002052655804436654
    },
    {
      "action_query_batch_std": 0.3506490021944046,
      "action_query_norm": 40.73988800048828,
      "action_relation_loss": 0.0019435611786320806,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.62114125350795,
      "flow_matching_loss": 0.018667061068117618,
      "grad_norm": 0.09788044542074203,
      "h_stock_valid_norm": 98.11617126464844,
      "invalid_tail_samples": 21.4,
      "learning_rate": 4.0598955957889876e-06,
      "loss": 0.0206,
      "positive_teacher_pairs": 27407.5,
      "query_h_norm_ratio": 0.4152232587337494,
      "relation_loss": 0.0019435611786320806,
      "selected_pairs": 27407.5,
      "step": 26320,
      "student_distance_mean": 22.940979957580566,
      "teacher_distance_mean": 0.8656389772891998,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.0019435611786320806,
      "weighted_relation_loss": 0.0019435611786320806
    },
    {
      "action_query_batch_std": 0.34098553359508516,
      "action_query_norm": 40.656067276000975,
      "action_relation_loss": 0.0019699632073752584,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.630495790458372,
      "flow_matching_loss": 0.01626026425510645,
      "grad_norm": 0.10083042830228806,
      "h_stock_valid_norm": 97.99658966064453,
      "invalid_tail_samples": 21.6,
      "learning_rate": 4.038168338096776e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 27360.3,
      "query_h_norm_ratio": 0.4148727774620056,
      "relation_loss": 0.0019699632073752584,
      "selected_pairs": 27360.3,
      "step": 26330,
      "student_distance_mean": 22.43275375366211,
      "teacher_distance_mean": 0.8444082736968994,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.0019699632073752584,
      "weighted_relation_loss": 0.0019699632073752584
    },
    {
      "action_query_batch_std": 0.3465292125940323,
      "action_query_norm": 40.67936325073242,
      "action_relation_loss": 0.002034902642481029,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.639850327408794,
      "flow_matching_loss": 0.015837622713297604,
      "grad_norm": 0.0916980504989624,
      "h_stock_valid_norm": 98.04437408447265,
      "invalid_tail_samples": 25.5,
      "learning_rate": 4.016496928363944e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 26462.4,
      "query_h_norm_ratio": 0.41490971446037295,
      "relation_loss": 0.002034902642481029,
      "selected_pairs": 26462.4,
      "step": 26340,
      "student_distance_mean": 22.73456745147705,
      "teacher_distance_mean": 0.8556677103042603,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.002034902642481029,
      "weighted_relation_loss": 0.002034902642481029
    },
    {
      "action_query_batch_std": 0.34750497341156006,
      "action_query_norm": 40.7140926361084,
      "action_relation_loss": 0.0019588835653848945,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.649204864359213,
      "flow_matching_loss": 0.01546412417665124,
      "grad_norm": 0.1713038980960846,
      "h_stock_valid_norm": 98.10043258666992,
      "invalid_tail_samples": 22.1,
      "learning_rate": 3.994881392923317e-06,
      "loss": 0.0174,
      "positive_teacher_pairs": 27248.1,
      "query_h_norm_ratio": 0.41502591371536257,
      "relation_loss": 0.0019588835653848945,
      "selected_pairs": 27248.1,
      "step": 26350,
      "student_distance_mean": 22.727522468566896,
      "teacher_distance_mean": 0.8560960471630097,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0019588835653848945,
      "weighted_relation_loss": 0.0019588835653848945
    },
    {
      "action_query_batch_std": 0.34525555968284605,
      "action_query_norm": 40.695771026611325,
      "action_relation_loss": 0.001964894263073802,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.658559401309635,
      "flow_matching_loss": 0.017589033488184212,
      "grad_norm": 0.0938817709684372,
      "h_stock_valid_norm": 97.92042465209961,
      "invalid_tail_samples": 25.3,
      "learning_rate": 3.9733217580397945e-06,
      "loss": 0.0196,
      "positive_teacher_pairs": 26503.0,
      "query_h_norm_ratio": 0.41560410559177396,
      "relation_loss": 0.001964894263073802,
      "selected_pairs": 26503.0,
      "step": 26360,
      "student_distance_mean": 22.798545837402344,
      "teacher_distance_mean": 0.8573636949062348,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.001964894263073802,
      "weighted_relation_loss": 0.001964894263073802
    },
    {
      "action_query_batch_std": 0.3429465562105179,
      "action_query_norm": 40.67837371826172,
      "action_relation_loss": 0.002047843614127487,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.667913938260057,
      "flow_matching_loss": 0.017184707336127758,
      "grad_norm": 0.13436353206634521,
      "h_stock_valid_norm": 97.84415969848632,
      "invalid_tail_samples": 25.0,
      "learning_rate": 3.951818049910366e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 26573.6,
      "query_h_norm_ratio": 0.4157486200332642,
      "relation_loss": 0.002047843614127487,
      "selected_pairs": 26573.6,
      "step": 26370,
      "student_distance_mean": 22.507000350952147,
      "teacher_distance_mean": 0.8468840897083283,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.002047843614127487,
      "weighted_relation_loss": 0.002047843614127487
    },
    {
      "action_query_batch_std": 0.34104676842689513,
      "action_query_norm": 40.6589771270752,
      "action_relation_loss": 0.0019099976285360753,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.67726847521048,
      "flow_matching_loss": 0.01543425964191556,
      "grad_norm": 0.1044015884399414,
      "h_stock_valid_norm": 97.8396240234375,
      "invalid_tail_samples": 24.9,
      "learning_rate": 3.930370294664071e-06,
      "loss": 0.0173,
      "positive_teacher_pairs": 26597.9,
      "query_h_norm_ratio": 0.4155688464641571,
      "relation_loss": 0.0019099976285360753,
      "selected_pairs": 26597.9,
      "step": 26380,
      "student_distance_mean": 22.538696098327637,
      "teacher_distance_mean": 0.8466160535812378,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0019099976285360753,
      "weighted_relation_loss": 0.0019099976285360753
    },
    {
      "action_query_batch_std": 0.3454879134893417,
      "action_query_norm": 40.690771865844724,
      "action_relation_loss": 0.0019222526345402003,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.686623012160897,
      "flow_matching_loss": 0.01482789684087038,
      "grad_norm": 0.09992438554763794,
      "h_stock_valid_norm": 97.86219940185546,
      "invalid_tail_samples": 21.9,
      "learning_rate": 3.908978518361938e-06,
      "loss": 0.0168,
      "positive_teacher_pairs": 27296.2,
      "query_h_norm_ratio": 0.41579862833023074,
      "relation_loss": 0.0019222526345402003,
      "selected_pairs": 27296.2,
      "step": 26390,
      "student_distance_mean": 22.65956974029541,
      "teacher_distance_mean": 0.8531278252601624,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.0019222526345402003,
      "weighted_relation_loss": 0.0019222526345402003
    },
    {
      "action_query_batch_std": 0.3477976113557816,
      "action_query_norm": 40.696902465820315,
      "action_relation_loss": 0.0019225807976908983,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.69597754911132,
      "flow_matching_loss": 0.016792679112404583,
      "grad_norm": 0.10789452493190765,
      "h_stock_valid_norm": 97.82762832641602,
      "invalid_tail_samples": 23.6,
      "learning_rate": 3.887642746997017e-06,
      "loss": 0.0187,
      "positive_teacher_pairs": 26890.6,
      "query_h_norm_ratio": 0.4160072773694992,
      "relation_loss": 0.0019225807976908983,
      "selected_pairs": 26890.6,
      "step": 26400,
      "student_distance_mean": 22.71832618713379,
      "teacher_distance_mean": 0.8567950189113617,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0019225807976908983,
      "weighted_relation_loss": 0.0019225807976908983
    },
    {
      "action_query_batch_std": 0.34779230654239657,
      "action_query_norm": 40.66339111328125,
      "action_relation_loss": 0.0020720209460705517,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.70533208606174,
      "flow_matching_loss": 0.01657564379274845,
      "grad_norm": 0.10061172395944595,
      "h_stock_valid_norm": 97.95223770141601,
      "invalid_tail_samples": 25.1,
      "learning_rate": 3.866363006494255e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 26556.2,
      "query_h_norm_ratio": 0.415138041973114,
      "relation_loss": 0.0020720209460705517,
      "selected_pairs": 26556.2,
      "step": 26410,
      "student_distance_mean": 22.749052619934083,
      "teacher_distance_mean": 0.855964207649231,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0020720209460705517,
      "weighted_relation_loss": 0.0020720209460705517
    },
    {
      "action_query_batch_std": 0.34720302522182467,
      "action_query_norm": 40.673278427124025,
      "action_relation_loss": 0.00200276697287336,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.71468662301216,
      "flow_matching_loss": 0.016823900770395993,
      "grad_norm": 0.09521540254354477,
      "h_stock_valid_norm": 97.7970199584961,
      "invalid_tail_samples": 22.1,
      "learning_rate": 3.8451393227105725e-06,
      "loss": 0.0188,
      "positive_teacher_pairs": 27245.9,
      "query_h_norm_ratio": 0.41590004563331606,
      "relation_loss": 0.00200276697287336,
      "selected_pairs": 27245.9,
      "step": 26420,
      "student_distance_mean": 22.863432121276855,
      "teacher_distance_mean": 0.8599147200584412,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.00200276697287336,
      "weighted_relation_loss": 0.00200276697287336
    },
    {
      "action_query_batch_std": 0.345606729388237,
      "action_query_norm": 40.68305549621582,
      "action_relation_loss": 0.0019542199559509756,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.72404115996258,
      "flow_matching_loss": 0.01543239140883088,
      "grad_norm": 0.08938373625278473,
      "h_stock_valid_norm": 97.89850769042968,
      "invalid_tail_samples": 23.8,
      "learning_rate": 3.823971721434727e-06,
      "loss": 0.0174,
      "positive_teacher_pairs": 26847.8,
      "query_h_norm_ratio": 0.4155645489692688,
      "relation_loss": 0.0019542199559509756,
      "selected_pairs": 26847.8,
      "step": 26430,
      "student_distance_mean": 22.73139133453369,
      "teacher_distance_mean": 0.8540280401706696,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0019542199559509756,
      "weighted_relation_loss": 0.0019542199559509756
    },
    {
      "action_query_batch_std": 0.3450278162956238,
      "action_query_norm": 40.69929466247559,
      "action_relation_loss": 0.0018907591002061963,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.733395696913004,
      "flow_matching_loss": 0.01531425267457962,
      "grad_norm": 0.08693420141935349,
      "h_stock_valid_norm": 97.87737655639648,
      "invalid_tail_samples": 23.3,
      "learning_rate": 3.8028602283873505e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 26970.7,
      "query_h_norm_ratio": 0.4158221423625946,
      "relation_loss": 0.0018907591002061963,
      "selected_pairs": 26970.7,
      "step": 26440,
      "student_distance_mean": 22.64361228942871,
      "teacher_distance_mean": 0.8519418597221374,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0018907591002061963,
      "weighted_relation_loss": 0.0018907591002061963
    },
    {
      "action_query_batch_std": 0.34783112406730654,
      "action_query_norm": 40.68293762207031,
      "action_relation_loss": 0.002026167616713792,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.742750233863426,
      "flow_matching_loss": 0.01706618517637253,
      "grad_norm": 0.10355259478092194,
      "h_stock_valid_norm": 97.92661972045899,
      "invalid_tail_samples": 25.3,
      "learning_rate": 3.781804869220912e-06,
      "loss": 0.0191,
      "positive_teacher_pairs": 26505.4,
      "query_h_norm_ratio": 0.4154451549053192,
      "relation_loss": 0.002026167616713792,
      "selected_pairs": 26505.4,
      "step": 26450,
      "student_distance_mean": 22.89730033874512,
      "teacher_distance_mean": 0.8622684121131897,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.002026167616713792,
      "weighted_relation_loss": 0.002026167616713792
    },
    {
      "action_query_batch_std": 0.34454052448272704,
      "action_query_norm": 40.70092849731445,
      "action_relation_loss": 0.0020853063324466348,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.752104770813844,
      "flow_matching_loss": 0.015840970259159804,
      "grad_norm": 0.3126986622810364,
      "h_stock_valid_norm": 98.1063232421875,
      "invalid_tail_samples": 23.3,
      "learning_rate": 3.760805669519646e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 26966.3,
      "query_h_norm_ratio": 0.41486703753471377,
      "relation_loss": 0.0020853063324466348,
      "selected_pairs": 26966.3,
      "step": 26460,
      "student_distance_mean": 22.655785751342773,
      "teacher_distance_mean": 0.8506422460079193,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0020853063324466348,
      "weighted_relation_loss": 0.0020853063324466348
    },
    {
      "action_query_batch_std": 0.34487585425376893,
      "action_query_norm": 40.63492546081543,
      "action_relation_loss": 0.002004219149239361,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.761459307764266,
      "flow_matching_loss": 0.016802537627518176,
      "grad_norm": 0.10173865407705307,
      "h_stock_valid_norm": 98.06369094848633,
      "invalid_tail_samples": 26.6,
      "learning_rate": 3.7398626547995587e-06,
      "loss": 0.0188,
      "positive_teacher_pairs": 26209.0,
      "query_h_norm_ratio": 0.414374440908432,
      "relation_loss": 0.002004219149239361,
      "selected_pairs": 26209.0,
      "step": 26470,
      "student_distance_mean": 22.581168174743652,
      "teacher_distance_mean": 0.8504232704639435,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.002004219149239361,
      "weighted_relation_loss": 0.002004219149239361
    },
    {
      "action_query_batch_std": 0.34071231484413145,
      "action_query_norm": 40.66885147094727,
      "action_relation_loss": 0.001982584164943546,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.770813844714688,
      "flow_matching_loss": 0.015145753882825375,
      "grad_norm": 0.08572811633348465,
      "h_stock_valid_norm": 97.84540252685547,
      "invalid_tail_samples": 22.3,
      "learning_rate": 3.7189758505083995e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 27201.9,
      "query_h_norm_ratio": 0.41564701199531556,
      "relation_loss": 0.001982584164943546,
      "selected_pairs": 27201.9,
      "step": 26480,
      "student_distance_mean": 22.385942268371583,
      "teacher_distance_mean": 0.843765264749527,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.001982584164943546,
      "weighted_relation_loss": 0.001982584164943546
    },
    {
      "action_query_batch_std": 0.34279951751232146,
      "action_query_norm": 40.66095809936523,
      "action_relation_loss": 0.0019604608649387957,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.780168381665106,
      "flow_matching_loss": 0.01715685296803713,
      "grad_norm": 0.10764091461896896,
      "h_stock_valid_norm": 98.07054061889649,
      "invalid_tail_samples": 24.3,
      "learning_rate": 3.6981452820255837e-06,
      "loss": 0.0191,
      "positive_teacher_pairs": 26734.9,
      "query_h_norm_ratio": 0.4146113395690918,
      "relation_loss": 0.0019604608649387957,
      "selected_pairs": 26734.9,
      "step": 26490,
      "student_distance_mean": 22.439784431457518,
      "teacher_distance_mean": 0.8450819492340088,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0019604608649387957,
      "weighted_relation_loss": 0.0019604608649387957
    },
    {
      "action_query_batch_std": 0.3426966041326523,
      "action_query_norm": 40.65768203735352,
      "action_relation_loss": 0.001949961401987821,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.78952291861553,
      "flow_matching_loss": 0.016437170468270777,
      "grad_norm": 0.23449085652828217,
      "h_stock_valid_norm": 98.07962036132812,
      "invalid_tail_samples": 24.8,
      "learning_rate": 3.6773709746622363e-06,
      "loss": 0.0184,
      "positive_teacher_pairs": 26624.0,
      "query_h_norm_ratio": 0.41453985273838045,
      "relation_loss": 0.001949961401987821,
      "selected_pairs": 26624.0,
      "step": 26500,
      "student_distance_mean": 22.644374275207518,
      "teacher_distance_mean": 0.8528895795345306,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.001949961401987821,
      "weighted_relation_loss": 0.001949961401987821
    },
    {
      "action_query_batch_std": 0.3480449229478836,
      "action_query_norm": 40.705601119995116,
      "action_relation_loss": 0.0019356238306500017,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.79887745556595,
      "flow_matching_loss": 0.015227765310555696,
      "grad_norm": 0.10488836467266083,
      "h_stock_valid_norm": 98.07556762695313,
      "invalid_tail_samples": 25.3,
      "learning_rate": 3.6566529536610995e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 26500.9,
      "query_h_norm_ratio": 0.41504487693309783,
      "relation_loss": 0.0019356238306500017,
      "selected_pairs": 26500.9,
      "step": 26510,
      "student_distance_mean": 22.880722427368163,
      "teacher_distance_mean": 0.8626170575618743,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0019356238306500017,
      "weighted_relation_loss": 0.0019356238306500017
    },
    {
      "action_query_batch_std": 0.34835530519485475,
      "action_query_norm": 40.6958065032959,
      "action_relation_loss": 0.0018942257738672197,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.80823199251637,
      "flow_matching_loss": 0.01438616756349802,
      "grad_norm": 0.108968585729599,
      "h_stock_valid_norm": 98.00640563964843,
      "invalid_tail_samples": 23.2,
      "learning_rate": 3.635991244196513e-06,
      "loss": 0.0163,
      "positive_teacher_pairs": 26987.8,
      "query_h_norm_ratio": 0.41523689925670626,
      "relation_loss": 0.0018942257738672197,
      "selected_pairs": 26987.8,
      "step": 26520,
      "student_distance_mean": 22.9161958694458,
      "teacher_distance_mean": 0.8635007858276367,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0018942257738672197,
      "weighted_relation_loss": 0.0018942257738672197
    },
    {
      "action_query_batch_std": 0.3467751830816269,
      "action_query_norm": 40.692800521850586,
      "action_relation_loss": 0.0019482884905301034,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.81758652946679,
      "flow_matching_loss": 0.018731429893523454,
      "grad_norm": 0.10325559228658676,
      "h_stock_valid_norm": 97.95403289794922,
      "invalid_tail_samples": 23.9,
      "learning_rate": 3.615385871374405e-06,
      "loss": 0.0207,
      "positive_teacher_pairs": 26823.8,
      "query_h_norm_ratio": 0.4154288738965988,
      "relation_loss": 0.0019482884905301034,
      "selected_pairs": 26823.8,
      "step": 26530,
      "student_distance_mean": 22.677438926696777,
      "teacher_distance_mean": 0.8536558747291565,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0019482884905301034,
      "weighted_relation_loss": 0.0019482884905301034
    },
    {
      "action_query_batch_std": 0.3475425481796265,
      "action_query_norm": 40.696706008911136,
      "action_relation_loss": 0.0019401596044190228,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.826941066417213,
      "flow_matching_loss": 0.017476718313992022,
      "grad_norm": 0.129449263215065,
      "h_stock_valid_norm": 97.99859466552735,
      "invalid_tail_samples": 22.8,
      "learning_rate": 3.5948368602322367e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 27078.5,
      "query_h_norm_ratio": 0.4152813881635666,
      "relation_loss": 0.0019401596044190228,
      "selected_pairs": 27078.5,
      "step": 26540,
      "student_distance_mean": 22.85547618865967,
      "teacher_distance_mean": 0.8604686617851257,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0019401596044190228,
      "weighted_relation_loss": 0.0019401596044190228
    },
    {
      "action_query_batch_std": 0.35056803822517396,
      "action_query_norm": 40.70787582397461,
      "action_relation_loss": 0.0019786858349107204,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.836295603367635,
      "flow_matching_loss": 0.01672425717115402,
      "grad_norm": 0.14247770607471466,
      "h_stock_valid_norm": 97.97924118041992,
      "invalid_tail_samples": 24.3,
      "learning_rate": 3.5743442357390187e-06,
      "loss": 0.0187,
      "positive_teacher_pairs": 26732.4,
      "query_h_norm_ratio": 0.4154755979776382,
      "relation_loss": 0.0019786858349107204,
      "selected_pairs": 26732.4,
      "step": 26550,
      "student_distance_mean": 22.988055419921874,
      "teacher_distance_mean": 0.8665674388408661,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0019786858349107204,
      "weighted_relation_loss": 0.0019786858349107204
    },
    {
      "action_query_batch_std": 0.344851553440094,
      "action_query_norm": 40.6720272064209,
      "action_relation_loss": 0.0019694863352924586,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.845650140318053,
      "flow_matching_loss": 0.01636503906920552,
      "grad_norm": 0.12025183439254761,
      "h_stock_valid_norm": 98.08115768432617,
      "invalid_tail_samples": 21.5,
      "learning_rate": 3.5539080227951937e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 27386.1,
      "query_h_norm_ratio": 0.41467897295951844,
      "relation_loss": 0.0019694863352924586,
      "selected_pairs": 27386.1,
      "step": 26560,
      "student_distance_mean": 22.506038093566893,
      "teacher_distance_mean": 0.8472756683826447,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.0019694863352924586,
      "weighted_relation_loss": 0.0019694863352924586
    },
    {
      "action_query_batch_std": 0.34272881150245665,
      "action_query_norm": 40.64541358947754,
      "action_relation_loss": 0.0019236169289797544,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.855004677268475,
      "flow_matching_loss": 0.015474700275808573,
      "grad_norm": 0.08868031203746796,
      "h_stock_valid_norm": 98.02170486450196,
      "invalid_tail_samples": 24.8,
      "learning_rate": 3.533528246232709e-06,
      "loss": 0.0174,
      "positive_teacher_pairs": 26621.6,
      "query_h_norm_ratio": 0.41466293334960935,
      "relation_loss": 0.0019236169289797544,
      "selected_pairs": 26621.6,
      "step": 26570,
      "student_distance_mean": 22.48790874481201,
      "teacher_distance_mean": 0.8457021832466125,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0019236169289797544,
      "weighted_relation_loss": 0.0019236169289797544
    },
    {
      "action_query_batch_std": 0.3439452737569809,
      "action_query_norm": 40.66818542480469,
      "action_relation_loss": 0.001996341336052865,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.864359214218897,
      "flow_matching_loss": 0.013859689328819513,
      "grad_norm": 0.08605322241783142,
      "h_stock_valid_norm": 98.11221389770508,
      "invalid_tail_samples": 22.5,
      "learning_rate": 3.5132049308149117e-06,
      "loss": 0.0159,
      "positive_teacher_pairs": 27155.4,
      "query_h_norm_ratio": 0.4145078927278519,
      "relation_loss": 0.001996341336052865,
      "selected_pairs": 27155.4,
      "step": 26580,
      "student_distance_mean": 22.604469299316406,
      "teacher_distance_mean": 0.8507469356060028,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.001996341336052865,
      "weighted_relation_loss": 0.001996341336052865
    },
    {
      "action_query_batch_std": 0.35045350790023805,
      "action_query_norm": 40.68764801025391,
      "action_relation_loss": 0.0019517453736625611,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.873713751169316,
      "flow_matching_loss": 0.015928878169506788,
      "grad_norm": 0.1130688339471817,
      "h_stock_valid_norm": 98.12228851318359,
      "invalid_tail_samples": 24.2,
      "learning_rate": 3.4929381012365425e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 26755.4,
      "query_h_norm_ratio": 0.4146651059389114,
      "relation_loss": 0.0019517453736625611,
      "selected_pairs": 26755.4,
      "step": 26590,
      "student_distance_mean": 22.993132019042967,
      "teacher_distance_mean": 0.8645830392837525,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0019517453736625611,
      "weighted_relation_loss": 0.0019517453736625611
    },
    {
      "action_query_batch_std": 0.3460786074399948,
      "action_query_norm": 40.69507331848145,
      "action_relation_loss": 0.0019139752141200006,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.883068288119738,
      "flow_matching_loss": 0.01564086154103279,
      "grad_norm": 0.09793597459793091,
      "h_stock_valid_norm": 98.01414794921875,
      "invalid_tail_samples": 21.1,
      "learning_rate": 3.472727782123697e-06,
      "loss": 0.0176,
      "positive_teacher_pairs": 27486.0,
      "query_h_norm_ratio": 0.41520085036754606,
      "relation_loss": 0.0019139752141200006,
      "selected_pairs": 27486.0,
      "step": 26600,
      "student_distance_mean": 22.630066108703613,
      "teacher_distance_mean": 0.8521816432476044,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.0019139752141200006,
      "weighted_relation_loss": 0.0019139752141200006
    },
    {
      "action_query_batch_std": 0.34282579123973844,
      "action_query_norm": 40.67916488647461,
      "action_relation_loss": 0.002037266024854034,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.89242282507016,
      "flow_matching_loss": 0.015134448558092118,
      "grad_norm": 0.10388446599245071,
      "h_stock_valid_norm": 97.8757209777832,
      "invalid_tail_samples": 23.2,
      "learning_rate": 3.4525739980338425e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 26988.9,
      "query_h_norm_ratio": 0.4156288206577301,
      "relation_loss": 0.002037266024854034,
      "selected_pairs": 26988.9,
      "step": 26610,
      "student_distance_mean": 22.531906127929688,
      "teacher_distance_mean": 0.8475111067295075,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.002037266024854034,
      "weighted_relation_loss": 0.002037266024854034
    },
    {
      "action_query_batch_std": 0.3438380122184753,
      "action_query_norm": 40.6661190032959,
      "action_relation_loss": 0.0019445643294602633,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.90177736202058,
      "flow_matching_loss": 0.016212190967053175,
      "grad_norm": 0.17809619009494781,
      "h_stock_valid_norm": 97.93863143920899,
      "invalid_tail_samples": 22.9,
      "learning_rate": 3.432476773455712e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 27061.7,
      "query_h_norm_ratio": 0.41522522568702697,
      "relation_loss": 0.0019445643294602633,
      "selected_pairs": 27061.7,
      "step": 26620,
      "student_distance_mean": 22.585935211181642,
      "teacher_distance_mean": 0.8501305937767029,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0019445643294602633,
      "weighted_relation_loss": 0.0019445643294602633
    },
    {
      "action_query_batch_std": 0.34669519364833834,
      "action_query_norm": 40.670335388183595,
      "action_relation_loss": 0.001953963888809085,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.911131898971,
      "flow_matching_loss": 0.015322363749146462,
      "grad_norm": 0.10126281529664993,
      "h_stock_valid_norm": 97.86461715698242,
      "invalid_tail_samples": 23.5,
      "learning_rate": 3.4124361328093378e-06,
      "loss": 0.0173,
      "positive_teacher_pairs": 26921.1,
      "query_h_norm_ratio": 0.41558080315589907,
      "relation_loss": 0.001953963888809085,
      "selected_pairs": 26921.1,
      "step": 26630,
      "student_distance_mean": 22.82313289642334,
      "teacher_distance_mean": 0.8595704674720764,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.001953963888809085,
      "weighted_relation_loss": 0.001953963888809085
    },
    {
      "action_query_batch_std": 0.3451068729162216,
      "action_query_norm": 40.66450157165527,
      "action_relation_loss": 0.0019473739201202988,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.920486435921422,
      "flow_matching_loss": 0.017468793597072364,
      "grad_norm": 0.0988059788942337,
      "h_stock_valid_norm": 97.85213623046874,
      "invalid_tail_samples": 26.8,
      "learning_rate": 3.392452100445975e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 26167.9,
      "query_h_norm_ratio": 0.4155716776847839,
      "relation_loss": 0.0019473739201202988,
      "selected_pairs": 26167.9,
      "step": 26640,
      "student_distance_mean": 22.754016876220703,
      "teacher_distance_mean": 0.8579560041427612,
      "valid_action_samples": 229.2,
      "weighted_auxiliary_loss": 0.0019473739201202988,
      "weighted_relation_loss": 0.0019473739201202988
    },
    {
      "action_query_batch_std": 0.3467362970113754,
      "action_query_norm": 40.69515151977539,
      "action_relation_loss": 0.0019058934063650667,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.929840972871844,
      "flow_matching_loss": 0.017802657466381787,
      "grad_norm": 0.10223283618688583,
      "h_stock_valid_norm": 97.9675422668457,
      "invalid_tail_samples": 24.0,
      "learning_rate": 3.3725247006481163e-06,
      "loss": 0.0197,
      "positive_teacher_pairs": 26805.8,
      "query_h_norm_ratio": 0.41539534628391267,
      "relation_loss": 0.0019058934063650667,
      "selected_pairs": 26805.8,
      "step": 26650,
      "student_distance_mean": 22.76315689086914,
      "teacher_distance_mean": 0.8569309413433075,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0019058934063650667,
      "weighted_relation_loss": 0.0019058934063650667
    },
    {
      "action_query_batch_std": 0.34706045091152193,
      "action_query_norm": 40.6755989074707,
      "action_relation_loss": 0.0019743863958865404,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.939195509822262,
      "flow_matching_loss": 0.016446453332901,
      "grad_norm": 0.0811353549361229,
      "h_stock_valid_norm": 97.85492858886718,
      "invalid_tail_samples": 23.9,
      "learning_rate": 3.3526539576294313e-06,
      "loss": 0.0184,
      "positive_teacher_pairs": 26824.6,
      "query_h_norm_ratio": 0.4156783729791641,
      "relation_loss": 0.0019743863958865404,
      "selected_pairs": 26824.6,
      "step": 26660,
      "student_distance_mean": 22.840336036682128,
      "teacher_distance_mean": 0.8621124148368835,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0019743863958865404,
      "weighted_relation_loss": 0.0019743863958865404
    },
    {
      "action_query_batch_std": 0.3445183277130127,
      "action_query_norm": 40.67184257507324,
      "action_relation_loss": 0.0019704361446201803,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.948550046772684,
      "flow_matching_loss": 0.014335073716938495,
      "grad_norm": 0.08687210083007812,
      "h_stock_valid_norm": 97.90335159301758,
      "invalid_tail_samples": 25.1,
      "learning_rate": 3.332839895534745e-06,
      "loss": 0.0163,
      "positive_teacher_pairs": 26559.0,
      "query_h_norm_ratio": 0.4154312163591385,
      "relation_loss": 0.0019704361446201803,
      "selected_pairs": 26559.0,
      "step": 26670,
      "student_distance_mean": 22.69031677246094,
      "teacher_distance_mean": 0.8539897322654724,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0019704361446201803,
      "weighted_relation_loss": 0.0019704361446201803
    },
    {
      "action_query_batch_std": 0.3467892646789551,
      "action_query_norm": 40.68534049987793,
      "action_relation_loss": 0.0019130464643239975,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.957904583723106,
      "flow_matching_loss": 0.014813712611794472,
      "grad_norm": 0.08647322654724121,
      "h_stock_valid_norm": 97.72708435058594,
      "invalid_tail_samples": 22.1,
      "learning_rate": 3.3130825384400155e-06,
      "loss": 0.0167,
      "positive_teacher_pairs": 27244.6,
      "query_h_norm_ratio": 0.4163191705942154,
      "relation_loss": 0.0019130464643239975,
      "selected_pairs": 27244.6,
      "step": 26680,
      "student_distance_mean": 22.792236328125,
      "teacher_distance_mean": 0.8576843738555908,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0019130464643239975,
      "weighted_relation_loss": 0.0019130464643239975
    },
    {
      "action_query_batch_std": 0.3457540422677994,
      "action_query_norm": 40.697197341918944,
      "action_relation_loss": 0.001960672379937023,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.96725912067353,
      "flow_matching_loss": 0.016816566791385414,
      "grad_norm": 0.10166017711162567,
      "h_stock_valid_norm": 97.7004150390625,
      "invalid_tail_samples": 23.0,
      "learning_rate": 3.293381910352278e-06,
      "loss": 0.0188,
      "positive_teacher_pairs": 27035.9,
      "query_h_norm_ratio": 0.41655449867248534,
      "relation_loss": 0.001960672379937023,
      "selected_pairs": 27035.9,
      "step": 26690,
      "student_distance_mean": 22.74029121398926,
      "teacher_distance_mean": 0.8549809157848358,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.001960672379937023,
      "weighted_relation_loss": 0.001960672379937023
    },
    {
      "action_query_batch_std": 0.346600678563118,
      "action_query_norm": 40.69897727966308,
      "action_relation_loss": 0.0018809766159392893,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.976613657623947,
      "flow_matching_loss": 0.016807017754763363,
      "grad_norm": 0.09584419429302216,
      "h_stock_valid_norm": 97.74853057861328,
      "invalid_tail_samples": 24.4,
      "learning_rate": 3.273738035209678e-06,
      "loss": 0.0187,
      "positive_teacher_pairs": 26709.1,
      "query_h_norm_ratio": 0.41636863350868225,
      "relation_loss": 0.0018809766159392893,
      "selected_pairs": 26709.1,
      "step": 26700,
      "student_distance_mean": 22.76158332824707,
      "teacher_distance_mean": 0.8558022916316986,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0018809766159392893,
      "weighted_relation_loss": 0.0018809766159392893
    },
    {
      "action_query_batch_std": 0.3445211827754974,
      "action_query_norm": 40.69261703491211,
      "action_relation_loss": 0.001878824911545962,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.98596819457437,
      "flow_matching_loss": 0.01645999038591981,
      "grad_norm": 0.11640273779630661,
      "h_stock_valid_norm": 97.77771606445313,
      "invalid_tail_samples": 25.4,
      "learning_rate": 3.254150936881356e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 26479.0,
      "query_h_norm_ratio": 0.4161758482456207,
      "relation_loss": 0.001878824911545962,
      "selected_pairs": 26479.0,
      "step": 26710,
      "student_distance_mean": 22.64390754699707,
      "teacher_distance_mean": 0.8531536757946014,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.001878824911545962,
      "weighted_relation_loss": 0.001878824911545962
    },
    {
      "action_query_batch_std": 0.3455888867378235,
      "action_query_norm": 40.663753509521484,
      "action_relation_loss": 0.0019593306700699032,
      "auxiliary_loss_weight": 1.0,
      "epoch": 24.99532273152479,
      "flow_matching_loss": 0.01516038728877902,
      "grad_norm": 0.15904125571250916,
      "h_stock_valid_norm": 97.85699005126953,
      "invalid_tail_samples": 26.3,
      "learning_rate": 3.234620639167496e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 26271.5,
      "query_h_norm_ratio": 0.41554788053035735,
      "relation_loss": 0.0019593306700699032,
      "selected_pairs": 26271.5,
      "step": 26720,
      "student_distance_mean": 22.765337181091308,
      "teacher_distance_mean": 0.8584648728370666,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.0019593306700699032,
      "weighted_relation_loss": 0.0019593306700699032
    },
    {
      "action_query_batch_std": 0.34755218029022217,
      "action_query_norm": 40.67998695373535,
      "action_relation_loss": 0.001956566807348281,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.00467726847521,
      "flow_matching_loss": 0.015079733077436685,
      "grad_norm": 0.08210590481758118,
      "h_stock_valid_norm": 98.00325164794921,
      "invalid_tail_samples": 21.2,
      "learning_rate": 3.2151471657992482e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 24424.8,
      "query_h_norm_ratio": 0.41509059965610506,
      "relation_loss": 0.001956566807348281,
      "selected_pairs": 24424.8,
      "step": 26730,
      "student_distance_mean": 22.786448097229005,
      "teacher_distance_mean": 0.8599992752075195,
      "valid_action_samples": 214.9,
      "weighted_auxiliary_loss": 0.001956566807348281,
      "weighted_relation_loss": 0.001956566807348281
    },
    {
      "action_query_batch_std": 0.3478986084461212,
      "action_query_norm": 40.68893737792969,
      "action_relation_loss": 0.0019800390116870405,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.01403180542563,
      "flow_matching_loss": 0.016846116073429583,
      "grad_norm": 0.14422698318958282,
      "h_stock_valid_norm": 97.72510223388672,
      "invalid_tail_samples": 25.2,
      "learning_rate": 3.195730540438718e-06,
      "loss": 0.0188,
      "positive_teacher_pairs": 26532.7,
      "query_h_norm_ratio": 0.4163664162158966,
      "relation_loss": 0.0019800390116870405,
      "selected_pairs": 26532.7,
      "step": 26740,
      "student_distance_mean": 22.891070938110353,
      "teacher_distance_mean": 0.8633697330951691,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0019800390116870405,
      "weighted_relation_loss": 0.0019800390116870405
    },
    {
      "action_query_batch_std": 0.34357230067253114,
      "action_query_norm": 40.650838470458986,
      "action_relation_loss": 0.0019510584068484604,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.023386342376053,
      "flow_matching_loss": 0.018362791929394006,
      "grad_norm": 0.08828077465295792,
      "h_stock_valid_norm": 97.80401153564453,
      "invalid_tail_samples": 23.7,
      "learning_rate": 3.1763707866789336e-06,
      "loss": 0.0203,
      "positive_teacher_pairs": 26872.2,
      "query_h_norm_ratio": 0.41563856303691865,
      "relation_loss": 0.0019510584068484604,
      "selected_pairs": 26872.2,
      "step": 26750,
      "student_distance_mean": 22.521900177001953,
      "teacher_distance_mean": 0.8482341170310974,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0019510584068484604,
      "weighted_relation_loss": 0.0019510584068484604
    },
    {
      "action_query_batch_std": 0.34768947064876554,
      "action_query_norm": 40.70719833374024,
      "action_relation_loss": 0.0018893347936682404,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.03274087932647,
      "flow_matching_loss": 0.0159782943315804,
      "grad_norm": 0.09924796968698502,
      "h_stock_valid_norm": 97.86937942504883,
      "invalid_tail_samples": 21.7,
      "learning_rate": 3.1570679280438165e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 27334.5,
      "query_h_norm_ratio": 0.41593533754348755,
      "relation_loss": 0.0018893347936682404,
      "selected_pairs": 27334.5,
      "step": 26760,
      "student_distance_mean": 22.802927780151368,
      "teacher_distance_mean": 0.8601479887962341,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0018893347936682404,
      "weighted_relation_loss": 0.0018893347936682404
    },
    {
      "action_query_batch_std": 0.3445039510726929,
      "action_query_norm": 40.66517601013184,
      "action_relation_loss": 0.0019283033441752195,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.042095416276894,
      "flow_matching_loss": 0.015691744349896907,
      "grad_norm": 0.16171902418136597,
      "h_stock_valid_norm": 97.90906219482422,
      "invalid_tail_samples": 24.2,
      "learning_rate": 3.1378219879881767e-06,
      "loss": 0.0176,
      "positive_teacher_pairs": 26759.2,
      "query_h_norm_ratio": 0.4153375387191772,
      "relation_loss": 0.0019283033441752195,
      "selected_pairs": 26759.2,
      "step": 26770,
      "student_distance_mean": 22.656945991516114,
      "teacher_distance_mean": 0.8541248261928558,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0019283033441752195,
      "weighted_relation_loss": 0.0019283033441752195
    },
    {
      "action_query_batch_std": 0.3503099292516708,
      "action_query_norm": 40.70689163208008,
      "action_relation_loss": 0.0019565678667277098,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.051449953227316,
      "flow_matching_loss": 0.017094272747635843,
      "grad_norm": 0.0960322692990303,
      "h_stock_valid_norm": 98.01139831542969,
      "invalid_tail_samples": 21.9,
      "learning_rate": 3.11863298989763e-06,
      "loss": 0.0191,
      "positive_teacher_pairs": 27291.6,
      "query_h_norm_ratio": 0.4153303027153015,
      "relation_loss": 0.0019565678667277098,
      "selected_pairs": 27291.6,
      "step": 26780,
      "student_distance_mean": 22.971653556823732,
      "teacher_distance_mean": 0.8670066177845002,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.0019565678667277098,
      "weighted_relation_loss": 0.0019565678667277098
    },
    {
      "action_query_batch_std": 0.3492968946695328,
      "action_query_norm": 40.7423454284668,
      "action_relation_loss": 0.0019085398525930941,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.060804490177738,
      "flow_matching_loss": 0.015334452409297228,
      "grad_norm": 0.08223196864128113,
      "h_stock_valid_norm": 97.8815788269043,
      "invalid_tail_samples": 21.7,
      "learning_rate": 3.0995009570886303e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 27340.9,
      "query_h_norm_ratio": 0.4162467747926712,
      "relation_loss": 0.0019085398525930941,
      "selected_pairs": 27340.9,
      "step": 26790,
      "student_distance_mean": 22.875923728942873,
      "teacher_distance_mean": 0.8618344008922577,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0019085398525930941,
      "weighted_relation_loss": 0.0019085398525930941
    },
    {
      "action_query_batch_std": 0.3454472064971924,
      "action_query_norm": 40.684017944335935,
      "action_relation_loss": 0.001972932496573776,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.070159027128156,
      "flow_matching_loss": 0.014645525813102722,
      "grad_norm": 0.11827743053436279,
      "h_stock_valid_norm": 98.18706283569335,
      "invalid_tail_samples": 23.0,
      "learning_rate": 3.0804259128083936e-06,
      "loss": 0.0166,
      "positive_teacher_pairs": 27036.1,
      "query_h_norm_ratio": 0.4143550664186478,
      "relation_loss": 0.001972932496573776,
      "selected_pairs": 27036.1,
      "step": 26800,
      "student_distance_mean": 22.690138626098634,
      "teacher_distance_mean": 0.8528465867042542,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.001972932496573776,
      "weighted_relation_loss": 0.001972932496573776
    },
    {
      "action_query_batch_std": 0.34726628065109255,
      "action_query_norm": 40.68882675170899,
      "action_relation_loss": 0.0018688746728003024,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.079513564078578,
      "flow_matching_loss": 0.015867495723068715,
      "grad_norm": 0.13893365859985352,
      "h_stock_valid_norm": 97.98335037231445,
      "invalid_tail_samples": 20.5,
      "learning_rate": 3.06140788023489e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 27618.5,
      "query_h_norm_ratio": 0.4152654200792313,
      "relation_loss": 0.0018688746728003024,
      "selected_pairs": 27618.5,
      "step": 26810,
      "student_distance_mean": 22.78143424987793,
      "teacher_distance_mean": 0.8567586123943329,
      "valid_action_samples": 235.5,
      "weighted_auxiliary_loss": 0.0018688746728003024,
      "weighted_relation_loss": 0.0018688746728003024
    },
    {
      "action_query_batch_std": 0.3442356437444687,
      "action_query_norm": 40.668635559082034,
      "action_relation_loss": 0.0020070894388481975,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.088868101029,
      "flow_matching_loss": 0.01579457111656666,
      "grad_norm": 0.10261648148298264,
      "h_stock_valid_norm": 97.84101867675781,
      "invalid_tail_samples": 20.6,
      "learning_rate": 3.042446882476846e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 27602.7,
      "query_h_norm_ratio": 0.41566104590892794,
      "relation_loss": 0.0020070894388481975,
      "selected_pairs": 27602.7,
      "step": 26820,
      "student_distance_mean": 22.523268699645996,
      "teacher_distance_mean": 0.8488270342350006,
      "valid_action_samples": 235.4,
      "weighted_auxiliary_loss": 0.0020070894388481975,
      "weighted_relation_loss": 0.0020070894388481975
    },
    {
      "action_query_batch_std": 0.34112181663513186,
      "action_query_norm": 40.65599136352539,
      "action_relation_loss": 0.0019376832293346523,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.09822263797942,
      "flow_matching_loss": 0.015596618130803108,
      "grad_norm": 0.09194627404212952,
      "h_stock_valid_norm": 97.90914535522461,
      "invalid_tail_samples": 24.0,
      "learning_rate": 3.023542942573643e-06,
      "loss": 0.0175,
      "positive_teacher_pairs": 26807.6,
      "query_h_norm_ratio": 0.41524293422698977,
      "relation_loss": 0.0019376832293346523,
      "selected_pairs": 26807.6,
      "step": 26830,
      "student_distance_mean": 22.43730010986328,
      "teacher_distance_mean": 0.8456153273582458,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0019376832293346523,
      "weighted_relation_loss": 0.0019376832293346523
    },
    {
      "action_query_batch_std": 0.34867856502532957,
      "action_query_norm": 40.6970817565918,
      "action_relation_loss": 0.0019542088150046764,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.10757717492984,
      "flow_matching_loss": 0.014740274846553802,
      "grad_norm": 0.11499129235744476,
      "h_stock_valid_norm": 97.93331527709961,
      "invalid_tail_samples": 22.6,
      "learning_rate": 3.004696083495351e-06,
      "loss": 0.0167,
      "positive_teacher_pairs": 27126.7,
      "query_h_norm_ratio": 0.41556265354156496,
      "relation_loss": 0.0019542088150046764,
      "selected_pairs": 27126.7,
      "step": 26840,
      "student_distance_mean": 22.989385414123536,
      "teacher_distance_mean": 0.8667202532291413,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.0019542088150046764,
      "weighted_relation_loss": 0.0019542088150046764
    },
    {
      "action_query_batch_std": 0.3430558830499649,
      "action_query_norm": 40.64410209655762,
      "action_relation_loss": 0.0019945801352150737,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.116931711880262,
      "flow_matching_loss": 0.016683919075876474,
      "grad_norm": 0.1388736516237259,
      "h_stock_valid_norm": 97.9672966003418,
      "invalid_tail_samples": 22.9,
      "learning_rate": 2.98590632814269e-06,
      "loss": 0.0187,
      "positive_teacher_pairs": 27054.0,
      "query_h_norm_ratio": 0.4148767739534378,
      "relation_loss": 0.0019945801352150737,
      "selected_pairs": 27054.0,
      "step": 26850,
      "student_distance_mean": 22.53719310760498,
      "teacher_distance_mean": 0.8503017246723175,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0019945801352150737,
      "weighted_relation_loss": 0.0019945801352150737
    },
    {
      "action_query_batch_std": 0.34319622814655304,
      "action_query_norm": 40.639752578735354,
      "action_relation_loss": 0.001968870195560157,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.126286248830684,
      "flow_matching_loss": 0.01668990748003125,
      "grad_norm": 0.10907869786024094,
      "h_stock_valid_norm": 97.9736717224121,
      "invalid_tail_samples": 26.6,
      "learning_rate": 2.9671736993469702e-06,
      "loss": 0.0187,
      "positive_teacher_pairs": 26206.7,
      "query_h_norm_ratio": 0.41480604708194735,
      "relation_loss": 0.001968870195560157,
      "selected_pairs": 26206.7,
      "step": 26860,
      "student_distance_mean": 22.633358001708984,
      "teacher_distance_mean": 0.8546272993087769,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.001968870195560157,
      "weighted_relation_loss": 0.001968870195560157
    },
    {
      "action_query_batch_std": 0.34681912660598757,
      "action_query_norm": 40.67931632995605,
      "action_relation_loss": 0.0018924801144748927,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.135640785781103,
      "flow_matching_loss": 0.015185260120779276,
      "grad_norm": 0.09460456669330597,
      "h_stock_valid_norm": 98.04712829589843,
      "invalid_tail_samples": 24.3,
      "learning_rate": 2.9484982198701216e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 26733.7,
      "query_h_norm_ratio": 0.4148980170488358,
      "relation_loss": 0.0018924801144748927,
      "selected_pairs": 26733.7,
      "step": 26870,
      "student_distance_mean": 22.653191375732423,
      "teacher_distance_mean": 0.8525093376636506,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0018924801144748927,
      "weighted_relation_loss": 0.0018924801144748927
    },
    {
      "action_query_batch_std": 0.341787314414978,
      "action_query_norm": 40.63343124389648,
      "action_relation_loss": 0.0019248241442255676,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.144995322731525,
      "flow_matching_loss": 0.015682105626910924,
      "grad_norm": 0.10535989701747894,
      "h_stock_valid_norm": 97.97815017700195,
      "invalid_tail_samples": 22.5,
      "learning_rate": 2.929879912404604e-06,
      "loss": 0.0176,
      "positive_teacher_pairs": 27156.1,
      "query_h_norm_ratio": 0.41472240090370177,
      "relation_loss": 0.0019248241442255676,
      "selected_pairs": 27156.1,
      "step": 26880,
      "student_distance_mean": 22.542717933654785,
      "teacher_distance_mean": 0.8497180342674255,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0019248241442255676,
      "weighted_relation_loss": 0.0019248241442255676
    },
    {
      "action_query_batch_std": 0.34935579299926756,
      "action_query_norm": 40.71403160095215,
      "action_relation_loss": 0.001886040112003684,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.154349859681947,
      "flow_matching_loss": 0.01619606204330921,
      "grad_norm": 0.11401388794183731,
      "h_stock_valid_norm": 98.16274032592773,
      "invalid_tail_samples": 21.6,
      "learning_rate": 2.9113187995734147e-06,
      "loss": 0.0181,
      "positive_teacher_pairs": 27372.3,
      "query_h_norm_ratio": 0.4147636145353317,
      "relation_loss": 0.001886040112003684,
      "selected_pairs": 27372.3,
      "step": 26890,
      "student_distance_mean": 22.882273864746093,
      "teacher_distance_mean": 0.8623098790645599,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.001886040112003684,
      "weighted_relation_loss": 0.001886040112003684
    },
    {
      "action_query_batch_std": 0.3450066864490509,
      "action_query_norm": 40.700876235961914,
      "action_relation_loss": 0.0019413597066886723,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.163704396632365,
      "flow_matching_loss": 0.01424277713522315,
      "grad_norm": 0.08704135566949844,
      "h_stock_valid_norm": 98.11770629882812,
      "invalid_tail_samples": 21.3,
      "learning_rate": 2.8928149039300524e-06,
      "loss": 0.0162,
      "positive_teacher_pairs": 27436.6,
      "query_h_norm_ratio": 0.41482037007808686,
      "relation_loss": 0.0019413597066886723,
      "selected_pairs": 27436.6,
      "step": 26900,
      "student_distance_mean": 22.6435338973999,
      "teacher_distance_mean": 0.8545122385025025,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.0019413597066886723,
      "weighted_relation_loss": 0.0019413597066886723
    },
    {
      "action_query_batch_std": 0.34564279913902285,
      "action_query_norm": 40.66800308227539,
      "action_relation_loss": 0.0019151049200445414,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.173058933582787,
      "flow_matching_loss": 0.014995530527085067,
      "grad_norm": 0.17835548520088196,
      "h_stock_valid_norm": 97.98607025146484,
      "invalid_tail_samples": 22.9,
      "learning_rate": 2.8743682479584975e-06,
      "loss": 0.0169,
      "positive_teacher_pairs": 27062.2,
      "query_h_norm_ratio": 0.4150415390729904,
      "relation_loss": 0.0019151049200445414,
      "selected_pairs": 27062.2,
      "step": 26910,
      "student_distance_mean": 22.58873062133789,
      "teacher_distance_mean": 0.851270443201065,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0019151049200445414,
      "weighted_relation_loss": 0.0019151049200445414
    },
    {
      "action_query_batch_std": 0.3443750202655792,
      "action_query_norm": 40.67404060363769,
      "action_relation_loss": 0.001891134143806994,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.18241347053321,
      "flow_matching_loss": 0.01624330235645175,
      "grad_norm": 0.09439922124147415,
      "h_stock_valid_norm": 98.11776199340821,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.8559788540731824e-06,
      "loss": 0.0181,
      "positive_teacher_pairs": 26963.9,
      "query_h_norm_ratio": 0.41454412043094635,
      "relation_loss": 0.001891134143806994,
      "selected_pairs": 26963.9,
      "step": 26920,
      "student_distance_mean": 22.564712524414062,
      "teacher_distance_mean": 0.8519084513187408,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.001891134143806994,
      "weighted_relation_loss": 0.001891134143806994
    },
    {
      "action_query_batch_std": 0.3457783430814743,
      "action_query_norm": 40.685856628417966,
      "action_relation_loss": 0.0020259558456018566,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.19176800748363,
      "flow_matching_loss": 0.016238643880933525,
      "grad_norm": 0.1274799406528473,
      "h_stock_valid_norm": 98.10090408325195,
      "invalid_tail_samples": 23.7,
      "learning_rate": 2.8376467446189493e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 26874.1,
      "query_h_norm_ratio": 0.41473732590675355,
      "relation_loss": 0.0020259558456018566,
      "selected_pairs": 26874.1,
      "step": 26930,
      "student_distance_mean": 22.664089965820313,
      "teacher_distance_mean": 0.8545173525810241,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0020259558456018566,
      "weighted_relation_loss": 0.0020259558456018566
    },
    {
      "action_query_batch_std": 0.3458989918231964,
      "action_query_norm": 40.662583923339845,
      "action_relation_loss": 0.002054530510213226,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.20112254443405,
      "flow_matching_loss": 0.015245129261165858,
      "grad_norm": 0.10061103850603104,
      "h_stock_valid_norm": 98.05417098999024,
      "invalid_tail_samples": 24.5,
      "learning_rate": 2.8193719418710406e-06,
      "loss": 0.0173,
      "positive_teacher_pairs": 26698.5,
      "query_h_norm_ratio": 0.4146965265274048,
      "relation_loss": 0.002054530510213226,
      "selected_pairs": 26698.5,
      "step": 26940,
      "student_distance_mean": 22.728022003173827,
      "teacher_distance_mean": 0.8573425769805908,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.002054530510213226,
      "weighted_relation_loss": 0.002054530510213226
    },
    {
      "action_query_batch_std": 0.3443688333034515,
      "action_query_norm": 40.66809730529785,
      "action_relation_loss": 0.00199502658797428,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.21047708138447,
      "flow_matching_loss": 0.01591977272182703,
      "grad_norm": 0.09148260951042175,
      "h_stock_valid_norm": 97.9744369506836,
      "invalid_tail_samples": 23.5,
      "learning_rate": 2.8011544680350667e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 26916.6,
      "query_h_norm_ratio": 0.4150922954082489,
      "relation_loss": 0.00199502658797428,
      "selected_pairs": 26916.6,
      "step": 26950,
      "student_distance_mean": 22.517795181274415,
      "teacher_distance_mean": 0.847713416814804,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.00199502658797428,
      "weighted_relation_loss": 0.00199502658797428
    },
    {
      "action_query_batch_std": 0.3459135115146637,
      "action_query_norm": 40.659433364868164,
      "action_relation_loss": 0.001896512566599995,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.219831618334894,
      "flow_matching_loss": 0.01588888168334961,
      "grad_norm": 0.09352084994316101,
      "h_stock_valid_norm": 97.9565658569336,
      "invalid_tail_samples": 26.4,
      "learning_rate": 2.7829943452469753e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 26252.4,
      "query_h_norm_ratio": 0.41507965624332427,
      "relation_loss": 0.001896512566599995,
      "selected_pairs": 26252.4,
      "step": 26960,
      "student_distance_mean": 22.785854721069335,
      "teacher_distance_mean": 0.8591368079185486,
      "valid_action_samples": 229.6,
      "weighted_auxiliary_loss": 0.001896512566599995,
      "weighted_relation_loss": 0.001896512566599995
    },
    {
      "action_query_batch_std": 0.3432978242635727,
      "action_query_norm": 40.66664505004883,
      "action_relation_loss": 0.0019634329597465693,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.229186155285312,
      "flow_matching_loss": 0.014921514131128789,
      "grad_norm": 0.12919357419013977,
      "h_stock_valid_norm": 98.14544296264648,
      "invalid_tail_samples": 22.1,
      "learning_rate": 2.7648915955730213e-06,
      "loss": 0.0169,
      "positive_teacher_pairs": 27243.6,
      "query_h_norm_ratio": 0.4143533378839493,
      "relation_loss": 0.0019634329597465693,
      "selected_pairs": 27243.6,
      "step": 26970,
      "student_distance_mean": 22.53144245147705,
      "teacher_distance_mean": 0.8480069100856781,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0019634329597465693,
      "weighted_relation_loss": 0.0019634329597465693
    },
    {
      "action_query_batch_std": 0.3475156217813492,
      "action_query_norm": 40.698239135742185,
      "action_relation_loss": 0.001965594687499106,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.238540692235734,
      "flow_matching_loss": 0.014388961996883155,
      "grad_norm": 0.10291537642478943,
      "h_stock_valid_norm": 98.10369644165038,
      "invalid_tail_samples": 21.3,
      "learning_rate": 2.746846241009765e-06,
      "loss": 0.0164,
      "positive_teacher_pairs": 27429.0,
      "query_h_norm_ratio": 0.41485156714916227,
      "relation_loss": 0.001965594687499106,
      "selected_pairs": 27429.0,
      "step": 26980,
      "student_distance_mean": 22.853397178649903,
      "teacher_distance_mean": 0.8603223741054535,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.001965594687499106,
      "weighted_relation_loss": 0.001965594687499106
    },
    {
      "action_query_batch_std": 0.3450963288545609,
      "action_query_norm": 40.691443252563474,
      "action_relation_loss": 0.0018755578552372754,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.247895229186156,
      "flow_matching_loss": 0.016229571122676135,
      "grad_norm": 0.09269007295370102,
      "h_stock_valid_norm": 97.99169616699218,
      "invalid_tail_samples": 24.2,
      "learning_rate": 2.7288583034839945e-06,
      "loss": 0.0181,
      "positive_teacher_pairs": 26765.5,
      "query_h_norm_ratio": 0.41525608003139497,
      "relation_loss": 0.0018755578552372754,
      "selected_pairs": 26765.5,
      "step": 26990,
      "student_distance_mean": 22.67597942352295,
      "teacher_distance_mean": 0.8549090027809143,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0018755578552372754,
      "weighted_relation_loss": 0.0018755578552372754
    },
    {
      "action_query_batch_std": 0.34809300005435945,
      "action_query_norm": 40.70483932495117,
      "action_relation_loss": 0.0019025601330213248,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.257249766136574,
      "flow_matching_loss": 0.016913557425141335,
      "grad_norm": 0.09123868495225906,
      "h_stock_valid_norm": 98.00548553466797,
      "invalid_tail_samples": 22.8,
      "learning_rate": 2.7109278048527752e-06,
      "loss": 0.0188,
      "positive_teacher_pairs": 27078.4,
      "query_h_norm_ratio": 0.41533846855163575,
      "relation_loss": 0.0019025601330213248,
      "selected_pairs": 27078.4,
      "step": 27000,
      "student_distance_mean": 22.87450466156006,
      "teacher_distance_mean": 0.8628093361854553,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0019025601330213248,
      "weighted_relation_loss": 0.0019025601330213248
    },
    {
      "action_query_batch_std": 0.34324542582035067,
      "action_query_norm": 40.63694839477539,
      "action_relation_loss": 0.0020025007193908094,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.266604303086996,
      "flow_matching_loss": 0.015215238090604544,
      "grad_norm": 0.10852787643671036,
      "h_stock_valid_norm": 98.06474227905274,
      "invalid_tail_samples": 25.3,
      "learning_rate": 2.6930547669033413e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 26503.1,
      "query_h_norm_ratio": 0.4143896162509918,
      "relation_loss": 0.0020025007193908094,
      "selected_pairs": 26503.1,
      "step": 27010,
      "student_distance_mean": 22.48435878753662,
      "teacher_distance_mean": 0.8484947085380554,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0020025007193908094,
      "weighted_relation_loss": 0.0020025007193908094
    },
    {
      "action_query_batch_std": 0.3476018369197845,
      "action_query_norm": 40.68470726013184,
      "action_relation_loss": 0.0019869290059432386,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.27595884003742,
      "flow_matching_loss": 0.015749535989016294,
      "grad_norm": 0.0925414189696312,
      "h_stock_valid_norm": 97.90652770996094,
      "invalid_tail_samples": 23.5,
      "learning_rate": 2.675239211353109e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 26916.3,
      "query_h_norm_ratio": 0.4155476987361908,
      "relation_loss": 0.0019869290059432386,
      "selected_pairs": 26916.3,
      "step": 27020,
      "student_distance_mean": 22.76321086883545,
      "teacher_distance_mean": 0.8582277595996857,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0019869290059432386,
      "weighted_relation_loss": 0.0019869290059432386
    },
    {
      "action_query_batch_std": 0.3499350905418396,
      "action_query_norm": 40.7013801574707,
      "action_relation_loss": 0.001836814743001014,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.28531337698784,
      "flow_matching_loss": 0.015091963578015566,
      "grad_norm": 0.11346624791622162,
      "h_stock_valid_norm": 97.91535720825195,
      "invalid_tail_samples": 24.7,
      "learning_rate": 2.6574811598496786e-06,
      "loss": 0.0169,
      "positive_teacher_pairs": 26648.8,
      "query_h_norm_ratio": 0.4156795680522919,
      "relation_loss": 0.001836814743001014,
      "selected_pairs": 26648.8,
      "step": 27030,
      "student_distance_mean": 23.05766067504883,
      "teacher_distance_mean": 0.8689804315567017,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.001836814743001014,
      "weighted_relation_loss": 0.001836814743001014
    },
    {
      "action_query_batch_std": 0.3486230790615082,
      "action_query_norm": 40.68735008239746,
      "action_relation_loss": 0.0019231655867770314,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.29466791393826,
      "flow_matching_loss": 0.01640039449557662,
      "grad_norm": 0.09035792946815491,
      "h_stock_valid_norm": 98.06042098999023,
      "invalid_tail_samples": 26.0,
      "learning_rate": 2.6397806339707456e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 26339.6,
      "query_h_norm_ratio": 0.4149227201938629,
      "relation_loss": 0.0019231655867770314,
      "selected_pairs": 26339.6,
      "step": 27040,
      "student_distance_mean": 22.83183937072754,
      "teacher_distance_mean": 0.861409479379654,
      "valid_action_samples": 230.0,
      "weighted_auxiliary_loss": 0.0019231655867770314,
      "weighted_relation_loss": 0.0019231655867770314
    },
    {
      "action_query_batch_std": 0.34473329186439516,
      "action_query_norm": 40.659247970581056,
      "action_relation_loss": 0.001945733791217208,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.30402245088868,
      "flow_matching_loss": 0.014701371546834707,
      "grad_norm": 0.09935439378023148,
      "h_stock_valid_norm": 97.89580841064453,
      "invalid_tail_samples": 25.0,
      "learning_rate": 2.6221376552241217e-06,
      "loss": 0.0166,
      "positive_teacher_pairs": 26583.7,
      "query_h_norm_ratio": 0.4153341829776764,
      "relation_loss": 0.001945733791217208,
      "selected_pairs": 26583.7,
      "step": 27050,
      "student_distance_mean": 22.645831298828124,
      "teacher_distance_mean": 0.8542539477348328,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.001945733791217208,
      "weighted_relation_loss": 0.001945733791217208
    },
    {
      "action_query_batch_std": 0.3426962047815323,
      "action_query_norm": 40.637240982055665,
      "action_relation_loss": 0.0019648698158562183,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.313376987839103,
      "flow_matching_loss": 0.014543025195598603,
      "grad_norm": 0.1105358749628067,
      "h_stock_valid_norm": 97.87279739379883,
      "invalid_tail_samples": 24.5,
      "learning_rate": 2.604552245047681e-06,
      "loss": 0.0165,
      "positive_teacher_pairs": 26685.0,
      "query_h_norm_ratio": 0.4152048945426941,
      "relation_loss": 0.0019648698158562183,
      "selected_pairs": 26685.0,
      "step": 27060,
      "student_distance_mean": 22.56572265625,
      "teacher_distance_mean": 0.8518096685409546,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0019648698158562183,
      "weighted_relation_loss": 0.0019648698158562183
    },
    {
      "action_query_batch_std": 0.34049591422080994,
      "action_query_norm": 40.61926345825195,
      "action_relation_loss": 0.0019043369335122406,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.32273152478952,
      "flow_matching_loss": 0.01574651850387454,
      "grad_norm": 0.07963034510612488,
      "h_stock_valid_norm": 98.09366455078126,
      "invalid_tail_samples": 25.2,
      "learning_rate": 2.587024424809359e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 26525.5,
      "query_h_norm_ratio": 0.41408749520778654,
      "relation_loss": 0.0019043369335122406,
      "selected_pairs": 26525.5,
      "step": 27070,
      "student_distance_mean": 22.385220527648926,
      "teacher_distance_mean": 0.8435228109359741,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0019043369335122406,
      "weighted_relation_loss": 0.0019043369335122406
    },
    {
      "action_query_batch_std": 0.34426186978816986,
      "action_query_norm": 40.653838348388675,
      "action_relation_loss": 0.0019301175023429095,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.332086061739943,
      "flow_matching_loss": 0.014898506551980972,
      "grad_norm": 0.10094238817691803,
      "h_stock_valid_norm": 97.96161804199218,
      "invalid_tail_samples": 23.5,
      "learning_rate": 2.5695542158071186e-06,
      "loss": 0.0168,
      "positive_teacher_pairs": 26923.0,
      "query_h_norm_ratio": 0.41499973833560944,
      "relation_loss": 0.0019301175023429095,
      "selected_pairs": 26923.0,
      "step": 27080,
      "student_distance_mean": 22.533505630493163,
      "teacher_distance_mean": 0.8505742609500885,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0019301175023429095,
      "weighted_relation_loss": 0.0019301175023429095
    },
    {
      "action_query_batch_std": 0.34348232448101046,
      "action_query_norm": 40.63966331481934,
      "action_relation_loss": 0.0019098313874565064,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.341440598690365,
      "flow_matching_loss": 0.015362250618636608,
      "grad_norm": 0.11867701262235641,
      "h_stock_valid_norm": 97.96804122924804,
      "invalid_tail_samples": 27.3,
      "learning_rate": 2.5521416392689068e-06,
      "loss": 0.0173,
      "positive_teacher_pairs": 26053.7,
      "query_h_norm_ratio": 0.4148290574550629,
      "relation_loss": 0.0019098313874565064,
      "selected_pairs": 26053.7,
      "step": 27090,
      "student_distance_mean": 22.603256225585938,
      "teacher_distance_mean": 0.851618093252182,
      "valid_action_samples": 228.7,
      "weighted_auxiliary_loss": 0.0019098313874565064,
      "weighted_relation_loss": 0.0019098313874565064
    },
    {
      "action_query_batch_std": 0.34703221917152405,
      "action_query_norm": 40.67220458984375,
      "action_relation_loss": 0.0018942214781418442,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.350795135640787,
      "flow_matching_loss": 0.015151983313262462,
      "grad_norm": 0.1340491622686386,
      "h_stock_valid_norm": 97.98514251708984,
      "invalid_tail_samples": 24.7,
      "learning_rate": 2.5347867163526384e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 26646.4,
      "query_h_norm_ratio": 0.41508733928203584,
      "relation_loss": 0.0018942214781418442,
      "selected_pairs": 26646.4,
      "step": 27100,
      "student_distance_mean": 22.77939338684082,
      "teacher_distance_mean": 0.8586500942707062,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0018942214781418442,
      "weighted_relation_loss": 0.0018942214781418442
    },
    {
      "action_query_batch_std": 0.34134902358055114,
      "action_query_norm": 40.618930435180665,
      "action_relation_loss": 0.002042641444131732,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.360149672591206,
      "flow_matching_loss": 0.016027580201625823,
      "grad_norm": 0.10823684185743332,
      "h_stock_valid_norm": 97.89800033569335,
      "invalid_tail_samples": 26.1,
      "learning_rate": 2.517489468146189e-06,
      "loss": 0.0181,
      "positive_teacher_pairs": 26331.1,
      "query_h_norm_ratio": 0.4149148315191269,
      "relation_loss": 0.002042641444131732,
      "selected_pairs": 26331.1,
      "step": 27110,
      "student_distance_mean": 22.4633150100708,
      "teacher_distance_mean": 0.8470806658267975,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.002042641444131732,
      "weighted_relation_loss": 0.002042641444131732
    },
    {
      "action_query_batch_std": 0.349056601524353,
      "action_query_norm": 40.700482559204104,
      "action_relation_loss": 0.001938857592176646,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.369504209541628,
      "flow_matching_loss": 0.01590910432860255,
      "grad_norm": 0.12095026671886444,
      "h_stock_valid_norm": 97.94656066894531,
      "invalid_tail_samples": 23.0,
      "learning_rate": 2.500249915667341e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 27032.1,
      "query_h_norm_ratio": 0.41554035246372223,
      "relation_loss": 0.001938857592176646,
      "selected_pairs": 27032.1,
      "step": 27120,
      "student_distance_mean": 22.903995513916016,
      "teacher_distance_mean": 0.8634238302707672,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.001938857592176646,
      "weighted_relation_loss": 0.001938857592176646
    },
    {
      "action_query_batch_std": 0.34799792766571047,
      "action_query_norm": 40.683467102050784,
      "action_relation_loss": 0.0018716322490945458,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.37885874649205,
      "flow_matching_loss": 0.014598044659942388,
      "grad_norm": 0.09559358656406403,
      "h_stock_valid_norm": 98.01266250610351,
      "invalid_tail_samples": 23.6,
      "learning_rate": 2.4830680798637817e-06,
      "loss": 0.0165,
      "positive_teacher_pairs": 26896.3,
      "query_h_norm_ratio": 0.41509132981300356,
      "relation_loss": 0.0018716322490945458,
      "selected_pairs": 26896.3,
      "step": 27130,
      "student_distance_mean": 22.77561836242676,
      "teacher_distance_mean": 0.8586519896984101,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0018716322490945458,
      "weighted_relation_loss": 0.0018716322490945458
    },
    {
      "action_query_batch_std": 0.34497091770172117,
      "action_query_norm": 40.67591171264648,
      "action_relation_loss": 0.0018527789041399955,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.388213283442468,
      "flow_matching_loss": 0.016418429277837276,
      "grad_norm": 0.09227987378835678,
      "h_stock_valid_norm": 98.0499755859375,
      "invalid_tail_samples": 21.7,
      "learning_rate": 2.465943981613056e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 27339.1,
      "query_h_norm_ratio": 0.4148548930883408,
      "relation_loss": 0.0018527789041399955,
      "selected_pairs": 27339.1,
      "step": 27140,
      "student_distance_mean": 22.610748100280762,
      "teacher_distance_mean": 0.8521281659603119,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0018527789041399955,
      "weighted_relation_loss": 0.0018527789041399955
    },
    {
      "action_query_batch_std": 0.3449141263961792,
      "action_query_norm": 40.65254898071289,
      "action_relation_loss": 0.0019770081737078725,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.39756782039289,
      "flow_matching_loss": 0.016286629252135754,
      "grad_norm": 0.07993543148040771,
      "h_stock_valid_norm": 97.9359230041504,
      "invalid_tail_samples": 22.4,
      "learning_rate": 2.448877641722569e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 27184.9,
      "query_h_norm_ratio": 0.4151004582643509,
      "relation_loss": 0.0019770081737078725,
      "selected_pairs": 27184.9,
      "step": 27150,
      "student_distance_mean": 22.610559463500977,
      "teacher_distance_mean": 0.8509602844715118,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.0019770081737078725,
      "weighted_relation_loss": 0.0019770081737078725
    },
    {
      "action_query_batch_std": 0.34555940628051757,
      "action_query_norm": 40.66870269775391,
      "action_relation_loss": 0.0019179247785359621,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.406922357343312,
      "flow_matching_loss": 0.01627912148833275,
      "grad_norm": 0.09121386706829071,
      "h_stock_valid_norm": 97.85817794799804,
      "invalid_tail_samples": 21.7,
      "learning_rate": 2.431869080929522e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 27335.9,
      "query_h_norm_ratio": 0.4155897706747055,
      "relation_loss": 0.0019179247785359621,
      "selected_pairs": 27335.9,
      "step": 27160,
      "student_distance_mean": 22.708698081970216,
      "teacher_distance_mean": 0.8553396344184876,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0019179247785359621,
      "weighted_relation_loss": 0.0019179247785359621
    },
    {
      "action_query_batch_std": 0.3488311797380447,
      "action_query_norm": 40.701083755493165,
      "action_relation_loss": 0.001873276918195188,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.416276894293734,
      "flow_matching_loss": 0.015647603012621403,
      "grad_norm": 0.09626397490501404,
      "h_stock_valid_norm": 98.00263442993165,
      "invalid_tail_samples": 23.7,
      "learning_rate": 2.4149183199009216e-06,
      "loss": 0.0175,
      "positive_teacher_pairs": 26875.0,
      "query_h_norm_ratio": 0.41530824899673463,
      "relation_loss": 0.001873276918195188,
      "selected_pairs": 26875.0,
      "step": 27170,
      "student_distance_mean": 22.88181495666504,
      "teacher_distance_mean": 0.8618137419223786,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.001873276918195188,
      "weighted_relation_loss": 0.001873276918195188
    },
    {
      "action_query_batch_std": 0.34292420744895935,
      "action_query_norm": 40.673356246948245,
      "action_relation_loss": 0.0018527515465393662,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.425631431244152,
      "flow_matching_loss": 0.01519375927746296,
      "grad_norm": 0.11646673083305359,
      "h_stock_valid_norm": 98.12925415039062,
      "invalid_tail_samples": 24.1,
      "learning_rate": 2.3980253792335427e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 26785.0,
      "query_h_norm_ratio": 0.4144929498434067,
      "relation_loss": 0.0018527515465393662,
      "selected_pairs": 26785.0,
      "step": 27180,
      "student_distance_mean": 22.607297706604005,
      "teacher_distance_mean": 0.8519021689891815,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0018527515465393662,
      "weighted_relation_loss": 0.0018527515465393662
    },
    {
      "action_query_batch_std": 0.34364646673202515,
      "action_query_norm": 40.65620422363281,
      "action_relation_loss": 0.0019122885656543075,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.434985968194574,
      "flow_matching_loss": 0.015298715513199567,
      "grad_norm": 0.0845905989408493,
      "h_stock_valid_norm": 97.96980514526368,
      "invalid_tail_samples": 24.2,
      "learning_rate": 2.381190279453899e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 26754.7,
      "query_h_norm_ratio": 0.41498822569847105,
      "relation_loss": 0.0019122885656543075,
      "selected_pairs": 26754.7,
      "step": 27190,
      "student_distance_mean": 22.498014068603517,
      "teacher_distance_mean": 0.8475596606731415,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0019122885656543075,
      "weighted_relation_loss": 0.0019122885656543075
    },
    {
      "action_query_batch_std": 0.34018336832523344,
      "action_query_norm": 40.62916107177735,
      "action_relation_loss": 0.001963926572352648,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.444340505144996,
      "flow_matching_loss": 0.014974153321236371,
      "grad_norm": 0.11981850117444992,
      "h_stock_valid_norm": 97.98533630371094,
      "invalid_tail_samples": 24.6,
      "learning_rate": 2.364413041018232e-06,
      "loss": 0.0169,
      "positive_teacher_pairs": 26662.0,
      "query_h_norm_ratio": 0.4146476000547409,
      "relation_loss": 0.001963926572352648,
      "selected_pairs": 26662.0,
      "step": 27200,
      "student_distance_mean": 22.367994689941405,
      "teacher_distance_mean": 0.8434945046901703,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.001963926572352648,
      "weighted_relation_loss": 0.001963926572352648
    },
    {
      "action_query_batch_std": 0.3465610176324844,
      "action_query_norm": 40.67606658935547,
      "action_relation_loss": 0.0019346918212249874,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.453695042095415,
      "flow_matching_loss": 0.015735315531492232,
      "grad_norm": 0.08239033818244934,
      "h_stock_valid_norm": 98.0483772277832,
      "invalid_tail_samples": 21.8,
      "learning_rate": 2.347693684312463e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 27317.7,
      "query_h_norm_ratio": 0.4148605912923813,
      "relation_loss": 0.0019346918212249874,
      "selected_pairs": 27317.7,
      "step": 27210,
      "student_distance_mean": 22.705980682373045,
      "teacher_distance_mean": 0.8561914145946503,
      "valid_action_samples": 234.2,
      "weighted_auxiliary_loss": 0.0019346918212249874,
      "weighted_relation_loss": 0.0019346918212249874
    },
    {
      "action_query_batch_std": 0.343648961186409,
      "action_query_norm": 40.65705947875976,
      "action_relation_loss": 0.0019367780769243836,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.463049579045837,
      "flow_matching_loss": 0.015755092725157738,
      "grad_norm": 0.11373892426490784,
      "h_stock_valid_norm": 97.96172332763672,
      "invalid_tail_samples": 24.6,
      "learning_rate": 2.331032229652186e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 26672.8,
      "query_h_norm_ratio": 0.4150357097387314,
      "relation_loss": 0.0019367780769243836,
      "selected_pairs": 26672.8,
      "step": 27220,
      "student_distance_mean": 22.527646446228026,
      "teacher_distance_mean": 0.8498398244380951,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0019367780769243836,
      "weighted_relation_loss": 0.0019367780769243836
    },
    {
      "action_query_batch_std": 0.34774781167507174,
      "action_query_norm": 40.695610427856444,
      "action_relation_loss": 0.0018979868735186755,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.47240411599626,
      "flow_matching_loss": 0.015210594981908798,
      "grad_norm": 0.08540168404579163,
      "h_stock_valid_norm": 98.03848876953126,
      "invalid_tail_samples": 25.0,
      "learning_rate": 2.314428697282628e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 26569.9,
      "query_h_norm_ratio": 0.4150995433330536,
      "relation_loss": 0.0018979868735186755,
      "selected_pairs": 26569.9,
      "step": 27230,
      "student_distance_mean": 22.81017017364502,
      "teacher_distance_mean": 0.8617422223091126,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0018979868735186755,
      "weighted_relation_loss": 0.0018979868735186755
    },
    {
      "action_query_batch_std": 0.34274354577064514,
      "action_query_norm": 40.677681350708006,
      "action_relation_loss": 0.001888043701183051,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.48175865294668,
      "flow_matching_loss": 0.015144103299826383,
      "grad_norm": 0.07712091505527496,
      "h_stock_valid_norm": 97.90103530883789,
      "invalid_tail_samples": 25.3,
      "learning_rate": 2.297883107378673e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 26504.0,
      "query_h_norm_ratio": 0.41550052762031553,
      "relation_loss": 0.001888043701183051,
      "selected_pairs": 26504.0,
      "step": 27240,
      "student_distance_mean": 22.47870864868164,
      "teacher_distance_mean": 0.8464844405651093,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.001888043701183051,
      "weighted_relation_loss": 0.001888043701183051
    },
    {
      "action_query_batch_std": 0.3495825886726379,
      "action_query_norm": 40.71270599365234,
      "action_relation_loss": 0.0019428591243922711,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.4911131898971,
      "flow_matching_loss": 0.015298534836620092,
      "grad_norm": 0.08141252398490906,
      "h_stock_valid_norm": 97.92979049682617,
      "invalid_tail_samples": 23.8,
      "learning_rate": 2.2813954800447513e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 26848.6,
      "query_h_norm_ratio": 0.4157359778881073,
      "relation_loss": 0.0019428591243922711,
      "selected_pairs": 26848.6,
      "step": 27250,
      "student_distance_mean": 22.935436058044434,
      "teacher_distance_mean": 0.8650087714195251,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0019428591243922711,
      "weighted_relation_loss": 0.0019428591243922711
    },
    {
      "action_query_batch_std": 0.3462283700704575,
      "action_query_norm": 40.66656303405762,
      "action_relation_loss": 0.001997482217848301,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.50046772684752,
      "flow_matching_loss": 0.015168244950473309,
      "grad_norm": 0.09995105862617493,
      "h_stock_valid_norm": 98.12328414916992,
      "invalid_tail_samples": 22.3,
      "learning_rate": 2.2649658353148974e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 27200.8,
      "query_h_norm_ratio": 0.41444429755210876,
      "relation_loss": 0.001997482217848301,
      "selected_pairs": 27200.8,
      "step": 27260,
      "student_distance_mean": 22.772772216796874,
      "teacher_distance_mean": 0.8588058054447174,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.001997482217848301,
      "weighted_relation_loss": 0.001997482217848301
    },
    {
      "action_query_batch_std": 0.3446668595075607,
      "action_query_norm": 40.69507827758789,
      "action_relation_loss": 0.0019340030616149307,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.509822263797943,
      "flow_matching_loss": 0.015718573797494174,
      "grad_norm": 0.10350213944911957,
      "h_stock_valid_norm": 97.81791381835937,
      "invalid_tail_samples": 23.6,
      "learning_rate": 2.2485941931526645e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 26897.3,
      "query_h_norm_ratio": 0.4160305172204971,
      "relation_loss": 0.0019340030616149307,
      "selected_pairs": 26897.3,
      "step": 27270,
      "student_distance_mean": 22.6410270690918,
      "teacher_distance_mean": 0.8533783793449402,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0019340030616149307,
      "weighted_relation_loss": 0.0019340030616149307
    },
    {
      "action_query_batch_std": 0.34349962770938874,
      "action_query_norm": 40.64754295349121,
      "action_relation_loss": 0.0018297760398127139,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.51917680074836,
      "flow_matching_loss": 0.01597054097801447,
      "grad_norm": 0.08683324605226517,
      "h_stock_valid_norm": 97.81614151000977,
      "invalid_tail_samples": 23.0,
      "learning_rate": 2.232280573451151e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 27030.8,
      "query_h_norm_ratio": 0.4155514776706696,
      "relation_loss": 0.0018297760398127139,
      "selected_pairs": 27030.8,
      "step": 27280,
      "student_distance_mean": 22.492611503601076,
      "teacher_distance_mean": 0.8479676723480225,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0018297760398127139,
      "weighted_relation_loss": 0.0018297760398127139
    },
    {
      "action_query_batch_std": 0.3432170420885086,
      "action_query_norm": 40.63569259643555,
      "action_relation_loss": 0.00203783109318465,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.528531337698784,
      "flow_matching_loss": 0.016564443241804838,
      "grad_norm": 0.06960930675268173,
      "h_stock_valid_norm": 97.93950653076172,
      "invalid_tail_samples": 25.5,
      "learning_rate": 2.216024996032945e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 26465.1,
      "query_h_norm_ratio": 0.41490898132324217,
      "relation_loss": 0.00203783109318465,
      "selected_pairs": 26465.1,
      "step": 27290,
      "student_distance_mean": 22.58141975402832,
      "teacher_distance_mean": 0.849332070350647,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.00203783109318465,
      "weighted_relation_loss": 0.00203783109318465
    },
    {
      "action_query_batch_std": 0.34639004766941073,
      "action_query_norm": 40.68256797790527,
      "action_relation_loss": 0.0020201773731969297,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.537885874649206,
      "flow_matching_loss": 0.01553366957232356,
      "grad_norm": 0.095244400203228,
      "h_stock_valid_norm": 97.89529876708984,
      "invalid_tail_samples": 22.6,
      "learning_rate": 2.1998274806501074e-06,
      "loss": 0.0176,
      "positive_teacher_pairs": 27131.8,
      "query_h_norm_ratio": 0.41557621359825136,
      "relation_loss": 0.0020201773731969297,
      "selected_pairs": 27131.8,
      "step": 27300,
      "student_distance_mean": 22.730785369873047,
      "teacher_distance_mean": 0.8571480274200439,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.0020201773731969297,
      "weighted_relation_loss": 0.0020201773731969297
    },
    {
      "action_query_batch_std": 0.3468508332967758,
      "action_query_norm": 40.69002342224121,
      "action_relation_loss": 0.0019598307088017465,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.547240411599624,
      "flow_matching_loss": 0.0162819960154593,
      "grad_norm": 0.13109542429447174,
      "h_stock_valid_norm": 98.04272003173828,
      "invalid_tail_samples": 24.4,
      "learning_rate": 2.183688046984139e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 26712.3,
      "query_h_norm_ratio": 0.41502794027328493,
      "relation_loss": 0.0019598307088017465,
      "selected_pairs": 26712.3,
      "step": 27310,
      "student_distance_mean": 22.73697166442871,
      "teacher_distance_mean": 0.8571777224540711,
      "valid_action_samples": 231.6,
      "weighted_auxiliary_loss": 0.0019598307088017465,
      "weighted_relation_loss": 0.0019598307088017465
    },
    {
      "action_query_batch_std": 0.33859330117702485,
      "action_query_norm": 40.63140296936035,
      "action_relation_loss": 0.0019798492081463336,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.556594948550046,
      "flow_matching_loss": 0.015661922004073858,
      "grad_norm": 0.09045828133821487,
      "h_stock_valid_norm": 98.10991821289062,
      "invalid_tail_samples": 22.3,
      "learning_rate": 2.167606714645981e-06,
      "loss": 0.0176,
      "positive_teacher_pairs": 27203.6,
      "query_h_norm_ratio": 0.41414450109004974,
      "relation_loss": 0.0019798492081463336,
      "selected_pairs": 27203.6,
      "step": 27320,
      "student_distance_mean": 22.333481979370116,
      "teacher_distance_mean": 0.8424164295196533,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0019798492081463336,
      "weighted_relation_loss": 0.0019798492081463336
    },
    {
      "action_query_batch_std": 0.3436519742012024,
      "action_query_norm": 40.66095504760742,
      "action_relation_loss": 0.0020543833612464367,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.565949485500468,
      "flow_matching_loss": 0.015910045895725488,
      "grad_norm": 0.11124507337808609,
      "h_stock_valid_norm": 97.83974761962891,
      "invalid_tail_samples": 25.8,
      "learning_rate": 2.151583503175958e-06,
      "loss": 0.018,
      "positive_teacher_pairs": 26394.5,
      "query_h_norm_ratio": 0.41558962762355806,
      "relation_loss": 0.0020543833612464367,
      "selected_pairs": 26394.5,
      "step": 27330,
      "student_distance_mean": 22.613255310058594,
      "teacher_distance_mean": 0.852581912279129,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0020543833612464367,
      "weighted_relation_loss": 0.0020543833612464367
    },
    {
      "action_query_batch_std": 0.3446756452322006,
      "action_query_norm": 40.67120628356933,
      "action_relation_loss": 0.001901570486370474,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.57530402245089,
      "flow_matching_loss": 0.014501342736184596,
      "grad_norm": 0.08874388784170151,
      "h_stock_valid_norm": 98.05528717041015,
      "invalid_tail_samples": 23.8,
      "learning_rate": 2.1356184320437957e-06,
      "loss": 0.0164,
      "positive_teacher_pairs": 26857.3,
      "query_h_norm_ratio": 0.41478319466114044,
      "relation_loss": 0.001901570486370474,
      "selected_pairs": 26857.3,
      "step": 27340,
      "student_distance_mean": 22.72081050872803,
      "teacher_distance_mean": 0.8569000601768494,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.001901570486370474,
      "weighted_relation_loss": 0.001901570486370474
    },
    {
      "action_query_batch_std": 0.3460662871599197,
      "action_query_norm": 40.67395706176758,
      "action_relation_loss": 0.001985070761293173,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.58465855940131,
      "flow_matching_loss": 0.014952819235622883,
      "grad_norm": 0.08767404407262802,
      "h_stock_valid_norm": 98.08633193969726,
      "invalid_tail_samples": 24.3,
      "learning_rate": 2.119711520648554e-06,
      "loss": 0.0169,
      "positive_teacher_pairs": 26735.5,
      "query_h_norm_ratio": 0.4146785408258438,
      "relation_loss": 0.001985070761293173,
      "selected_pairs": 26735.5,
      "step": 27350,
      "student_distance_mean": 22.691884231567382,
      "teacher_distance_mean": 0.8559800982475281,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.001985070761293173,
      "weighted_relation_loss": 0.001985070761293173
    },
    {
      "action_query_batch_std": 0.34484581351280214,
      "action_query_norm": 40.659262084960936,
      "action_relation_loss": 0.0019720967393368483,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.59401309635173,
      "flow_matching_loss": 0.015447068493813276,
      "grad_norm": 0.13890589773654938,
      "h_stock_valid_norm": 98.02774810791016,
      "invalid_tail_samples": 24.6,
      "learning_rate": 2.103862788318628e-06,
      "loss": 0.0174,
      "positive_teacher_pairs": 26667.0,
      "query_h_norm_ratio": 0.4147760570049286,
      "relation_loss": 0.0019720967393368483,
      "selected_pairs": 26667.0,
      "step": 27360,
      "student_distance_mean": 22.601177978515626,
      "teacher_distance_mean": 0.8518382549285889,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0019720967393368483,
      "weighted_relation_loss": 0.0019720967393368483
    },
    {
      "action_query_batch_std": 0.34517434537410735,
      "action_query_norm": 40.693154907226564,
      "action_relation_loss": 0.001977439795155078,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.603367633302152,
      "flow_matching_loss": 0.01619754396378994,
      "grad_norm": 0.1295357197523117,
      "h_stock_valid_norm": 97.99426803588867,
      "invalid_tail_samples": 21.5,
      "learning_rate": 2.088072254311729e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 27383.4,
      "query_h_norm_ratio": 0.41526309251785276,
      "relation_loss": 0.001977439795155078,
      "selected_pairs": 27383.4,
      "step": 27370,
      "student_distance_mean": 22.56978874206543,
      "teacher_distance_mean": 0.8513988018035888,
      "valid_action_samples": 234.5,
      "weighted_auxiliary_loss": 0.001977439795155078,
      "weighted_relation_loss": 0.001977439795155078
    },
    {
      "action_query_batch_std": 0.3492521524429321,
      "action_query_norm": 40.707894515991214,
      "action_relation_loss": 0.0019693309091962873,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.61272217025257,
      "flow_matching_loss": 0.01515886876732111,
      "grad_norm": 0.10722094774246216,
      "h_stock_valid_norm": 97.9738998413086,
      "invalid_tail_samples": 20.4,
      "learning_rate": 2.0723399378148435e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 27645.2,
      "query_h_norm_ratio": 0.4155005425214767,
      "relation_loss": 0.0019693309091962873,
      "selected_pairs": 27645.2,
      "step": 27380,
      "student_distance_mean": 22.87575569152832,
      "teacher_distance_mean": 0.863320791721344,
      "valid_action_samples": 235.6,
      "weighted_auxiliary_loss": 0.0019693309091962873,
      "weighted_relation_loss": 0.0019693309091962873
    },
    {
      "action_query_batch_std": 0.3447621017694473,
      "action_query_norm": 40.660438537597656,
      "action_relation_loss": 0.001867728680372238,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.622076707202993,
      "flow_matching_loss": 0.01529544936493039,
      "grad_norm": 0.08114355802536011,
      "h_stock_valid_norm": 98.04294891357422,
      "invalid_tail_samples": 21.4,
      "learning_rate": 2.0566658579442065e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 27403.9,
      "query_h_norm_ratio": 0.41472279727458955,
      "relation_loss": 0.001867728680372238,
      "selected_pairs": 27403.9,
      "step": 27390,
      "student_distance_mean": 22.562289810180665,
      "teacher_distance_mean": 0.8509093761444092,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.001867728680372238,
      "weighted_relation_loss": 0.001867728680372238
    },
    {
      "action_query_batch_std": 0.34330333173274996,
      "action_query_norm": 40.668484115600585,
      "action_relation_loss": 0.0019465878838673234,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.631431244153415,
      "flow_matching_loss": 0.015670732129365206,
      "grad_norm": 0.1333768218755722,
      "h_stock_valid_norm": 98.01232833862305,
      "invalid_tail_samples": 24.6,
      "learning_rate": 2.0410500337453176e-06,
      "loss": 0.0176,
      "positive_teacher_pairs": 26668.3,
      "query_h_norm_ratio": 0.4149348229169846,
      "relation_loss": 0.0019465878838673234,
      "selected_pairs": 26668.3,
      "step": 27400,
      "student_distance_mean": 22.631255531311034,
      "teacher_distance_mean": 0.8528038382530212,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0019465878838673234,
      "weighted_relation_loss": 0.0019465878838673234
    },
    {
      "action_query_batch_std": 0.34649922549724577,
      "action_query_norm": 40.678989028930665,
      "action_relation_loss": 0.001867644931189716,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.640785781103837,
      "flow_matching_loss": 0.015129910316318274,
      "grad_norm": 0.08835698664188385,
      "h_stock_valid_norm": 97.98555297851563,
      "invalid_tail_samples": 25.2,
      "learning_rate": 2.0254924841928645e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 26539.0,
      "query_h_norm_ratio": 0.41515462696552274,
      "relation_loss": 0.001867644931189716,
      "selected_pairs": 26539.0,
      "step": 27410,
      "student_distance_mean": 22.782732009887695,
      "teacher_distance_mean": 0.860043340921402,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.001867644931189716,
      "weighted_relation_loss": 0.001867644931189716
    },
    {
      "action_query_batch_std": 0.3468936085700989,
      "action_query_norm": 40.67963829040527,
      "action_relation_loss": 0.0018984768190421163,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.650140318054255,
      "flow_matching_loss": 0.014350426569581031,
      "grad_norm": 0.08989267796278,
      "h_stock_valid_norm": 97.77962951660156,
      "invalid_tail_samples": 25.2,
      "learning_rate": 2.009993228190754e-06,
      "loss": 0.0162,
      "positive_teacher_pairs": 26538.6,
      "query_h_norm_ratio": 0.41603688299655917,
      "relation_loss": 0.0018984768190421163,
      "selected_pairs": 26538.6,
      "step": 27420,
      "student_distance_mean": 22.731153869628905,
      "teacher_distance_mean": 0.8571206033229828,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0018984768190421163,
      "weighted_relation_loss": 0.0018984768190421163
    },
    {
      "action_query_batch_std": 0.34572671055793763,
      "action_query_norm": 40.6726016998291,
      "action_relation_loss": 0.0018173379823565484,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.659494855004677,
      "flow_matching_loss": 0.015732401795685292,
      "grad_norm": 0.10680362582206726,
      "h_stock_valid_norm": 97.8979476928711,
      "invalid_tail_samples": 29.2,
      "learning_rate": 1.9945522845720323e-06,
      "loss": 0.0175,
      "positive_teacher_pairs": 25618.6,
      "query_h_norm_ratio": 0.4154618978500366,
      "relation_loss": 0.0018173379823565484,
      "selected_pairs": 25618.6,
      "step": 27430,
      "student_distance_mean": 22.63339786529541,
      "teacher_distance_mean": 0.8543327510356903,
      "valid_action_samples": 226.8,
      "weighted_auxiliary_loss": 0.0018173379823565484,
      "weighted_relation_loss": 0.0018173379823565484
    },
    {
      "action_query_batch_std": 0.3480029582977295,
      "action_query_norm": 40.6808895111084,
      "action_relation_loss": 0.0019632114679552616,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.6688493919551,
      "flow_matching_loss": 0.01517362855374813,
      "grad_norm": 0.1219492256641388,
      "h_stock_valid_norm": 98.00642166137695,
      "invalid_tail_samples": 22.6,
      "learning_rate": 1.9791696720988963e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 27127.9,
      "query_h_norm_ratio": 0.4150881379842758,
      "relation_loss": 0.0019632114679552616,
      "selected_pairs": 27127.9,
      "step": 27440,
      "student_distance_mean": 22.8330379486084,
      "teacher_distance_mean": 0.8610141038894653,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.0019632114679552616,
      "weighted_relation_loss": 0.0019632114679552616
    },
    {
      "action_query_batch_std": 0.3418247550725937,
      "action_query_norm": 40.646942901611325,
      "action_relation_loss": 0.0019601422711275516,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.678203928905518,
      "flow_matching_loss": 0.015718217939138412,
      "grad_norm": 0.09429173171520233,
      "h_stock_valid_norm": 97.82794799804688,
      "invalid_tail_samples": 23.3,
      "learning_rate": 1.9638454094626833e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 26978.1,
      "query_h_norm_ratio": 0.4154958128929138,
      "relation_loss": 0.0019601422711275516,
      "selected_pairs": 26978.1,
      "step": 27450,
      "student_distance_mean": 22.450417137145998,
      "teacher_distance_mean": 0.8481356918811798,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0019601422711275516,
      "weighted_relation_loss": 0.0019601422711275516
    },
    {
      "action_query_batch_std": 0.3462774395942688,
      "action_query_norm": 40.67125244140625,
      "action_relation_loss": 0.0019686106010340155,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.68755846585594,
      "flow_matching_loss": 0.01517038606107235,
      "grad_norm": 0.10863795876502991,
      "h_stock_valid_norm": 98.14228744506836,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.9485795152838104e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 26769.2,
      "query_h_norm_ratio": 0.41441144049167633,
      "relation_loss": 0.0019686106010340155,
      "selected_pairs": 26769.2,
      "step": 27460,
      "student_distance_mean": 22.801881408691408,
      "teacher_distance_mean": 0.8596313059329986,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0019686106010340155,
      "weighted_relation_loss": 0.0019686106010340155
    },
    {
      "action_query_batch_std": 0.34688309133052825,
      "action_query_norm": 40.6953540802002,
      "action_relation_loss": 0.0019425052683800458,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.69691300280636,
      "flow_matching_loss": 0.014265317004173994,
      "grad_norm": 0.09490471333265305,
      "h_stock_valid_norm": 97.90883941650391,
      "invalid_tail_samples": 21.2,
      "learning_rate": 1.933372008111778e-06,
      "loss": 0.0162,
      "positive_teacher_pairs": 27462.2,
      "query_h_norm_ratio": 0.4156494677066803,
      "relation_loss": 0.0019425052683800458,
      "selected_pairs": 27462.2,
      "step": 27470,
      "student_distance_mean": 22.68049011230469,
      "teacher_distance_mean": 0.8567264437675476,
      "valid_action_samples": 234.8,
      "weighted_auxiliary_loss": 0.0019425052683800458,
      "weighted_relation_loss": 0.0019425052683800458
    },
    {
      "action_query_batch_std": 0.34694035053253175,
      "action_query_norm": 40.679284286499026,
      "action_relation_loss": 0.001946013094857335,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.706267539756784,
      "flow_matching_loss": 0.015832567121833563,
      "grad_norm": 0.12471000850200653,
      "h_stock_valid_norm": 97.9573371887207,
      "invalid_tail_samples": 23.3,
      "learning_rate": 1.918222906425143e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 26966.5,
      "query_h_norm_ratio": 0.4152766793966293,
      "relation_loss": 0.001946013094857335,
      "selected_pairs": 26966.5,
      "step": 27480,
      "student_distance_mean": 22.648840713500977,
      "teacher_distance_mean": 0.8545786619186402,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.001946013094857335,
      "weighted_relation_loss": 0.001946013094857335
    },
    {
      "action_query_batch_std": 0.3462535500526428,
      "action_query_norm": 40.70270538330078,
      "action_relation_loss": 0.0018754968419671058,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.715622076707202,
      "flow_matching_loss": 0.016096751671284437,
      "grad_norm": 0.09775518625974655,
      "h_stock_valid_norm": 98.0369384765625,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.903132228631488e-06,
      "loss": 0.018,
      "positive_teacher_pairs": 26807.4,
      "query_h_norm_ratio": 0.4151851564645767,
      "relation_loss": 0.0018754968419671058,
      "selected_pairs": 26807.4,
      "step": 27490,
      "student_distance_mean": 22.737805557250976,
      "teacher_distance_mean": 0.8582633793354034,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0018754968419671058,
      "weighted_relation_loss": 0.0018754968419671058
    },
    {
      "action_query_batch_std": 0.3422227561473846,
      "action_query_norm": 40.64460639953613,
      "action_relation_loss": 0.001961816963739693,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.724976613657624,
      "flow_matching_loss": 0.01629590392112732,
      "grad_norm": 0.09052807092666626,
      "h_stock_valid_norm": 97.98960800170899,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.8880999930674215e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 26761.1,
      "query_h_norm_ratio": 0.4147872656583786,
      "relation_loss": 0.001961816963739693,
      "selected_pairs": 26761.1,
      "step": 27500,
      "student_distance_mean": 22.371813774108887,
      "teacher_distance_mean": 0.8426270961761475,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.001961816963739693,
      "weighted_relation_loss": 0.001961816963739693
    },
    {
      "action_query_batch_std": 0.33769217133522034,
      "action_query_norm": 40.638393020629884,
      "action_relation_loss": 0.0019196264329366386,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.734331150608046,
      "flow_matching_loss": 0.015495941787958146,
      "grad_norm": 0.11130250990390778,
      "h_stock_valid_norm": 97.92063217163086,
      "invalid_tail_samples": 22.1,
      "learning_rate": 1.8731262179985166e-06,
      "loss": 0.0174,
      "positive_teacher_pairs": 27240.0,
      "query_h_norm_ratio": 0.41501742601394653,
      "relation_loss": 0.0019196264329366386,
      "selected_pairs": 27240.0,
      "step": 27510,
      "student_distance_mean": 22.180055618286133,
      "teacher_distance_mean": 0.8373531103134155,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0019196264329366386,
      "weighted_relation_loss": 0.0019196264329366386
    },
    {
      "action_query_batch_std": 0.3446566641330719,
      "action_query_norm": 40.67860717773438,
      "action_relation_loss": 0.0018804908846504985,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.743685687558465,
      "flow_matching_loss": 0.013745760172605514,
      "grad_norm": 0.09847285598516464,
      "h_stock_valid_norm": 97.80088577270507,
      "invalid_tail_samples": 24.1,
      "learning_rate": 1.8582109216193244e-06,
      "loss": 0.0156,
      "positive_teacher_pairs": 26778.9,
      "query_h_norm_ratio": 0.4159346103668213,
      "relation_loss": 0.0018804908846504985,
      "selected_pairs": 26778.9,
      "step": 27520,
      "student_distance_mean": 22.687015342712403,
      "teacher_distance_mean": 0.8551554203033447,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0018804908846504985,
      "weighted_relation_loss": 0.0018804908846504985
    },
    {
      "action_query_batch_std": 0.3445038259029388,
      "action_query_norm": 40.64909744262695,
      "action_relation_loss": 0.002012085972819477,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.753040224508887,
      "flow_matching_loss": 0.016186445392668246,
      "grad_norm": 0.10289158672094345,
      "h_stock_valid_norm": 97.89090957641602,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.8433541220533368e-06,
      "loss": 0.0182,
      "positive_teacher_pairs": 26755.4,
      "query_h_norm_ratio": 0.41525009274482727,
      "relation_loss": 0.002012085972819477,
      "selected_pairs": 26755.4,
      "step": 27530,
      "student_distance_mean": 22.694115447998048,
      "teacher_distance_mean": 0.8558400094509124,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.002012085972819477,
      "weighted_relation_loss": 0.002012085972819477
    },
    {
      "action_query_batch_std": 0.34833302795886995,
      "action_query_norm": 40.68312759399414,
      "action_relation_loss": 0.0018754060845822095,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.76239476145931,
      "flow_matching_loss": 0.015104466304183006,
      "grad_norm": 0.14078478515148163,
      "h_stock_valid_norm": 97.94605560302735,
      "invalid_tail_samples": 22.9,
      "learning_rate": 1.8285558373529577e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 27057.7,
      "query_h_norm_ratio": 0.41536572575569153,
      "relation_loss": 0.0018754060845822095,
      "selected_pairs": 27057.7,
      "step": 27540,
      "student_distance_mean": 22.81731719970703,
      "teacher_distance_mean": 0.8612569332122803,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0018754060845822095,
      "weighted_relation_loss": 0.0018754060845822095
    },
    {
      "action_query_batch_std": 0.3452854245901108,
      "action_query_norm": 40.669948959350585,
      "action_relation_loss": 0.001882235158700496,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.77174929840973,
      "flow_matching_loss": 0.015081393346190453,
      "grad_norm": 0.08718887716531754,
      "h_stock_valid_norm": 98.00013580322266,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.8138160854995145e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 27013.4,
      "query_h_norm_ratio": 0.4150016218423843,
      "relation_loss": 0.001882235158700496,
      "selected_pairs": 27013.4,
      "step": 27550,
      "student_distance_mean": 22.636791610717772,
      "teacher_distance_mean": 0.853718489408493,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.001882235158700496,
      "weighted_relation_loss": 0.001882235158700496
    },
    {
      "action_query_batch_std": 0.3436843127012253,
      "action_query_norm": 40.652788162231445,
      "action_relation_loss": 0.0019130176748149096,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.78110383536015,
      "flow_matching_loss": 0.01593713564798236,
      "grad_norm": 0.10097099095582962,
      "h_stock_valid_norm": 97.9473274230957,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.7991348844031864e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 26875.3,
      "query_h_norm_ratio": 0.41505221128463743,
      "relation_loss": 0.0019130176748149096,
      "selected_pairs": 26875.3,
      "step": 27560,
      "student_distance_mean": 22.555413055419923,
      "teacher_distance_mean": 0.8498208940029144,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0019130176748149096,
      "weighted_relation_loss": 0.0019130176748149096
    },
    {
      "action_query_batch_std": 0.3475106328725815,
      "action_query_norm": 40.67262306213379,
      "action_relation_loss": 0.0018866243888624012,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.79045837231057,
      "flow_matching_loss": 0.014889500103890896,
      "grad_norm": 0.09366689622402191,
      "h_stock_valid_norm": 97.86790008544922,
      "invalid_tail_samples": 24.9,
      "learning_rate": 1.78451225190302e-06,
      "loss": 0.0168,
      "positive_teacher_pairs": 26593.5,
      "query_h_norm_ratio": 0.4155891716480255,
      "relation_loss": 0.0018866243888624012,
      "selected_pairs": 26593.5,
      "step": 27570,
      "student_distance_mean": 22.906238746643066,
      "teacher_distance_mean": 0.8642394483089447,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0018866243888624012,
      "weighted_relation_loss": 0.0018866243888624012
    },
    {
      "action_query_batch_std": 0.3432707339525223,
      "action_query_norm": 40.629630661010744,
      "action_relation_loss": 0.0018726255744695663,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.799812909260993,
      "flow_matching_loss": 0.014927390310913325,
      "grad_norm": 0.10433848202228546,
      "h_stock_valid_norm": 98.11483612060547,
      "invalid_tail_samples": 26.2,
      "learning_rate": 1.7699482057668914e-06,
      "loss": 0.0168,
      "positive_teacher_pairs": 26303.4,
      "query_h_norm_ratio": 0.4141051650047302,
      "relation_loss": 0.0018726255744695663,
      "selected_pairs": 26303.4,
      "step": 27580,
      "student_distance_mean": 22.574846839904787,
      "teacher_distance_mean": 0.8515571236610413,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.0018726255744695663,
      "weighted_relation_loss": 0.0018726255744695663
    },
    {
      "action_query_batch_std": 0.34633790254592894,
      "action_query_norm": 40.69858245849609,
      "action_relation_loss": 0.0018031562794931233,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.80916744621141,
      "flow_matching_loss": 0.015341474115848542,
      "grad_norm": 0.07897685468196869,
      "h_stock_valid_norm": 98.06564178466797,
      "invalid_tail_samples": 23.2,
      "learning_rate": 1.7554427636914895e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 26986.1,
      "query_h_norm_ratio": 0.41501672863960265,
      "relation_loss": 0.0018031562794931233,
      "selected_pairs": 26986.1,
      "step": 27590,
      "student_distance_mean": 22.87466106414795,
      "teacher_distance_mean": 0.860962551832199,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0018031562794931233,
      "weighted_relation_loss": 0.0018031562794931233
    },
    {
      "action_query_batch_std": 0.3449224293231964,
      "action_query_norm": 40.669497680664065,
      "action_relation_loss": 0.001889489160384983,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.818521983161833,
      "flow_matching_loss": 0.015878608729690314,
      "grad_norm": 0.07303120940923691,
      "h_stock_valid_norm": 98.24813919067383,
      "invalid_tail_samples": 25.3,
      "learning_rate": 1.7409959433022871e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 26508.6,
      "query_h_norm_ratio": 0.4139492392539978,
      "relation_loss": 0.001889489160384983,
      "selected_pairs": 26508.6,
      "step": 27600,
      "student_distance_mean": 22.710091400146485,
      "teacher_distance_mean": 0.8558596909046173,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.001889489160384983,
      "weighted_relation_loss": 0.001889489160384983
    },
    {
      "action_query_batch_std": 0.342643266916275,
      "action_query_norm": 40.64864807128906,
      "action_relation_loss": 0.0019231330719776452,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.827876520112255,
      "flow_matching_loss": 0.015672573633491994,
      "grad_norm": 0.0831264927983284,
      "h_stock_valid_norm": 97.99297103881835,
      "invalid_tail_samples": 22.6,
      "learning_rate": 1.7266077621535481e-06,
      "loss": 0.0176,
      "positive_teacher_pairs": 27127.3,
      "query_h_norm_ratio": 0.4148151695728302,
      "relation_loss": 0.0019231330719776452,
      "selected_pairs": 27127.3,
      "step": 27610,
      "student_distance_mean": 22.490702056884764,
      "teacher_distance_mean": 0.8468204855918884,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.0019231330719776452,
      "weighted_relation_loss": 0.0019231330719776452
    },
    {
      "action_query_batch_std": 0.3456541568040848,
      "action_query_norm": 40.67337188720703,
      "action_relation_loss": 0.001885942614171654,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.837231057062674,
      "flow_matching_loss": 0.016071194689720868,
      "grad_norm": 0.11129511147737503,
      "h_stock_valid_norm": 98.104443359375,
      "invalid_tail_samples": 22.6,
      "learning_rate": 1.7122782377282597e-06,
      "loss": 0.018,
      "positive_teacher_pairs": 27129.8,
      "query_h_norm_ratio": 0.4145975857973099,
      "relation_loss": 0.001885942614171654,
      "selected_pairs": 27129.8,
      "step": 27620,
      "student_distance_mean": 22.683224487304688,
      "teacher_distance_mean": 0.8549609899520874,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.001885942614171654,
      "weighted_relation_loss": 0.001885942614171654
    },
    {
      "action_query_batch_std": 0.3478016197681427,
      "action_query_norm": 40.67577476501465,
      "action_relation_loss": 0.0019243088783696294,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.846585594013096,
      "flow_matching_loss": 0.017037657741457223,
      "grad_norm": 0.09247106313705444,
      "h_stock_valid_norm": 97.9863265991211,
      "invalid_tail_samples": 25.7,
      "learning_rate": 1.6980073874381497e-06,
      "loss": 0.019,
      "positive_teacher_pairs": 26411.4,
      "query_h_norm_ratio": 0.4151207089424133,
      "relation_loss": 0.0019243088783696294,
      "selected_pairs": 26411.4,
      "step": 27630,
      "student_distance_mean": 22.85819911956787,
      "teacher_distance_mean": 0.8621093928813934,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0019243088783696294,
      "weighted_relation_loss": 0.0019243088783696294
    },
    {
      "action_query_batch_std": 0.340828537940979,
      "action_query_norm": 40.62031898498535,
      "action_relation_loss": 0.00195371329318732,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.855940130963518,
      "flow_matching_loss": 0.01564602702856064,
      "grad_norm": 0.08119816333055496,
      "h_stock_valid_norm": 97.90443420410156,
      "invalid_tail_samples": 26.1,
      "learning_rate": 1.6837952286236413e-06,
      "loss": 0.0176,
      "positive_teacher_pairs": 26317.5,
      "query_h_norm_ratio": 0.4148999750614166,
      "relation_loss": 0.00195371329318732,
      "selected_pairs": 26317.5,
      "step": 27640,
      "student_distance_mean": 22.488066101074217,
      "teacher_distance_mean": 0.8490593671798706,
      "valid_action_samples": 229.9,
      "weighted_auxiliary_loss": 0.00195371329318732,
      "weighted_relation_loss": 0.00195371329318732
    },
    {
      "action_query_batch_std": 0.34614368081092833,
      "action_query_norm": 40.66418952941895,
      "action_relation_loss": 0.0018698583473451436,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.86529466791394,
      "flow_matching_loss": 0.014887991640716792,
      "grad_norm": 0.09505595266819,
      "h_stock_valid_norm": 97.90302429199218,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.6696417785538488e-06,
      "loss": 0.0168,
      "positive_teacher_pairs": 27108.6,
      "query_h_norm_ratio": 0.4153537154197693,
      "relation_loss": 0.0018698583473451436,
      "selected_pairs": 27108.6,
      "step": 27650,
      "student_distance_mean": 22.67315101623535,
      "teacher_distance_mean": 0.8557172834873199,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0018698583473451436,
      "weighted_relation_loss": 0.0018698583473451436
    },
    {
      "action_query_batch_std": 0.34385876953601835,
      "action_query_norm": 40.64622764587402,
      "action_relation_loss": 0.002010425354819745,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.874649204864358,
      "flow_matching_loss": 0.017402350064367057,
      "grad_norm": 0.11040088534355164,
      "h_stock_valid_norm": 97.82875595092773,
      "invalid_tail_samples": 24.7,
      "learning_rate": 1.6555470544265538e-06,
      "loss": 0.0194,
      "positive_teacher_pairs": 26638.3,
      "query_h_norm_ratio": 0.415485680103302,
      "relation_loss": 0.002010425354819745,
      "selected_pairs": 26638.3,
      "step": 27660,
      "student_distance_mean": 22.526560592651368,
      "teacher_distance_mean": 0.8495086193084717,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.002010425354819745,
      "weighted_relation_loss": 0.002010425354819745
    },
    {
      "action_query_batch_std": 0.34824413657188413,
      "action_query_norm": 40.68526153564453,
      "action_relation_loss": 0.0019127762876451015,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.88400374181478,
      "flow_matching_loss": 0.015212023444473744,
      "grad_norm": 0.07336866110563278,
      "h_stock_valid_norm": 97.98473052978515,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.6415110733681738e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 26759.9,
      "query_h_norm_ratio": 0.41522409319877623,
      "relation_loss": 0.0019127762876451015,
      "selected_pairs": 26759.9,
      "step": 27670,
      "student_distance_mean": 22.810822868347167,
      "teacher_distance_mean": 0.8586301922798156,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0019127762876451015,
      "weighted_relation_loss": 0.0019127762876451015
    },
    {
      "action_query_batch_std": 0.3453317642211914,
      "action_query_norm": 40.6770133972168,
      "action_relation_loss": 0.0018368169665336608,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.893358278765202,
      "flow_matching_loss": 0.014689847175031901,
      "grad_norm": 0.09461969137191772,
      "h_stock_valid_norm": 98.00641403198242,
      "invalid_tail_samples": 22.9,
      "learning_rate": 1.6275338524337436e-06,
      "loss": 0.0165,
      "positive_teacher_pairs": 27058.3,
      "query_h_norm_ratio": 0.4150456845760345,
      "relation_loss": 0.0018368169665336608,
      "selected_pairs": 27058.3,
      "step": 27680,
      "student_distance_mean": 22.601072311401367,
      "teacher_distance_mean": 0.8515997648239135,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0018368169665336608,
      "weighted_relation_loss": 0.0018368169665336608
    },
    {
      "action_query_batch_std": 0.3472306340932846,
      "action_query_norm": 40.67530632019043,
      "action_relation_loss": 0.001953534584026784,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.90271281571562,
      "flow_matching_loss": 0.016775991301983595,
      "grad_norm": 0.11086966842412949,
      "h_stock_valid_norm": 97.90462265014648,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.6136154086069056e-06,
      "loss": 0.0187,
      "positive_teacher_pairs": 26735.9,
      "query_h_norm_ratio": 0.41546447277069093,
      "relation_loss": 0.001953534584026784,
      "selected_pairs": 26735.9,
      "step": 27690,
      "student_distance_mean": 22.76692295074463,
      "teacher_distance_mean": 0.8599176526069641,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.001953534584026784,
      "weighted_relation_loss": 0.001953534584026784
    },
    {
      "action_query_batch_std": 0.34580897986888887,
      "action_query_norm": 40.67375144958496,
      "action_relation_loss": 0.0019915198208764195,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.912067352666043,
      "flow_matching_loss": 0.015617303270846605,
      "grad_norm": 0.12360355257987976,
      "h_stock_valid_norm": 97.91469497680664,
      "invalid_tail_samples": 25.4,
      "learning_rate": 1.5997557587998868e-06,
      "loss": 0.0176,
      "positive_teacher_pairs": 26480.1,
      "query_h_norm_ratio": 0.4154015302658081,
      "relation_loss": 0.0019915198208764195,
      "selected_pairs": 26480.1,
      "step": 27700,
      "student_distance_mean": 22.80695686340332,
      "teacher_distance_mean": 0.8592385947704315,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0019915198208764195,
      "weighted_relation_loss": 0.0019915198208764195
    },
    {
      "action_query_batch_std": 0.3437846839427948,
      "action_query_norm": 40.647728729248044,
      "action_relation_loss": 0.001996013068128377,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.921421889616465,
      "flow_matching_loss": 0.014668552204966545,
      "grad_norm": 0.08866694569587708,
      "h_stock_valid_norm": 97.94498138427734,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.5859549198534606e-06,
      "loss": 0.0167,
      "positive_teacher_pairs": 27104.8,
      "query_h_norm_ratio": 0.4150070697069168,
      "relation_loss": 0.001996013068128377,
      "selected_pairs": 27104.8,
      "step": 27710,
      "student_distance_mean": 22.510415458679198,
      "teacher_distance_mean": 0.850192004442215,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.001996013068128377,
      "weighted_relation_loss": 0.001996013068128377
    },
    {
      "action_query_batch_std": 0.3486610382795334,
      "action_query_norm": 40.681784057617186,
      "action_relation_loss": 0.0019222061149775983,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.930776426566887,
      "flow_matching_loss": 0.017254369147121907,
      "grad_norm": 0.11378847062587738,
      "h_stock_valid_norm": 97.75437698364257,
      "invalid_tail_samples": 24.9,
      "learning_rate": 1.572212908536963e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 26603.9,
      "query_h_norm_ratio": 0.41616451144218447,
      "relation_loss": 0.0019222061149775983,
      "selected_pairs": 26603.9,
      "step": 27720,
      "student_distance_mean": 22.901157569885253,
      "teacher_distance_mean": 0.8635348558425904,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0019222061149775983,
      "weighted_relation_loss": 0.0019222061149775983
    },
    {
      "action_query_batch_std": 0.3414117842912674,
      "action_query_norm": 40.64189682006836,
      "action_relation_loss": 0.0019644925836473704,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.940130963517305,
      "flow_matching_loss": 0.01721981419250369,
      "grad_norm": 0.1351652294397354,
      "h_stock_valid_norm": 97.90289611816407,
      "invalid_tail_samples": 22.1,
      "learning_rate": 1.5585297415482202e-06,
      "loss": 0.0192,
      "positive_teacher_pairs": 27251.4,
      "query_h_norm_ratio": 0.4151283085346222,
      "relation_loss": 0.0019644925836473704,
      "selected_pairs": 27251.4,
      "step": 27730,
      "student_distance_mean": 22.467040824890137,
      "teacher_distance_mean": 0.8472377181053161,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0019644925836473704,
      "weighted_relation_loss": 0.0019644925836473704
    },
    {
      "action_query_batch_std": 0.3464480131864548,
      "action_query_norm": 40.678554916381835,
      "action_relation_loss": 0.001906657626386732,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.949485500467727,
      "flow_matching_loss": 0.015551782678812742,
      "grad_norm": 0.2208515703678131,
      "h_stock_valid_norm": 97.80889511108398,
      "invalid_tail_samples": 22.5,
      "learning_rate": 1.5449054355135717e-06,
      "loss": 0.0175,
      "positive_teacher_pairs": 27147.6,
      "query_h_norm_ratio": 0.41590021550655365,
      "relation_loss": 0.001906657626386732,
      "selected_pairs": 27147.6,
      "step": 27740,
      "student_distance_mean": 22.687776374816895,
      "teacher_distance_mean": 0.8557334005832672,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.001906657626386732,
      "weighted_relation_loss": 0.001906657626386732
    },
    {
      "action_query_batch_std": 0.3481185257434845,
      "action_query_norm": 40.675279998779295,
      "action_relation_loss": 0.0019142200821079315,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.95884003741815,
      "flow_matching_loss": 0.014639060944318771,
      "grad_norm": 0.08018824458122253,
      "h_stock_valid_norm": 97.91208572387696,
      "invalid_tail_samples": 23.4,
      "learning_rate": 1.5313400069878415e-06,
      "loss": 0.0166,
      "positive_teacher_pairs": 26940.9,
      "query_h_norm_ratio": 0.41542963981628417,
      "relation_loss": 0.0019142200821079315,
      "selected_pairs": 26940.9,
      "step": 27750,
      "student_distance_mean": 22.872536277770998,
      "teacher_distance_mean": 0.8624405026435852,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0019142200821079315,
      "weighted_relation_loss": 0.0019142200821079315
    },
    {
      "action_query_batch_std": 0.3465215742588043,
      "action_query_norm": 40.666434097290036,
      "action_relation_loss": 0.001957190129905939,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.968194574368567,
      "flow_matching_loss": 0.015247601363807917,
      "grad_norm": 0.10719849914312363,
      "h_stock_valid_norm": 97.96526794433593,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.5178334724542887e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 26800.9,
      "query_h_norm_ratio": 0.41511242687702177,
      "relation_loss": 0.001957190129905939,
      "selected_pairs": 26800.9,
      "step": 27760,
      "student_distance_mean": 22.832557487487794,
      "teacher_distance_mean": 0.862733519077301,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.001957190129905939,
      "weighted_relation_loss": 0.001957190129905939
    },
    {
      "action_query_batch_std": 0.3466698616743088,
      "action_query_norm": 40.67626800537109,
      "action_relation_loss": 0.0019636669545434414,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.97754911131899,
      "flow_matching_loss": 0.015066086500883102,
      "grad_norm": 0.06839372217655182,
      "h_stock_valid_norm": 97.9912010192871,
      "invalid_tail_samples": 25.3,
      "learning_rate": 1.504385848324641e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 26500.4,
      "query_h_norm_ratio": 0.4151020646095276,
      "relation_loss": 0.0019636669545434414,
      "selected_pairs": 26500.4,
      "step": 27770,
      "student_distance_mean": 22.756070137023926,
      "teacher_distance_mean": 0.8582752406597137,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0019636669545434414,
      "weighted_relation_loss": 0.0019636669545434414
    },
    {
      "action_query_batch_std": 0.3430729418992996,
      "action_query_norm": 40.62811470031738,
      "action_relation_loss": 0.0019444279256276785,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.98690364826941,
      "flow_matching_loss": 0.015836095064878465,
      "grad_norm": 0.08578963577747345,
      "h_stock_valid_norm": 97.9369499206543,
      "invalid_tail_samples": 25.8,
      "learning_rate": 1.490997150939033e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 26393.8,
      "query_h_norm_ratio": 0.414840567111969,
      "relation_loss": 0.0019444279256276785,
      "selected_pairs": 26393.8,
      "step": 27780,
      "student_distance_mean": 22.558677101135252,
      "teacher_distance_mean": 0.8496857762336731,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0019444279256276785,
      "weighted_relation_loss": 0.0019444279256276785
    },
    {
      "action_query_batch_std": 0.34901848137378694,
      "action_query_norm": 40.685404205322264,
      "action_relation_loss": 0.0018562147743068635,
      "auxiliary_loss_weight": 1.0,
      "epoch": 25.996258185219833,
      "flow_matching_loss": 0.01640510745346546,
      "grad_norm": 0.0955105572938919,
      "h_stock_valid_norm": 98.10201644897461,
      "invalid_tail_samples": 26.5,
      "learning_rate": 1.4776673965659793e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 26226.8,
      "query_h_norm_ratio": 0.41473206877708435,
      "relation_loss": 0.0018562147743068635,
      "selected_pairs": 26226.8,
      "step": 27790,
      "student_distance_mean": 22.801817893981934,
      "teacher_distance_mean": 0.8611262500286102,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.0018562147743068635,
      "weighted_relation_loss": 0.0018562147743068635
    },
    {
      "action_query_batch_std": 0.34620744585990904,
      "action_query_norm": 40.67781295776367,
      "action_relation_loss": 0.001849536201916635,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.00561272217025,
      "flow_matching_loss": 0.015609824936836958,
      "grad_norm": 0.09482758492231369,
      "h_stock_valid_norm": 98.04662857055663,
      "invalid_tail_samples": 25.6,
      "learning_rate": 1.4643966014023957e-06,
      "loss": 0.0175,
      "positive_teacher_pairs": 23525.5,
      "query_h_norm_ratio": 0.41488408148288725,
      "relation_loss": 0.001849536201916635,
      "selected_pairs": 23525.5,
      "step": 27800,
      "student_distance_mean": 22.94160270690918,
      "teacher_distance_mean": 0.8657264351844788,
      "valid_action_samples": 210.5,
      "weighted_auxiliary_loss": 0.001849536201916635,
      "weighted_relation_loss": 0.001849536201916635
    },
    {
      "action_query_batch_std": 0.3454247921705246,
      "action_query_norm": 40.65677108764648,
      "action_relation_loss": 0.0018485679640434683,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.014967259120674,
      "flow_matching_loss": 0.01593731278553605,
      "grad_norm": 0.10647010058164597,
      "h_stock_valid_norm": 98.17292785644531,
      "invalid_tail_samples": 25.6,
      "learning_rate": 1.4511847815735503e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 26433.4,
      "query_h_norm_ratio": 0.41413558423519137,
      "relation_loss": 0.0018485679640434683,
      "selected_pairs": 26433.4,
      "step": 27810,
      "student_distance_mean": 22.732944297790528,
      "teacher_distance_mean": 0.8567986607551574,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0018485679640434683,
      "weighted_relation_loss": 0.0018485679640434683
    },
    {
      "action_query_batch_std": 0.3464484065771103,
      "action_query_norm": 40.679765701293945,
      "action_relation_loss": 0.0018720796797424555,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.024321796071096,
      "flow_matching_loss": 0.014659443311393261,
      "grad_norm": 0.09368552267551422,
      "h_stock_valid_norm": 97.92438659667968,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.438031953133051e-06,
      "loss": 0.0165,
      "positive_teacher_pairs": 26899.3,
      "query_h_norm_ratio": 0.41542198956012727,
      "relation_loss": 0.0018720796797424555,
      "selected_pairs": 26899.3,
      "step": 27820,
      "student_distance_mean": 22.72631130218506,
      "teacher_distance_mean": 0.8582377970218659,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0018720796797424555,
      "weighted_relation_loss": 0.0018720796797424555
    },
    {
      "action_query_batch_std": 0.34544448256492616,
      "action_query_norm": 40.66093521118164,
      "action_relation_loss": 0.0019824937451630832,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.033676333021514,
      "flow_matching_loss": 0.015018806234002114,
      "grad_norm": 0.20346751809120178,
      "h_stock_valid_norm": 98.00668182373047,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.4249381320628197e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 26736.4,
      "query_h_norm_ratio": 0.4148836970329285,
      "relation_loss": 0.0019824937451630832,
      "selected_pairs": 26736.4,
      "step": 27830,
      "student_distance_mean": 22.600723648071288,
      "teacher_distance_mean": 0.8535449981689454,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0019824937451630832,
      "weighted_relation_loss": 0.0019824937451630832
    },
    {
      "action_query_batch_std": 0.3456861227750778,
      "action_query_norm": 40.664506149291995,
      "action_relation_loss": 0.0018798833247274161,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.043030869971936,
      "flow_matching_loss": 0.01580160791054368,
      "grad_norm": 0.07708290964365005,
      "h_stock_valid_norm": 98.23995590209961,
      "invalid_tail_samples": 25.8,
      "learning_rate": 1.4119033342730902e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 26393.2,
      "query_h_norm_ratio": 0.41393386423587797,
      "relation_loss": 0.0018798833247274161,
      "selected_pairs": 26393.2,
      "step": 27840,
      "student_distance_mean": 22.648949241638185,
      "teacher_distance_mean": 0.8549577295780182,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0018798833247274161,
      "weighted_relation_loss": 0.0018798833247274161
    },
    {
      "action_query_batch_std": 0.34645108282566073,
      "action_query_norm": 40.67871627807617,
      "action_relation_loss": 0.0019368792651221155,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.052385406922358,
      "flow_matching_loss": 0.015816517733037473,
      "grad_norm": 0.07202862203121185,
      "h_stock_valid_norm": 97.99668350219727,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.3989275756023656e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 26730.6,
      "query_h_norm_ratio": 0.41510542631149294,
      "relation_loss": 0.0019368792651221155,
      "selected_pairs": 26730.6,
      "step": 27850,
      "student_distance_mean": 22.713900756835937,
      "teacher_distance_mean": 0.8573049783706665,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0019368792651221155,
      "weighted_relation_loss": 0.0019368792651221155
    },
    {
      "action_query_batch_std": 0.3438083529472351,
      "action_query_norm": 40.65938758850098,
      "action_relation_loss": 0.0019182341638952494,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.061739943872777,
      "flow_matching_loss": 0.015150076244026423,
      "grad_norm": 0.09640977531671524,
      "h_stock_valid_norm": 97.84561462402344,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.386010871817417e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 26758.3,
      "query_h_norm_ratio": 0.4155478686094284,
      "relation_loss": 0.0019182341638952494,
      "selected_pairs": 26758.3,
      "step": 27860,
      "student_distance_mean": 22.569282722473144,
      "teacher_distance_mean": 0.8505372166633606,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0019182341638952494,
      "weighted_relation_loss": 0.0019182341638952494
    },
    {
      "action_query_batch_std": 0.34071981608867646,
      "action_query_norm": 40.6286865234375,
      "action_relation_loss": 0.0019337303470820188,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.0710944808232,
      "flow_matching_loss": 0.014592022262513637,
      "grad_norm": 0.10052046179771423,
      "h_stock_valid_norm": 98.04203948974609,
      "invalid_tail_samples": 21.3,
      "learning_rate": 1.3731532386132616e-06,
      "loss": 0.0165,
      "positive_teacher_pairs": 27431.0,
      "query_h_norm_ratio": 0.41440220177173615,
      "relation_loss": 0.0019337303470820188,
      "selected_pairs": 27431.0,
      "step": 27870,
      "student_distance_mean": 22.37364444732666,
      "teacher_distance_mean": 0.843695592880249,
      "valid_action_samples": 234.7,
      "weighted_auxiliary_loss": 0.0019337303470820188,
      "weighted_relation_loss": 0.0019337303470820188
    },
    {
      "action_query_batch_std": 0.3476030617952347,
      "action_query_norm": 40.683447265625,
      "action_relation_loss": 0.0019015777274034918,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.08044901777362,
      "flow_matching_loss": 0.01716818017885089,
      "grad_norm": 0.08640792220830917,
      "h_stock_valid_norm": 97.99008255004883,
      "invalid_tail_samples": 24.9,
      "learning_rate": 1.360354691613136e-06,
      "loss": 0.0191,
      "positive_teacher_pairs": 26596.1,
      "query_h_norm_ratio": 0.4151805579662323,
      "relation_loss": 0.0019015777274034918,
      "selected_pairs": 26596.1,
      "step": 27880,
      "student_distance_mean": 22.91471576690674,
      "teacher_distance_mean": 0.8644738376140595,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0019015777274034918,
      "weighted_relation_loss": 0.0019015777274034918
    },
    {
      "action_query_batch_std": 0.34459719359874724,
      "action_query_norm": 40.66192817687988,
      "action_relation_loss": 0.0018872283515520393,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.089803554724043,
      "flow_matching_loss": 0.014593594335019588,
      "grad_norm": 0.16912955045700073,
      "h_stock_valid_norm": 98.01200256347656,
      "invalid_tail_samples": 20.4,
      "learning_rate": 1.3476152463684776e-06,
      "loss": 0.0165,
      "positive_teacher_pairs": 27646.8,
      "query_h_norm_ratio": 0.4148697227239609,
      "relation_loss": 0.0018872283515520393,
      "selected_pairs": 27646.8,
      "step": 27890,
      "student_distance_mean": 22.607382202148436,
      "teacher_distance_mean": 0.851783037185669,
      "valid_action_samples": 235.6,
      "weighted_auxiliary_loss": 0.0018872283515520393,
      "weighted_relation_loss": 0.0018872283515520393
    },
    {
      "action_query_batch_std": 0.3426231533288956,
      "action_query_norm": 40.6297607421875,
      "action_relation_loss": 0.001989100081846118,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.09915809167446,
      "flow_matching_loss": 0.016321318317204715,
      "grad_norm": 0.11035033315420151,
      "h_stock_valid_norm": 97.90178680419922,
      "invalid_tail_samples": 27.9,
      "learning_rate": 1.3349349183589155e-06,
      "loss": 0.0183,
      "positive_teacher_pairs": 25914.5,
      "query_h_norm_ratio": 0.4150074452161789,
      "relation_loss": 0.001989100081846118,
      "selected_pairs": 25914.5,
      "step": 27900,
      "student_distance_mean": 22.51450996398926,
      "teacher_distance_mean": 0.8486063539981842,
      "valid_action_samples": 228.1,
      "weighted_auxiliary_loss": 0.001989100081846118,
      "weighted_relation_loss": 0.001989100081846118
    },
    {
      "action_query_batch_std": 0.34559584558010104,
      "action_query_norm": 40.68274993896485,
      "action_relation_loss": 0.0018693870515562594,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.108512628624883,
      "flow_matching_loss": 0.01618640162050724,
      "grad_norm": 0.07916800677776337,
      "h_stock_valid_norm": 97.92732696533203,
      "invalid_tail_samples": 23.8,
      "learning_rate": 1.3223137229922356e-06,
      "loss": 0.0181,
      "positive_teacher_pairs": 26860.2,
      "query_h_norm_ratio": 0.4154412358999252,
      "relation_loss": 0.0018693870515562594,
      "selected_pairs": 26860.2,
      "step": 27910,
      "student_distance_mean": 22.69701156616211,
      "teacher_distance_mean": 0.8557035803794861,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0018693870515562594,
      "weighted_relation_loss": 0.0018693870515562594
    },
    {
      "action_query_batch_std": 0.3448881387710571,
      "action_query_norm": 40.6649829864502,
      "action_relation_loss": 0.0019453573971986771,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.117867165575305,
      "flow_matching_loss": 0.01583222234621644,
      "grad_norm": 0.10679709166288376,
      "h_stock_valid_norm": 97.98823318481445,
      "invalid_tail_samples": 22.2,
      "learning_rate": 1.3097516756043981e-06,
      "loss": 0.0178,
      "positive_teacher_pairs": 27218.2,
      "query_h_norm_ratio": 0.4149998098611832,
      "relation_loss": 0.0019453573971986771,
      "selected_pairs": 27218.2,
      "step": 27920,
      "student_distance_mean": 22.688258552551268,
      "teacher_distance_mean": 0.8559043824672699,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0019453573971986771,
      "weighted_relation_loss": 0.0019453573971986771
    },
    {
      "action_query_batch_std": 0.3458287715911865,
      "action_query_norm": 40.6914867401123,
      "action_relation_loss": 0.0018160188919864595,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.127221702525723,
      "flow_matching_loss": 0.015180032886564731,
      "grad_norm": 0.12782080471515656,
      "h_stock_valid_norm": 97.88893966674804,
      "invalid_tail_samples": 22.3,
      "learning_rate": 1.29724879145946e-06,
      "loss": 0.017,
      "positive_teacher_pairs": 27204.3,
      "query_h_norm_ratio": 0.41569280326366426,
      "relation_loss": 0.0018160188919864595,
      "selected_pairs": 27204.3,
      "step": 27930,
      "student_distance_mean": 22.747132301330566,
      "teacher_distance_mean": 0.8595836937427521,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0018160188919864595,
      "weighted_relation_loss": 0.0018160188919864595
    },
    {
      "action_query_batch_std": 0.34253057837486267,
      "action_query_norm": 40.6515869140625,
      "action_relation_loss": 0.0018357882392592729,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.136576239476145,
      "flow_matching_loss": 0.014950050693005323,
      "grad_norm": 0.11143189668655396,
      "h_stock_valid_norm": 97.80337982177734,
      "invalid_tail_samples": 24.8,
      "learning_rate": 1.284805085749613e-06,
      "loss": 0.0168,
      "positive_teacher_pairs": 26619.8,
      "query_h_norm_ratio": 0.4156494617462158,
      "relation_loss": 0.0018357882392592729,
      "selected_pairs": 26619.8,
      "step": 27940,
      "student_distance_mean": 22.5445499420166,
      "teacher_distance_mean": 0.850050288438797,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0018357882392592729,
      "weighted_relation_loss": 0.0018357882392592729
    },
    {
      "action_query_batch_std": 0.3457911193370819,
      "action_query_norm": 40.67887878417969,
      "action_relation_loss": 0.0020114769693464042,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.145930776426567,
      "flow_matching_loss": 0.015277356933802366,
      "grad_norm": 0.08312679082155228,
      "h_stock_valid_norm": 98.01051712036133,
      "invalid_tail_samples": 23.5,
      "learning_rate": 1.2724205735951288e-06,
      "loss": 0.0173,
      "positive_teacher_pairs": 26919.6,
      "query_h_norm_ratio": 0.4150503247976303,
      "relation_loss": 0.0020114769693464042,
      "selected_pairs": 26919.6,
      "step": 27950,
      "student_distance_mean": 22.620624923706053,
      "teacher_distance_mean": 0.8518655359745025,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0020114769693464042,
      "weighted_relation_loss": 0.0020114769693464042
    },
    {
      "action_query_batch_std": 0.3430414766073227,
      "action_query_norm": 40.66130447387695,
      "action_relation_loss": 0.0019059222540818155,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.15528531337699,
      "flow_matching_loss": 0.015328416600823402,
      "grad_norm": 0.07914888113737106,
      "h_stock_valid_norm": 97.97368087768555,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.2600952700443591e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 26902.0,
      "query_h_norm_ratio": 0.41502542793750763,
      "relation_loss": 0.0019059222540818155,
      "selected_pairs": 26902.0,
      "step": 27960,
      "student_distance_mean": 22.471041107177733,
      "teacher_distance_mean": 0.8472964465618134,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0019059222540818155,
      "weighted_relation_loss": 0.0019059222540818155
    },
    {
      "action_query_batch_std": 0.348100820183754,
      "action_query_norm": 40.695603942871095,
      "action_relation_loss": 0.0018773377407342196,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.164639850327408,
      "flow_matching_loss": 0.01463415352627635,
      "grad_norm": 0.12171586602926254,
      "h_stock_valid_norm": 98.02064971923828,
      "invalid_tail_samples": 23.8,
      "learning_rate": 1.247829190073707e-06,
      "loss": 0.0165,
      "positive_teacher_pairs": 26850.0,
      "query_h_norm_ratio": 0.41517473459243776,
      "relation_loss": 0.0018773377407342196,
      "selected_pairs": 26850.0,
      "step": 27970,
      "student_distance_mean": 22.82338104248047,
      "teacher_distance_mean": 0.8597489595413208,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0018773377407342196,
      "weighted_relation_loss": 0.0018773377407342196
    },
    {
      "action_query_batch_std": 0.3459133207798004,
      "action_query_norm": 40.688786315917966,
      "action_relation_loss": 0.001886849629227072,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.17399438727783,
      "flow_matching_loss": 0.014415446016937495,
      "grad_norm": 0.1904418021440506,
      "h_stock_valid_norm": 97.99309768676758,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.2356223485876173e-06,
      "loss": 0.0163,
      "positive_teacher_pairs": 26825.6,
      "query_h_norm_ratio": 0.415221232175827,
      "relation_loss": 0.001886849629227072,
      "selected_pairs": 26825.6,
      "step": 27980,
      "student_distance_mean": 22.810619354248047,
      "teacher_distance_mean": 0.8609384298324585,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.001886849629227072,
      "weighted_relation_loss": 0.001886849629227072
    },
    {
      "action_query_batch_std": 0.34710111618041994,
      "action_query_norm": 40.66386070251465,
      "action_relation_loss": 0.0019073382718488574,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.18334892422825,
      "flow_matching_loss": 0.014358511101454497,
      "grad_norm": 0.12334979325532913,
      "h_stock_valid_norm": 98.09782180786132,
      "invalid_tail_samples": 23.5,
      "learning_rate": 1.2234747604185526e-06,
      "loss": 0.0163,
      "positive_teacher_pairs": 26924.3,
      "query_h_norm_ratio": 0.4145275682210922,
      "relation_loss": 0.0019073382718488574,
      "selected_pairs": 26924.3,
      "step": 27990,
      "student_distance_mean": 22.778536987304687,
      "teacher_distance_mean": 0.8593181252479554,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0019073382718488574,
      "weighted_relation_loss": 0.0019073382718488574
    },
    {
      "action_query_batch_std": 0.34648674428462983,
      "action_query_norm": 40.67065505981445,
      "action_relation_loss": 0.0019668272230774163,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.19270346117867,
      "flow_matching_loss": 0.015349578112363815,
      "grad_norm": 0.09338141232728958,
      "h_stock_valid_norm": 98.12902069091797,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.2113864403269836e-06,
      "loss": 0.0173,
      "positive_teacher_pairs": 26739.9,
      "query_h_norm_ratio": 0.4144624561071396,
      "relation_loss": 0.0019668272230774163,
      "selected_pairs": 26739.9,
      "step": 28000,
      "student_distance_mean": 22.755931282043456,
      "teacher_distance_mean": 0.8585870504379273,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0019668272230774163,
      "weighted_relation_loss": 0.0019668272230774163
    },
    {
      "action_query_batch_std": 0.34317547976970675,
      "action_query_norm": 40.65479431152344,
      "action_relation_loss": 0.0018869090592488647,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.202057998129092,
      "flow_matching_loss": 0.0167318943887949,
      "grad_norm": 0.15203148126602173,
      "h_stock_valid_norm": 97.99799957275391,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.1993574030013555e-06,
      "loss": 0.0186,
      "positive_teacher_pairs": 26754.3,
      "query_h_norm_ratio": 0.4148542761802673,
      "relation_loss": 0.0018869090592488647,
      "selected_pairs": 26754.3,
      "step": 28010,
      "student_distance_mean": 22.391164016723632,
      "teacher_distance_mean": 0.8444019854068756,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0018869090592488647,
      "weighted_relation_loss": 0.0018869090592488647
    },
    {
      "action_query_batch_std": 0.34333547949790955,
      "action_query_norm": 40.666264343261716,
      "action_relation_loss": 0.002008301403839141,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.211412535079514,
      "flow_matching_loss": 0.015650799963623285,
      "grad_norm": 0.08485472947359085,
      "h_stock_valid_norm": 97.94643020629883,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.1873876630580816e-06,
      "loss": 0.0177,
      "positive_teacher_pairs": 26880.9,
      "query_h_norm_ratio": 0.4151911586523056,
      "relation_loss": 0.002008301403839141,
      "selected_pairs": 26880.9,
      "step": 28020,
      "student_distance_mean": 22.570996856689455,
      "teacher_distance_mean": 0.8527510106563568,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.002008301403839141,
      "weighted_relation_loss": 0.002008301403839141
    },
    {
      "action_query_batch_std": 0.3455253005027771,
      "action_query_norm": 40.66507110595703,
      "action_relation_loss": 0.001960700913332403,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.220767072029936,
      "flow_matching_loss": 0.01616257494315505,
      "grad_norm": 0.09744007140398026,
      "h_stock_valid_norm": 97.96774291992188,
      "invalid_tail_samples": 20.8,
      "learning_rate": 1.1754772350415278e-06,
      "loss": 0.0181,
      "positive_teacher_pairs": 27549.8,
      "query_h_norm_ratio": 0.4150902032852173,
      "relation_loss": 0.001960700913332403,
      "selected_pairs": 27549.8,
      "step": 28030,
      "student_distance_mean": 22.592205429077147,
      "teacher_distance_mean": 0.853096854686737,
      "valid_action_samples": 235.2,
      "weighted_auxiliary_loss": 0.001960700913332403,
      "weighted_relation_loss": 0.001960700913332403
    },
    {
      "action_query_batch_std": 0.3454154312610626,
      "action_query_norm": 40.63577423095703,
      "action_relation_loss": 0.0018798045697622002,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.230121608980355,
      "flow_matching_loss": 0.016091903671622276,
      "grad_norm": 0.09282291680574417,
      "h_stock_valid_norm": 97.99980087280274,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.163626133423984e-06,
      "loss": 0.018,
      "positive_teacher_pairs": 26825.9,
      "query_h_norm_ratio": 0.41465300917625425,
      "relation_loss": 0.0018798045697622002,
      "selected_pairs": 26825.9,
      "step": 28040,
      "student_distance_mean": 22.546019554138184,
      "teacher_distance_mean": 0.8503889918327332,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0018798045697622002,
      "weighted_relation_loss": 0.0018798045697622002
    },
    {
      "action_query_batch_std": 0.3466485351324081,
      "action_query_norm": 40.67114143371582,
      "action_relation_loss": 0.0019355240394361318,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.239476145930777,
      "flow_matching_loss": 0.015176859125494957,
      "grad_norm": 0.10109438002109528,
      "h_stock_valid_norm": 97.94048156738282,
      "invalid_tail_samples": 25.3,
      "learning_rate": 1.1518343726056591e-06,
      "loss": 0.0171,
      "positive_teacher_pairs": 26512.2,
      "query_h_norm_ratio": 0.4152653872966766,
      "relation_loss": 0.0019355240394361318,
      "selected_pairs": 26512.2,
      "step": 28050,
      "student_distance_mean": 22.805475425720214,
      "teacher_distance_mean": 0.8615305066108704,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0019355240394361318,
      "weighted_relation_loss": 0.0019355240394361318
    },
    {
      "action_query_batch_std": 0.3463587939739227,
      "action_query_norm": 40.676153564453124,
      "action_relation_loss": 0.0019134448026306926,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.2488306828812,
      "flow_matching_loss": 0.015304749179631472,
      "grad_norm": 0.0786181092262268,
      "h_stock_valid_norm": 97.9233627319336,
      "invalid_tail_samples": 21.6,
      "learning_rate": 1.1401019669146474e-06,
      "loss": 0.0172,
      "positive_teacher_pairs": 27363.8,
      "query_h_norm_ratio": 0.41538981199264524,
      "relation_loss": 0.0019134448026306926,
      "selected_pairs": 27363.8,
      "step": 28060,
      "student_distance_mean": 22.64251537322998,
      "teacher_distance_mean": 0.8541577756404877,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.0019134448026306926,
      "weighted_relation_loss": 0.0019134448026306926
    },
    {
      "action_query_batch_std": 0.3484364151954651,
      "action_query_norm": 40.70608406066894,
      "action_relation_loss": 0.0019658653298392894,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.258185219831617,
      "flow_matching_loss": 0.015522914100438356,
      "grad_norm": 0.13341854512691498,
      "h_stock_valid_norm": 97.93275680541993,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.128428930606934e-06,
      "loss": 0.0175,
      "positive_teacher_pairs": 27109.1,
      "query_h_norm_ratio": 0.41565498411655427,
      "relation_loss": 0.0019658653298392894,
      "selected_pairs": 27109.1,
      "step": 28070,
      "student_distance_mean": 22.932575607299803,
      "teacher_distance_mean": 0.8668529093265533,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0019658653298392894,
      "weighted_relation_loss": 0.0019658653298392894
    },
    {
      "action_query_batch_std": 0.3457250863313675,
      "action_query_norm": 40.672982406616214,
      "action_relation_loss": 0.0018796037533320487,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.26753975678204,
      "flow_matching_loss": 0.014702590648084879,
      "grad_norm": 0.1229785606265068,
      "h_stock_valid_norm": 97.8920783996582,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.1168152778663621e-06,
      "loss": 0.0166,
      "positive_teacher_pairs": 27081.4,
      "query_h_norm_ratio": 0.4154894948005676,
      "relation_loss": 0.0018796037533320487,
      "selected_pairs": 27081.4,
      "step": 28080,
      "student_distance_mean": 22.731570625305174,
      "teacher_distance_mean": 0.8581295669078827,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0018796037533320487,
      "weighted_relation_loss": 0.0018796037533320487
    },
    {
      "action_query_batch_std": 0.34601543843746185,
      "action_query_norm": 40.64436416625976,
      "action_relation_loss": 0.001868439989630133,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.27689429373246,
      "flow_matching_loss": 0.015999548230320217,
      "grad_norm": 0.0954301580786705,
      "h_stock_valid_norm": 97.9897590637207,
      "invalid_tail_samples": 21.1,
      "learning_rate": 1.10526102280461e-06,
      "loss": 0.0179,
      "positive_teacher_pairs": 27477.3,
      "query_h_norm_ratio": 0.41478410363197327,
      "relation_loss": 0.001868439989630133,
      "selected_pairs": 27477.3,
      "step": 28090,
      "student_distance_mean": 22.68352222442627,
      "teacher_distance_mean": 0.8571509778499603,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.001868439989630133,
      "weighted_relation_loss": 0.001868439989630133
    },
    {
      "action_query_batch_std": 0.34525828063488007,
      "action_query_norm": 40.68850936889648,
      "action_relation_loss": 0.001878742256667465,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.28624883068288,
      "flow_matching_loss": 0.014188459981232882,
      "grad_norm": 0.08394384384155273,
      "h_stock_valid_norm": 98.18284378051757,
      "invalid_tail_samples": 22.9,
      "learning_rate": 1.0937661794611864e-06,
      "loss": 0.0161,
      "positive_teacher_pairs": 27060.9,
      "query_h_norm_ratio": 0.4144157409667969,
      "relation_loss": 0.001878742256667465,
      "selected_pairs": 27060.9,
      "step": 28100,
      "student_distance_mean": 22.70667095184326,
      "teacher_distance_mean": 0.8584912240505218,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.001878742256667465,
      "weighted_relation_loss": 0.001878742256667465
    },
    {
      "action_query_batch_std": 0.34206669926643374,
      "action_query_norm": 40.6250301361084,
      "action_relation_loss": 0.0018700283719226718,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.2956033676333,
      "flow_matching_loss": 0.014752068836241961,
      "grad_norm": 0.11979079246520996,
      "h_stock_valid_norm": 97.97719497680664,
      "invalid_tail_samples": 26.2,
      "learning_rate": 1.082330761803413e-06,
      "loss": 0.0166,
      "positive_teacher_pairs": 26294.4,
      "query_h_norm_ratio": 0.4146383047103882,
      "relation_loss": 0.0018700283719226718,
      "selected_pairs": 26294.4,
      "step": 28110,
      "student_distance_mean": 22.483547592163085,
      "teacher_distance_mean": 0.8486299812793732,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.0018700283719226718,
      "weighted_relation_loss": 0.0018700283719226718
    },
    {
      "action_query_batch_std": 0.3460875809192657,
      "action_query_norm": 40.69765281677246,
      "action_relation_loss": 0.001799963926896453,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.304957904583723,
      "flow_matching_loss": 0.014611440245062113,
      "grad_norm": 0.11160798370838165,
      "h_stock_valid_norm": 97.99574813842773,
      "invalid_tail_samples": 23.3,
      "learning_rate": 1.0709547837263966e-06,
      "loss": 0.0164,
      "positive_teacher_pairs": 26966.2,
      "query_h_norm_ratio": 0.4153028309345245,
      "relation_loss": 0.001799963926896453,
      "selected_pairs": 26966.2,
      "step": 28120,
      "student_distance_mean": 22.705718994140625,
      "teacher_distance_mean": 0.8568882942199707,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.001799963926896453,
      "weighted_relation_loss": 0.001799963926896453
    },
    {
      "action_query_batch_std": 0.3458553373813629,
      "action_query_norm": 40.66542472839355,
      "action_relation_loss": 0.0019044291693717241,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.314312441534145,
      "flow_matching_loss": 0.014414284378290176,
      "grad_norm": 0.12132956832647324,
      "h_stock_valid_norm": 98.01465148925782,
      "invalid_tail_samples": 23.0,
      "learning_rate": 1.059638259053025e-06,
      "loss": 0.0163,
      "positive_teacher_pairs": 27038.1,
      "query_h_norm_ratio": 0.4148944795131683,
      "relation_loss": 0.0019044291693717241,
      "selected_pairs": 27038.1,
      "step": 28130,
      "student_distance_mean": 22.68190860748291,
      "teacher_distance_mean": 0.8557204842567444,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0019044291693717241,
      "weighted_relation_loss": 0.0019044291693717241
    },
    {
      "action_query_batch_std": 0.34884674847126007,
      "action_query_norm": 40.69457893371582,
      "action_relation_loss": 0.0019203124102205038,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.323666978484564,
      "flow_matching_loss": 0.01460331454873085,
      "grad_norm": 0.10926099121570587,
      "h_stock_valid_norm": 97.8804702758789,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.0483812015339434e-06,
      "loss": 0.0165,
      "positive_teacher_pairs": 26833.2,
      "query_h_norm_ratio": 0.4157613039016724,
      "relation_loss": 0.0019203124102205038,
      "selected_pairs": 26833.2,
      "step": 28140,
      "student_distance_mean": 22.797181701660158,
      "teacher_distance_mean": 0.8598420321941376,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0019203124102205038,
      "weighted_relation_loss": 0.0019203124102205038
    },
    {
      "action_query_batch_std": 0.3425750911235809,
      "action_query_norm": 40.651860046386716,
      "action_relation_loss": 0.0019725770107470453,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.333021515434986,
      "flow_matching_loss": 0.01407088441774249,
      "grad_norm": 0.0919799879193306,
      "h_stock_valid_norm": 97.94916000366212,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.037183624847543e-06,
      "loss": 0.016,
      "positive_teacher_pairs": 27104.8,
      "query_h_norm_ratio": 0.41503130197525023,
      "relation_loss": 0.0019725770107470453,
      "selected_pairs": 27104.8,
      "step": 28150,
      "student_distance_mean": 22.49333076477051,
      "teacher_distance_mean": 0.8492944478988648,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0019725770107470453,
      "weighted_relation_loss": 0.0019725770107470453
    },
    {
      "action_query_batch_std": 0.3457699567079544,
      "action_query_norm": 40.67704315185547,
      "action_relation_loss": 0.0018825257080607115,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.342376052385408,
      "flow_matching_loss": 0.014962442219257355,
      "grad_norm": 0.09770698845386505,
      "h_stock_valid_norm": 97.98836288452148,
      "invalid_tail_samples": 23.4,
      "learning_rate": 1.0260455425999238e-06,
      "loss": 0.0168,
      "positive_teacher_pairs": 26944.4,
      "query_h_norm_ratio": 0.41512311697006227,
      "relation_loss": 0.0018825257080607115,
      "selected_pairs": 26944.4,
      "step": 28160,
      "student_distance_mean": 22.68253536224365,
      "teacher_distance_mean": 0.8561046659946442,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0018825257080607115,
      "weighted_relation_loss": 0.0018825257080607115
    },
    {
      "action_query_batch_std": 0.34395570755004884,
      "action_query_norm": 40.62764625549316,
      "action_relation_loss": 0.0018901347531937063,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.351730589335826,
      "flow_matching_loss": 0.014816903322935105,
      "grad_norm": 0.10876839607954025,
      "h_stock_valid_norm": 98.03649673461913,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.0149669683249208e-06,
      "loss": 0.0167,
      "positive_teacher_pairs": 27103.4,
      "query_h_norm_ratio": 0.4144178181886673,
      "relation_loss": 0.0018901347531937063,
      "selected_pairs": 27103.4,
      "step": 28170,
      "student_distance_mean": 22.53576717376709,
      "teacher_distance_mean": 0.8499638974666596,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0018901347531937063,
      "weighted_relation_loss": 0.0018901347531937063
    },
    {
      "action_query_batch_std": 0.3430181711912155,
      "action_query_norm": 40.63842124938965,
      "action_relation_loss": 0.001890819298569113,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.36108512628625,
      "flow_matching_loss": 0.014284553285688163,
      "grad_norm": 0.10023196786642075,
      "h_stock_valid_norm": 98.13195495605468,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.003947915484027e-06,
      "loss": 0.0162,
      "positive_teacher_pairs": 26802.4,
      "query_h_norm_ratio": 0.41412135660648347,
      "relation_loss": 0.001890819298569113,
      "selected_pairs": 26802.4,
      "step": 28180,
      "student_distance_mean": 22.443724060058592,
      "teacher_distance_mean": 0.8469112753868103,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.001890819298569113,
      "weighted_relation_loss": 0.001890819298569113
    },
    {
      "action_query_batch_std": 0.34356655478477477,
      "action_query_norm": 40.68004264831543,
      "action_relation_loss": 0.00195414531044662,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.37043966323667,
      "flow_matching_loss": 0.015117150079458953,
      "grad_norm": 0.35749441385269165,
      "h_stock_valid_norm": 97.91492156982422,
      "invalid_tail_samples": 22.9,
      "learning_rate": 9.92988397466449e-07,
      "loss": 0.0171,
      "positive_teacher_pairs": 27061.0,
      "query_h_norm_ratio": 0.4154694378376007,
      "relation_loss": 0.00195414531044662,
      "selected_pairs": 27061.0,
      "step": 28190,
      "student_distance_mean": 22.610592651367188,
      "teacher_distance_mean": 0.8543682515621185,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.00195414531044662,
      "weighted_relation_loss": 0.00195414531044662
    },
    {
      "action_query_batch_std": 0.3422421872615814,
      "action_query_norm": 40.64904136657715,
      "action_relation_loss": 0.0018587692524306477,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.379794200187092,
      "flow_matching_loss": 0.015196861885488033,
      "grad_norm": 0.14655059576034546,
      "h_stock_valid_norm": 98.05332717895507,
      "invalid_tail_samples": 26.2,
      "learning_rate": 9.820884275890286e-07,
      "loss": 0.0171,
      "positive_teacher_pairs": 26301.3,
      "query_h_norm_ratio": 0.41456357538700106,
      "relation_loss": 0.0018587692524306477,
      "selected_pairs": 26301.3,
      "step": 28200,
      "student_distance_mean": 22.486140060424805,
      "teacher_distance_mean": 0.8480534613132477,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.0018587692524306477,
      "weighted_relation_loss": 0.0018587692524306477
    },
    {
      "action_query_batch_std": 0.34753909409046174,
      "action_query_norm": 40.67670783996582,
      "action_relation_loss": 0.002043371566105634,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.38914873713751,
      "flow_matching_loss": 0.01616587396711111,
      "grad_norm": 0.0929461419582367,
      "h_stock_valid_norm": 98.00739135742188,
      "invalid_tail_samples": 23.4,
      "learning_rate": 9.712480190962546e-07,
      "loss": 0.0182,
      "positive_teacher_pairs": 26941.7,
      "query_h_norm_ratio": 0.4150390952825546,
      "relation_loss": 0.002043371566105634,
      "selected_pairs": 26941.7,
      "step": 28210,
      "student_distance_mean": 22.74064826965332,
      "teacher_distance_mean": 0.8588350296020508,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.002043371566105634,
      "weighted_relation_loss": 0.002043371566105634
    },
    {
      "action_query_batch_std": 0.3425449639558792,
      "action_query_norm": 40.645348739624026,
      "action_relation_loss": 0.0019858134677633644,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.398503274087933,
      "flow_matching_loss": 0.015134617034345866,
      "grad_norm": 0.11313315480947495,
      "h_stock_valid_norm": 97.9986686706543,
      "invalid_tail_samples": 25.6,
      "learning_rate": 9.604671851602464e-07,
      "loss": 0.0171,
      "positive_teacher_pairs": 26439.2,
      "query_h_norm_ratio": 0.41475659012794497,
      "relation_loss": 0.0019858134677633644,
      "selected_pairs": 26439.2,
      "step": 28220,
      "student_distance_mean": 22.477580070495605,
      "teacher_distance_mean": 0.8485673666000366,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0019858134677633644,
      "weighted_relation_loss": 0.0019858134677633644
    },
    {
      "action_query_batch_std": 0.3498458474874496,
      "action_query_norm": 40.6781982421875,
      "action_relation_loss": 0.0018865717225708067,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.407857811038355,
      "flow_matching_loss": 0.014292856864631175,
      "grad_norm": 0.07487934827804565,
      "h_stock_valid_norm": 98.1226707458496,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.497459388807306e-07,
      "loss": 0.0162,
      "positive_teacher_pairs": 26755.4,
      "query_h_norm_ratio": 0.4145659238100052,
      "relation_loss": 0.0018865717225708067,
      "selected_pairs": 26755.4,
      "step": 28230,
      "student_distance_mean": 22.98610954284668,
      "teacher_distance_mean": 0.8681500256061554,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0018865717225708067,
      "weighted_relation_loss": 0.0018865717225708067
    },
    {
      "action_query_batch_std": 0.348723778128624,
      "action_query_norm": 40.67290992736817,
      "action_relation_loss": 0.0018444214132614435,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.417212347988773,
      "flow_matching_loss": 0.01614487525075674,
      "grad_norm": 0.0890076532959938,
      "h_stock_valid_norm": 98.15604400634766,
      "invalid_tail_samples": 23.3,
      "learning_rate": 9.390842932850364e-07,
      "loss": 0.018,
      "positive_teacher_pairs": 26965.5,
      "query_h_norm_ratio": 0.4143732100725174,
      "relation_loss": 0.0018444214132614435,
      "selected_pairs": 26965.5,
      "step": 28240,
      "student_distance_mean": 22.76611919403076,
      "teacher_distance_mean": 0.8591656625270844,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0018444214132614435,
      "weighted_relation_loss": 0.0018444214132614435
    },
    {
      "action_query_batch_std": 0.3459518551826477,
      "action_query_norm": 40.657052993774414,
      "action_relation_loss": 0.001967150333803147,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.426566884939195,
      "flow_matching_loss": 0.014088431000709533,
      "grad_norm": 0.08985194563865662,
      "h_stock_valid_norm": 98.0487678527832,
      "invalid_tail_samples": 24.2,
      "learning_rate": 9.28482261328073e-07,
      "loss": 0.0161,
      "positive_teacher_pairs": 26763.9,
      "query_h_norm_ratio": 0.41466403007507324,
      "relation_loss": 0.001967150333803147,
      "selected_pairs": 26763.9,
      "step": 28250,
      "student_distance_mean": 22.715204429626464,
      "teacher_distance_mean": 0.8579949617385865,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.001967150333803147,
      "weighted_relation_loss": 0.001967150333803147
    },
    {
      "action_query_batch_std": 0.3456383913755417,
      "action_query_norm": 40.66974296569824,
      "action_relation_loss": 0.0019020536565221846,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.435921421889617,
      "flow_matching_loss": 0.015802106633782387,
      "grad_norm": 0.07933132350444794,
      "h_stock_valid_norm": 97.96704788208008,
      "invalid_tail_samples": 24.9,
      "learning_rate": 9.179398558923025e-07,
      "loss": 0.0177,
      "positive_teacher_pairs": 26596.3,
      "query_h_norm_ratio": 0.41514116823673247,
      "relation_loss": 0.0019020536565221846,
      "selected_pairs": 26596.3,
      "step": 28260,
      "student_distance_mean": 22.746760368347168,
      "teacher_distance_mean": 0.8584149181842804,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0019020536565221846,
      "weighted_relation_loss": 0.0019020536565221846
    },
    {
      "action_query_batch_std": 0.3422913432121277,
      "action_query_norm": 40.63134536743164,
      "action_relation_loss": 0.001950336736626923,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.44527595884004,
      "flow_matching_loss": 0.015377387031912803,
      "grad_norm": 0.09064161777496338,
      "h_stock_valid_norm": 98.06315307617187,
      "invalid_tail_samples": 23.9,
      "learning_rate": 9.074570897877388e-07,
      "loss": 0.0173,
      "positive_teacher_pairs": 26827.6,
      "query_h_norm_ratio": 0.4143424302339554,
      "relation_loss": 0.001950336736626923,
      "selected_pairs": 26827.6,
      "step": 28270,
      "student_distance_mean": 22.456009483337404,
      "teacher_distance_mean": 0.8475736498832702,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.001950336736626923,
      "weighted_relation_loss": 0.001950336736626923
    },
    {
      "action_query_batch_std": 0.3466391533613205,
      "action_query_norm": 40.673088455200194,
      "action_relation_loss": 0.0018839239492081106,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.454630495790457,
      "flow_matching_loss": 0.015093783941119909,
      "grad_norm": 0.08474863320589066,
      "h_stock_valid_norm": 97.95908737182617,
      "invalid_tail_samples": 25.5,
      "learning_rate": 8.970339757519375e-07,
      "loss": 0.017,
      "positive_teacher_pairs": 26453.5,
      "query_h_norm_ratio": 0.4152084618806839,
      "relation_loss": 0.0018839239492081106,
      "selected_pairs": 26453.5,
      "step": 28280,
      "student_distance_mean": 22.71503448486328,
      "teacher_distance_mean": 0.8563388228416443,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0018839239492081106,
      "weighted_relation_loss": 0.0018839239492081106
    },
    {
      "action_query_batch_std": 0.34387975335121157,
      "action_query_norm": 40.65600852966308,
      "action_relation_loss": 0.0018928476260043682,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.46398503274088,
      "flow_matching_loss": 0.013852775190025568,
      "grad_norm": 0.0748753473162651,
      "h_stock_valid_norm": 98.039453125,
      "invalid_tail_samples": 21.7,
      "learning_rate": 8.866705264499619e-07,
      "loss": 0.0157,
      "positive_teacher_pairs": 27339.7,
      "query_h_norm_ratio": 0.4146912157535553,
      "relation_loss": 0.0018928476260043682,
      "selected_pairs": 27339.7,
      "step": 28290,
      "student_distance_mean": 22.607766342163085,
      "teacher_distance_mean": 0.8530316948890686,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0018928476260043682,
      "weighted_relation_loss": 0.0018928476260043682
    },
    {
      "action_query_batch_std": 0.3473688870668411,
      "action_query_norm": 40.67500343322754,
      "action_relation_loss": 0.0018882961245253681,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.4733395696913,
      "flow_matching_loss": 0.014938625507056713,
      "grad_norm": 0.08862782269716263,
      "h_stock_valid_norm": 98.10865631103516,
      "invalid_tail_samples": 23.6,
      "learning_rate": 8.763667544743836e-07,
      "loss": 0.0168,
      "positive_teacher_pairs": 26894.4,
      "query_h_norm_ratio": 0.41459438800811765,
      "relation_loss": 0.0018882961245253681,
      "selected_pairs": 26894.4,
      "step": 28300,
      "student_distance_mean": 22.72402992248535,
      "teacher_distance_mean": 0.8569600880146027,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0018882961245253681,
      "weighted_relation_loss": 0.0018882961245253681
    },
    {
      "action_query_batch_std": 0.34414624869823457,
      "action_query_norm": 40.65187530517578,
      "action_relation_loss": 0.0019095975905656815,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.48269410664172,
      "flow_matching_loss": 0.0175497330725193,
      "grad_norm": 0.1090473011136055,
      "h_stock_valid_norm": 98.11175765991212,
      "invalid_tail_samples": 22.8,
      "learning_rate": 8.661226723452543e-07,
      "loss": 0.0195,
      "positive_teacher_pairs": 27079.4,
      "query_h_norm_ratio": 0.41434520184993745,
      "relation_loss": 0.0019095975905656815,
      "selected_pairs": 27079.4,
      "step": 28310,
      "student_distance_mean": 22.545166969299316,
      "teacher_distance_mean": 0.8499829590320587,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0019095975905656815,
      "weighted_relation_loss": 0.0019095975905656815
    },
    {
      "action_query_batch_std": 0.3444894254207611,
      "action_query_norm": 40.66260528564453,
      "action_relation_loss": 0.0018755345256067813,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.492048643592142,
      "flow_matching_loss": 0.01513704750686884,
      "grad_norm": 0.10727544873952866,
      "h_stock_valid_norm": 97.95141372680663,
      "invalid_tail_samples": 25.0,
      "learning_rate": 8.559382925101001e-07,
      "loss": 0.017,
      "positive_teacher_pairs": 26578.0,
      "query_h_norm_ratio": 0.4151322454214096,
      "relation_loss": 0.0018755345256067813,
      "selected_pairs": 26578.0,
      "step": 28320,
      "student_distance_mean": 22.562969970703126,
      "teacher_distance_mean": 0.8517216801643371,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0018755345256067813,
      "weighted_relation_loss": 0.0018755345256067813
    },
    {
      "action_query_batch_std": 0.3459402322769165,
      "action_query_norm": 40.67229118347168,
      "action_relation_loss": 0.001969429396558553,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.501403180542564,
      "flow_matching_loss": 0.014547730889171362,
      "grad_norm": 0.06048057600855827,
      "h_stock_valid_norm": 98.02919235229493,
      "invalid_tail_samples": 21.0,
      "learning_rate": 8.458136273438943e-07,
      "loss": 0.0165,
      "positive_teacher_pairs": 27505.6,
      "query_h_norm_ratio": 0.4149033933877945,
      "relation_loss": 0.001969429396558553,
      "selected_pairs": 27505.6,
      "step": 28330,
      "student_distance_mean": 22.717128372192384,
      "teacher_distance_mean": 0.8562718331813812,
      "valid_action_samples": 235.0,
      "weighted_auxiliary_loss": 0.001969429396558553,
      "weighted_relation_loss": 0.001969429396558553
    },
    {
      "action_query_batch_std": 0.34538787603378296,
      "action_query_norm": 40.651708984375,
      "action_relation_loss": 0.0019310611998662353,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.510757717492986,
      "flow_matching_loss": 0.014964998234063387,
      "grad_norm": 0.08070029318332672,
      "h_stock_valid_norm": 98.1090576171875,
      "invalid_tail_samples": 24.6,
      "learning_rate": 8.357486891490795e-07,
      "loss": 0.0169,
      "positive_teacher_pairs": 26677.5,
      "query_h_norm_ratio": 0.4143552243709564,
      "relation_loss": 0.0019310611998662353,
      "selected_pairs": 26677.5,
      "step": 28340,
      "student_distance_mean": 22.62691707611084,
      "teacher_distance_mean": 0.8554877758026123,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0019310611998662353,
      "weighted_relation_loss": 0.0019310611998662353
    },
    {
      "action_query_batch_std": 0.34280786514282224,
      "action_query_norm": 40.609968948364255,
      "action_relation_loss": 0.001974352903198451,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.520112254443404,
      "flow_matching_loss": 0.014869397971779108,
      "grad_norm": 0.09963560104370117,
      "h_stock_valid_norm": 98.01836547851562,
      "invalid_tail_samples": 26.6,
      "learning_rate": 8.257434901554895e-07,
      "loss": 0.0168,
      "positive_teacher_pairs": 26203.8,
      "query_h_norm_ratio": 0.4143129289150238,
      "relation_loss": 0.001974352903198451,
      "selected_pairs": 26203.8,
      "step": 28350,
      "student_distance_mean": 22.46621780395508,
      "teacher_distance_mean": 0.8471720695495606,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.001974352903198451,
      "weighted_relation_loss": 0.001974352903198451
    },
    {
      "action_query_batch_std": 0.35105487108230593,
      "action_query_norm": 40.703474044799805,
      "action_relation_loss": 0.0019430051324889064,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.529466791393826,
      "flow_matching_loss": 0.016058266814798116,
      "grad_norm": 0.10332192480564117,
      "h_stock_valid_norm": 98.06542129516602,
      "invalid_tail_samples": 19.7,
      "learning_rate": 8.157980425203937e-07,
      "loss": 0.018,
      "positive_teacher_pairs": 27811.4,
      "query_h_norm_ratio": 0.41506657302379607,
      "relation_loss": 0.0019430051324889064,
      "selected_pairs": 27811.4,
      "step": 28360,
      "student_distance_mean": 22.975078773498534,
      "teacher_distance_mean": 0.8673199236392974,
      "valid_action_samples": 236.3,
      "weighted_auxiliary_loss": 0.0019430051324889064,
      "weighted_relation_loss": 0.0019430051324889064
    },
    {
      "action_query_batch_std": 0.3449658930301666,
      "action_query_norm": 40.64685897827148,
      "action_relation_loss": 0.0019045793334953487,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.53882132834425,
      "flow_matching_loss": 0.016061093378812074,
      "grad_norm": 0.08599183708429337,
      "h_stock_valid_norm": 97.89244766235352,
      "invalid_tail_samples": 23.4,
      "learning_rate": 8.059123583284367e-07,
      "loss": 0.018,
      "positive_teacher_pairs": 26949.1,
      "query_h_norm_ratio": 0.4152203381061554,
      "relation_loss": 0.0019045793334953487,
      "selected_pairs": 26949.1,
      "step": 28370,
      "student_distance_mean": 22.60891799926758,
      "teacher_distance_mean": 0.855192244052887,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0019045793334953487,
      "weighted_relation_loss": 0.0019045793334953487
    },
    {
      "action_query_batch_std": 0.34441368877887724,
      "action_query_norm": 40.63159065246582,
      "action_relation_loss": 0.001909806637559086,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.548175865294667,
      "flow_matching_loss": 0.017204261478036643,
      "grad_norm": 0.10500700771808624,
      "h_stock_valid_norm": 98.15668563842773,
      "invalid_tail_samples": 24.2,
      "learning_rate": 7.960864495916654e-07,
      "loss": 0.0191,
      "positive_teacher_pairs": 26763.2,
      "query_h_norm_ratio": 0.4139470189809799,
      "relation_loss": 0.001909806637559086,
      "selected_pairs": 26763.2,
      "step": 28380,
      "student_distance_mean": 22.635890197753906,
      "teacher_distance_mean": 0.8548388957977295,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.001909806637559086,
      "weighted_relation_loss": 0.001909806637559086
    },
    {
      "action_query_batch_std": 0.3446394205093384,
      "action_query_norm": 40.653283309936526,
      "action_relation_loss": 0.001891755359247327,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.55753040224509,
      "flow_matching_loss": 0.014316547941416503,
      "grad_norm": 0.09753444045782089,
      "h_stock_valid_norm": 98.0890869140625,
      "invalid_tail_samples": 24.2,
      "learning_rate": 7.863203282494847e-07,
      "loss": 0.0162,
      "positive_teacher_pairs": 26762.7,
      "query_h_norm_ratio": 0.41445459723472594,
      "relation_loss": 0.001891755359247327,
      "selected_pairs": 26762.7,
      "step": 28390,
      "student_distance_mean": 22.544791984558106,
      "teacher_distance_mean": 0.8504279732704163,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.001891755359247327,
      "weighted_relation_loss": 0.001891755359247327
    },
    {
      "action_query_batch_std": 0.34644554555416107,
      "action_query_norm": 40.65263557434082,
      "action_relation_loss": 0.0019218853209167718,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.56688493919551,
      "flow_matching_loss": 0.017350408341735603,
      "grad_norm": 0.08295311033725739,
      "h_stock_valid_norm": 97.94049224853515,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.766140061686522e-07,
      "loss": 0.0193,
      "positive_teacher_pairs": 27036.6,
      "query_h_norm_ratio": 0.4150779336690903,
      "relation_loss": 0.0019218853209167718,
      "selected_pairs": 27036.6,
      "step": 28400,
      "student_distance_mean": 22.632478713989258,
      "teacher_distance_mean": 0.8548456370830536,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0019218853209167718,
      "weighted_relation_loss": 0.0019218853209167718
    },
    {
      "action_query_batch_std": 0.34502761662006376,
      "action_query_norm": 40.67035064697266,
      "action_relation_loss": 0.001857769617345184,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.57623947614593,
      "flow_matching_loss": 0.01652173148468137,
      "grad_norm": 0.09681697934865952,
      "h_stock_valid_norm": 98.06165924072266,
      "invalid_tail_samples": 23.4,
      "learning_rate": 7.669674951432615e-07,
      "loss": 0.0184,
      "positive_teacher_pairs": 26948.5,
      "query_h_norm_ratio": 0.4147443205118179,
      "relation_loss": 0.001857769617345184,
      "selected_pairs": 26948.5,
      "step": 28410,
      "student_distance_mean": 22.70509567260742,
      "teacher_distance_mean": 0.8581543505191803,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.001857769617345184,
      "weighted_relation_loss": 0.001857769617345184
    },
    {
      "action_query_batch_std": 0.3453625500202179,
      "action_query_norm": 40.6449577331543,
      "action_relation_loss": 0.0018810732290148735,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.58559401309635,
      "flow_matching_loss": 0.01450381837785244,
      "grad_norm": 0.07870914041996002,
      "h_stock_valid_norm": 98.03775634765626,
      "invalid_tail_samples": 23.7,
      "learning_rate": 7.573808068947363e-07,
      "loss": 0.0164,
      "positive_teacher_pairs": 26869.9,
      "query_h_norm_ratio": 0.41458560526371,
      "relation_loss": 0.0018810732290148735,
      "selected_pairs": 26869.9,
      "step": 28420,
      "student_distance_mean": 22.706098556518555,
      "teacher_distance_mean": 0.8575934290885925,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0018810732290148735,
      "weighted_relation_loss": 0.0018810732290148735
    },
    {
      "action_query_batch_std": 0.34388970732688906,
      "action_query_norm": 40.648625183105466,
      "action_relation_loss": 0.0019300182699225842,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.594948550046773,
      "flow_matching_loss": 0.016297975182533266,
      "grad_norm": 0.08757095038890839,
      "h_stock_valid_norm": 98.0457862854004,
      "invalid_tail_samples": 23.7,
      "learning_rate": 7.478539530718087e-07,
      "loss": 0.0182,
      "positive_teacher_pairs": 26874.9,
      "query_h_norm_ratio": 0.41459035873413086,
      "relation_loss": 0.0019300182699225842,
      "selected_pairs": 26874.9,
      "step": 28430,
      "student_distance_mean": 22.535107612609863,
      "teacher_distance_mean": 0.850175416469574,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0019300182699225842,
      "weighted_relation_loss": 0.0019300182699225842
    },
    {
      "action_query_batch_std": 0.347376412153244,
      "action_query_norm": 40.67503890991211,
      "action_relation_loss": 0.0019085540203377604,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.604303086997195,
      "flow_matching_loss": 0.016051025409251452,
      "grad_norm": 0.07844194024801254,
      "h_stock_valid_norm": 97.94168167114258,
      "invalid_tail_samples": 25.1,
      "learning_rate": 7.383869452504965e-07,
      "loss": 0.018,
      "positive_teacher_pairs": 26556.0,
      "query_h_norm_ratio": 0.41530259847640993,
      "relation_loss": 0.0019085540203377604,
      "selected_pairs": 26556.0,
      "step": 28440,
      "student_distance_mean": 22.784778594970703,
      "teacher_distance_mean": 0.8607399702072144,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0019085540203377604,
      "weighted_relation_loss": 0.0019085540203377604
    },
    {
      "action_query_batch_std": 0.34470994472503663,
      "action_query_norm": 40.655541610717776,
      "action_relation_loss": 0.0019059932208620011,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.613657623947613,
      "flow_matching_loss": 0.014676209166646003,
      "grad_norm": 0.09062466770410538,
      "h_stock_valid_norm": 98.06273498535157,
      "invalid_tail_samples": 24.0,
      "learning_rate": 7.289797949341204e-07,
      "loss": 0.0166,
      "positive_teacher_pairs": 26809.8,
      "query_h_norm_ratio": 0.41459025740623473,
      "relation_loss": 0.0019059932208620011,
      "selected_pairs": 26809.8,
      "step": 28450,
      "student_distance_mean": 22.59243564605713,
      "teacher_distance_mean": 0.8531899094581604,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0019059932208620011,
      "weighted_relation_loss": 0.0019059932208620011
    },
    {
      "action_query_batch_std": 0.3447964698076248,
      "action_query_norm": 40.66020240783691,
      "action_relation_loss": 0.0018250245251692832,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.623012160898035,
      "flow_matching_loss": 0.014434600621461869,
      "grad_norm": 0.13803601264953613,
      "h_stock_valid_norm": 97.93531646728516,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.196325135532423e-07,
      "loss": 0.0163,
      "positive_teacher_pairs": 27050.9,
      "query_h_norm_ratio": 0.4151787251234055,
      "relation_loss": 0.0018250245251692832,
      "selected_pairs": 27050.9,
      "step": 28460,
      "student_distance_mean": 22.669398307800293,
      "teacher_distance_mean": 0.8555469512939453,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0018250245251692832,
      "weighted_relation_loss": 0.0018250245251692832
    },
    {
      "action_query_batch_std": 0.34168158769607543,
      "action_query_norm": 40.62483215332031,
      "action_relation_loss": 0.0019250184181146325,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.632366697848457,
      "flow_matching_loss": 0.01620433423668146,
      "grad_norm": 0.14038628339767456,
      "h_stock_valid_norm": 98.03528366088867,
      "invalid_tail_samples": 25.1,
      "learning_rate": 7.10345112465699e-07,
      "loss": 0.0181,
      "positive_teacher_pairs": 26548.4,
      "query_h_norm_ratio": 0.4143921881914139,
      "relation_loss": 0.0019250184181146325,
      "selected_pairs": 26548.4,
      "step": 28470,
      "student_distance_mean": 22.439685440063478,
      "teacher_distance_mean": 0.8468271970748902,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0019250184181146325,
      "weighted_relation_loss": 0.0019250184181146325
    },
    {
      "action_query_batch_std": 0.3458171248435974,
      "action_query_norm": 40.64672431945801,
      "action_relation_loss": 0.001987006212584674,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.641721234798876,
      "flow_matching_loss": 0.01573573099449277,
      "grad_norm": 0.09151118248701096,
      "h_stock_valid_norm": 97.9300765991211,
      "invalid_tail_samples": 23.5,
      "learning_rate": 7.011176029565525e-07,
      "loss": 0.0177,
      "positive_teacher_pairs": 26919.3,
      "query_h_norm_ratio": 0.4150622248649597,
      "relation_loss": 0.001987006212584674,
      "selected_pairs": 26919.3,
      "step": 28480,
      "student_distance_mean": 22.635318183898924,
      "teacher_distance_mean": 0.8534184396266937,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.001987006212584674,
      "weighted_relation_loss": 0.001987006212584674
    },
    {
      "action_query_batch_std": 0.3416070371866226,
      "action_query_norm": 40.62473564147949,
      "action_relation_loss": 0.00194532253080979,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.651075771749298,
      "flow_matching_loss": 0.013285716902464629,
      "grad_norm": 0.10016264766454697,
      "h_stock_valid_norm": 98.0619125366211,
      "invalid_tail_samples": 23.8,
      "learning_rate": 6.919499962381004e-07,
      "loss": 0.0152,
      "positive_teacher_pairs": 26856.5,
      "query_h_norm_ratio": 0.41427911818027496,
      "relation_loss": 0.00194532253080979,
      "selected_pairs": 26856.5,
      "step": 28490,
      "student_distance_mean": 22.442816734313965,
      "teacher_distance_mean": 0.8486289858818055,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.00194532253080979,
      "weighted_relation_loss": 0.00194532253080979
    },
    {
      "action_query_batch_std": 0.341568061709404,
      "action_query_norm": 40.652225494384766,
      "action_relation_loss": 0.0018855810281820594,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.66043030869972,
      "flow_matching_loss": 0.0160297485999763,
      "grad_norm": 0.10587362945079803,
      "h_stock_valid_norm": 98.20681457519531,
      "invalid_tail_samples": 25.7,
      "learning_rate": 6.828423034498488e-07,
      "loss": 0.0179,
      "positive_teacher_pairs": 26417.8,
      "query_h_norm_ratio": 0.4139466017484665,
      "relation_loss": 0.0018855810281820594,
      "selected_pairs": 26417.8,
      "step": 28500,
      "student_distance_mean": 22.444453048706055,
      "teacher_distance_mean": 0.8477632701396942,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0018855810281820594,
      "weighted_relation_loss": 0.0018855810281820594
    },
    {
      "action_query_batch_std": 0.3453330546617508,
      "action_query_norm": 40.65935173034668,
      "action_relation_loss": 0.001920159487053752,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.669784845650142,
      "flow_matching_loss": 0.016041974443942307,
      "grad_norm": 0.07946043461561203,
      "h_stock_valid_norm": 97.9972915649414,
      "invalid_tail_samples": 24.3,
      "learning_rate": 6.737945356585007e-07,
      "loss": 0.018,
      "positive_teacher_pairs": 26731.4,
      "query_h_norm_ratio": 0.41490461230278014,
      "relation_loss": 0.001920159487053752,
      "selected_pairs": 26731.4,
      "step": 28510,
      "student_distance_mean": 22.705199813842775,
      "teacher_distance_mean": 0.8567954659461975,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.001920159487053752,
      "weighted_relation_loss": 0.001920159487053752
    },
    {
      "action_query_batch_std": 0.3428832530975342,
      "action_query_norm": 40.61707992553711,
      "action_relation_loss": 0.0019216106738895178,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.67913938260056,
      "flow_matching_loss": 0.014914426021277905,
      "grad_norm": 0.061718206852674484,
      "h_stock_valid_norm": 98.0884620666504,
      "invalid_tail_samples": 26.2,
      "learning_rate": 6.648067038579508e-07,
      "loss": 0.0168,
      "positive_teacher_pairs": 26295.4,
      "query_h_norm_ratio": 0.4140878677368164,
      "relation_loss": 0.0019216106738895178,
      "selected_pairs": 26295.4,
      "step": 28520,
      "student_distance_mean": 22.54668560028076,
      "teacher_distance_mean": 0.8512162327766418,
      "valid_action_samples": 229.8,
      "weighted_auxiliary_loss": 0.0019216106738895178,
      "weighted_relation_loss": 0.0019216106738895178
    },
    {
      "action_query_batch_std": 0.3457224369049072,
      "action_query_norm": 40.653957748413085,
      "action_relation_loss": 0.001903487544041127,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.688493919550982,
      "flow_matching_loss": 0.01568887960165739,
      "grad_norm": 0.09753907471895218,
      "h_stock_valid_norm": 98.08378677368164,
      "invalid_tail_samples": 25.5,
      "learning_rate": 6.558788189692578e-07,
      "loss": 0.0176,
      "positive_teacher_pairs": 26465.6,
      "query_h_norm_ratio": 0.41448464393615725,
      "relation_loss": 0.001903487544041127,
      "selected_pairs": 26465.6,
      "step": 28530,
      "student_distance_mean": 22.71395378112793,
      "teacher_distance_mean": 0.8580532133579254,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.001903487544041127,
      "weighted_relation_loss": 0.001903487544041127
    },
    {
      "action_query_batch_std": 0.3483422338962555,
      "action_query_norm": 40.670661544799806,
      "action_relation_loss": 0.001999036455526948,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.697848456501404,
      "flow_matching_loss": 0.016574554331600665,
      "grad_norm": 0.10577826201915741,
      "h_stock_valid_norm": 98.07505645751954,
      "invalid_tail_samples": 23.3,
      "learning_rate": 6.470108918406492e-07,
      "loss": 0.0186,
      "positive_teacher_pairs": 26969.8,
      "query_h_norm_ratio": 0.4146911323070526,
      "relation_loss": 0.001999036455526948,
      "selected_pairs": 26969.8,
      "step": 28540,
      "student_distance_mean": 22.86358222961426,
      "teacher_distance_mean": 0.8623511373996735,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.001999036455526948,
      "weighted_relation_loss": 0.001999036455526948
    },
    {
      "action_query_batch_std": 0.3459888160228729,
      "action_query_norm": 40.66293601989746,
      "action_relation_loss": 0.0019242855720221995,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.707202993451823,
      "flow_matching_loss": 0.01592422565445304,
      "grad_norm": 0.08486241847276688,
      "h_stock_valid_norm": 98.15690765380859,
      "invalid_tail_samples": 25.5,
      "learning_rate": 6.382029332474893e-07,
      "loss": 0.0178,
      "positive_teacher_pairs": 26464.8,
      "query_h_norm_ratio": 0.41426857113838195,
      "relation_loss": 0.0019242855720221995,
      "selected_pairs": 26464.8,
      "step": 28550,
      "student_distance_mean": 22.755645561218262,
      "teacher_distance_mean": 0.8584609389305115,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0019242855720221995,
      "weighted_relation_loss": 0.0019242855720221995
    },
    {
      "action_query_batch_std": 0.34326491951942445,
      "action_query_norm": 40.64050636291504,
      "action_relation_loss": 0.0021409867564216255,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.716557530402245,
      "flow_matching_loss": 0.01577318236231804,
      "grad_norm": 0.09414679557085037,
      "h_stock_valid_norm": 98.02385635375977,
      "invalid_tail_samples": 22.2,
      "learning_rate": 6.294549538922778e-07,
      "loss": 0.0179,
      "positive_teacher_pairs": 27220.1,
      "query_h_norm_ratio": 0.41460283994674685,
      "relation_loss": 0.0021409867564216255,
      "selected_pairs": 27220.1,
      "step": 28560,
      "student_distance_mean": 22.505900955200197,
      "teacher_distance_mean": 0.850541865825653,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.0021409867564216255,
      "weighted_relation_loss": 0.0021409867564216255
    },
    {
      "action_query_batch_std": 0.3407123953104019,
      "action_query_norm": 40.65793800354004,
      "action_relation_loss": 0.0019530417514033616,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.725912067352667,
      "flow_matching_loss": 0.015501117054373025,
      "grad_norm": 0.09705690294504166,
      "h_stock_valid_norm": 97.97909927368164,
      "invalid_tail_samples": 26.3,
      "learning_rate": 6.207669644046344e-07,
      "loss": 0.0175,
      "positive_teacher_pairs": 26270.6,
      "query_h_norm_ratio": 0.41496730744838717,
      "relation_loss": 0.0019530417514033616,
      "selected_pairs": 26270.6,
      "step": 28570,
      "student_distance_mean": 22.35112762451172,
      "teacher_distance_mean": 0.8431589782238007,
      "valid_action_samples": 229.7,
      "weighted_auxiliary_loss": 0.0019530417514033616,
      "weighted_relation_loss": 0.0019530417514033616
    },
    {
      "action_query_batch_std": 0.34506458044052124,
      "action_query_norm": 40.64891395568848,
      "action_relation_loss": 0.0019663577666506173,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.73526660430309,
      "flow_matching_loss": 0.01558061558753252,
      "grad_norm": 0.07689682394266129,
      "h_stock_valid_norm": 98.0078010559082,
      "invalid_tail_samples": 25.2,
      "learning_rate": 6.121389753412865e-07,
      "loss": 0.0175,
      "positive_teacher_pairs": 26530.8,
      "query_h_norm_ratio": 0.4147532433271408,
      "relation_loss": 0.0019663577666506173,
      "selected_pairs": 26530.8,
      "step": 28580,
      "student_distance_mean": 22.751455879211427,
      "teacher_distance_mean": 0.8585484027862549,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0019663577666506173,
      "weighted_relation_loss": 0.0019663577666506173
    },
    {
      "action_query_batch_std": 0.345592674612999,
      "action_query_norm": 40.686194229125974,
      "action_relation_loss": 0.001864227221813053,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.744621141253507,
      "flow_matching_loss": 0.01564035601913929,
      "grad_norm": 0.09621748328208923,
      "h_stock_valid_norm": 97.99930725097656,
      "invalid_tail_samples": 22.3,
      "learning_rate": 6.035709971860592e-07,
      "loss": 0.0175,
      "positive_teacher_pairs": 27202.7,
      "query_h_norm_ratio": 0.41517068445682526,
      "relation_loss": 0.001864227221813053,
      "selected_pairs": 27202.7,
      "step": 28590,
      "student_distance_mean": 22.71358127593994,
      "teacher_distance_mean": 0.8577908873558044,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.001864227221813053,
      "weighted_relation_loss": 0.001864227221813053
    },
    {
      "action_query_batch_std": 0.33898238837718964,
      "action_query_norm": 40.59423751831055,
      "action_relation_loss": 0.0019406433217227458,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.75397567820393,
      "flow_matching_loss": 0.015530369989573956,
      "grad_norm": 0.09825153648853302,
      "h_stock_valid_norm": 97.90933990478516,
      "invalid_tail_samples": 25.2,
      "learning_rate": 5.950630403498469e-07,
      "loss": 0.0175,
      "positive_teacher_pairs": 26524.3,
      "query_h_norm_ratio": 0.4146134525537491,
      "relation_loss": 0.0019406433217227458,
      "selected_pairs": 26524.3,
      "step": 28600,
      "student_distance_mean": 22.303813552856447,
      "teacher_distance_mean": 0.8417169392108917,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.0019406433217227458,
      "weighted_relation_loss": 0.0019406433217227458
    },
    {
      "action_query_batch_std": 0.3509583830833435,
      "action_query_norm": 40.7065128326416,
      "action_relation_loss": 0.0018210150068625809,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.76333021515435,
      "flow_matching_loss": 0.014579904172569514,
      "grad_norm": 0.10436809808015823,
      "h_stock_valid_norm": 97.95241165161133,
      "invalid_tail_samples": 23.5,
      "learning_rate": 5.866151151706189e-07,
      "loss": 0.0164,
      "positive_teacher_pairs": 26920.2,
      "query_h_norm_ratio": 0.41557946503162385,
      "relation_loss": 0.0018210150068625809,
      "selected_pairs": 26920.2,
      "step": 28610,
      "student_distance_mean": 22.968615913391112,
      "teacher_distance_mean": 0.8678649067878723,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0018210150068625809,
      "weighted_relation_loss": 0.0018210150068625809
    },
    {
      "action_query_batch_std": 0.3467617005109787,
      "action_query_norm": 40.6765754699707,
      "action_relation_loss": 0.001868311723228544,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.77268475210477,
      "flow_matching_loss": 0.015408106241375208,
      "grad_norm": 0.10326805710792542,
      "h_stock_valid_norm": 98.1033920288086,
      "invalid_tail_samples": 25.2,
      "learning_rate": 5.782272319134086e-07,
      "loss": 0.0173,
      "positive_teacher_pairs": 26525.9,
      "query_h_norm_ratio": 0.41463262438774107,
      "relation_loss": 0.001868311723228544,
      "selected_pairs": 26525.9,
      "step": 28620,
      "student_distance_mean": 22.78101215362549,
      "teacher_distance_mean": 0.860746318101883,
      "valid_action_samples": 230.8,
      "weighted_auxiliary_loss": 0.001868311723228544,
      "weighted_relation_loss": 0.001868311723228544
    },
    {
      "action_query_batch_std": 0.3496330320835114,
      "action_query_norm": 40.710746383666994,
      "action_relation_loss": 0.0018316762172617019,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.78203928905519,
      "flow_matching_loss": 0.01646695742383599,
      "grad_norm": 0.08988454937934875,
      "h_stock_valid_norm": 97.90828323364258,
      "invalid_tail_samples": 20.5,
      "learning_rate": 5.698994007702796e-07,
      "loss": 0.0183,
      "positive_teacher_pairs": 27614.9,
      "query_h_norm_ratio": 0.41580714881420133,
      "relation_loss": 0.0018316762172617019,
      "selected_pairs": 27614.9,
      "step": 28630,
      "student_distance_mean": 22.949913597106935,
      "teacher_distance_mean": 0.865999561548233,
      "valid_action_samples": 235.5,
      "weighted_auxiliary_loss": 0.0018316762172617019,
      "weighted_relation_loss": 0.0018316762172617019
    },
    {
      "action_query_batch_std": 0.3447363436222076,
      "action_query_norm": 40.67645378112793,
      "action_relation_loss": 0.001839390303939581,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.791393826005613,
      "flow_matching_loss": 0.016477269772440194,
      "grad_norm": 0.10883059352636337,
      "h_stock_valid_norm": 97.97100448608398,
      "invalid_tail_samples": 23.2,
      "learning_rate": 5.616316318603321e-07,
      "loss": 0.0183,
      "positive_teacher_pairs": 26989.4,
      "query_h_norm_ratio": 0.415190601348877,
      "relation_loss": 0.001839390303939581,
      "selected_pairs": 26989.4,
      "step": 28640,
      "student_distance_mean": 22.623092651367188,
      "teacher_distance_mean": 0.8547655522823334,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.001839390303939581,
      "weighted_relation_loss": 0.001839390303939581
    },
    {
      "action_query_batch_std": 0.34542000889778135,
      "action_query_norm": 40.670651245117185,
      "action_relation_loss": 0.0019261471345089376,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.800748362956035,
      "flow_matching_loss": 0.01686685290187597,
      "grad_norm": 0.22059273719787598,
      "h_stock_valid_norm": 98.11084289550782,
      "invalid_tail_samples": 23.0,
      "learning_rate": 5.534239352296799e-07,
      "loss": 0.0188,
      "positive_teacher_pairs": 27045.7,
      "query_h_norm_ratio": 0.4145396053791046,
      "relation_loss": 0.0019261471345089376,
      "selected_pairs": 27045.7,
      "step": 28650,
      "student_distance_mean": 22.677691078186037,
      "teacher_distance_mean": 0.8579043328762055,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0019261471345089376,
      "weighted_relation_loss": 0.0019261471345089376
    },
    {
      "action_query_batch_std": 0.3458202421665192,
      "action_query_norm": 40.67505035400391,
      "action_relation_loss": 0.0018667754950001836,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.810102899906454,
      "flow_matching_loss": 0.016106653865426777,
      "grad_norm": 0.10795186460018158,
      "h_stock_valid_norm": 98.09868392944335,
      "invalid_tail_samples": 22.8,
      "learning_rate": 5.452763208514621e-07,
      "loss": 0.018,
      "positive_teacher_pairs": 27084.3,
      "query_h_norm_ratio": 0.4146366655826569,
      "relation_loss": 0.0018667754950001836,
      "selected_pairs": 27084.3,
      "step": 28660,
      "student_distance_mean": 22.7391637802124,
      "teacher_distance_mean": 0.8595351696014404,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0018667754950001836,
      "weighted_relation_loss": 0.0018667754950001836
    },
    {
      "action_query_batch_std": 0.3429455190896988,
      "action_query_norm": 40.642880249023435,
      "action_relation_loss": 0.0019140218151733278,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.819457436856876,
      "flow_matching_loss": 0.01478933310136199,
      "grad_norm": 0.10106346756219864,
      "h_stock_valid_norm": 98.01764373779297,
      "invalid_tail_samples": 25.7,
      "learning_rate": 5.371887986257873e-07,
      "loss": 0.0167,
      "positive_teacher_pairs": 26415.9,
      "query_h_norm_ratio": 0.41464942693710327,
      "relation_loss": 0.0019140218151733278,
      "selected_pairs": 26415.9,
      "step": 28670,
      "student_distance_mean": 22.487228775024413,
      "teacher_distance_mean": 0.8499682664871215,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0019140218151733278,
      "weighted_relation_loss": 0.0019140218151733278
    },
    {
      "action_query_batch_std": 0.3480007767677307,
      "action_query_norm": 40.67230339050293,
      "action_relation_loss": 0.0018912755069322885,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.828811973807298,
      "flow_matching_loss": 0.01495769377797842,
      "grad_norm": 0.15521307289600372,
      "h_stock_valid_norm": 98.04936370849609,
      "invalid_tail_samples": 23.7,
      "learning_rate": 5.291613783797611e-07,
      "loss": 0.0168,
      "positive_teacher_pairs": 26874.0,
      "query_h_norm_ratio": 0.4148165285587311,
      "relation_loss": 0.0018912755069322885,
      "selected_pairs": 26874.0,
      "step": 28680,
      "student_distance_mean": 22.819181442260742,
      "teacher_distance_mean": 0.8619510054588317,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0018912755069322885,
      "weighted_relation_loss": 0.0018912755069322885
    },
    {
      "action_query_batch_std": 0.34410865902900695,
      "action_query_norm": 40.6488956451416,
      "action_relation_loss": 0.0019063864601776004,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.838166510757716,
      "flow_matching_loss": 0.015718712005764246,
      "grad_norm": 0.10687731951475143,
      "h_stock_valid_norm": 98.03108673095703,
      "invalid_tail_samples": 22.7,
      "learning_rate": 5.211940698674534e-07,
      "loss": 0.0176,
      "positive_teacher_pairs": 27107.9,
      "query_h_norm_ratio": 0.4146540343761444,
      "relation_loss": 0.0019063864601776004,
      "selected_pairs": 27107.9,
      "step": 28690,
      "student_distance_mean": 22.690375518798827,
      "teacher_distance_mean": 0.8573776960372925,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0019063864601776004,
      "weighted_relation_loss": 0.0019063864601776004
    },
    {
      "action_query_batch_std": 0.3476702481508255,
      "action_query_norm": 40.69908599853515,
      "action_relation_loss": 0.0018979432294145226,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.84752104770814,
      "flow_matching_loss": 0.015477770939469337,
      "grad_norm": 0.08365975320339203,
      "h_stock_valid_norm": 98.07321624755859,
      "invalid_tail_samples": 20.5,
      "learning_rate": 5.132868827698978e-07,
      "loss": 0.0174,
      "positive_teacher_pairs": 27614.3,
      "query_h_norm_ratio": 0.4149881362915039,
      "relation_loss": 0.0018979432294145226,
      "selected_pairs": 27614.3,
      "step": 28700,
      "student_distance_mean": 22.813249588012695,
      "teacher_distance_mean": 0.8616655707359314,
      "valid_action_samples": 235.5,
      "weighted_auxiliary_loss": 0.0018979432294145226,
      "weighted_relation_loss": 0.0018979432294145226
    },
    {
      "action_query_batch_std": 0.3429137974977493,
      "action_query_norm": 40.64133186340332,
      "action_relation_loss": 0.0018985578208230436,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.85687558465856,
      "flow_matching_loss": 0.01623966982588172,
      "grad_norm": 0.13146062195301056,
      "h_stock_valid_norm": 97.98089599609375,
      "invalid_tail_samples": 23.3,
      "learning_rate": 5.054398266950755e-07,
      "loss": 0.0181,
      "positive_teacher_pairs": 26969.0,
      "query_h_norm_ratio": 0.4147909551858902,
      "relation_loss": 0.0018985578208230436,
      "selected_pairs": 26969.0,
      "step": 28710,
      "student_distance_mean": 22.555505180358885,
      "teacher_distance_mean": 0.8527831256389617,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0018985578208230436,
      "weighted_relation_loss": 0.0018985578208230436
    },
    {
      "action_query_batch_std": 0.3446595162153244,
      "action_query_norm": 40.65447235107422,
      "action_relation_loss": 0.0019151362939737737,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.86623012160898,
      "flow_matching_loss": 0.016960871033370496,
      "grad_norm": 0.12362152338027954,
      "h_stock_valid_norm": 97.92088165283204,
      "invalid_tail_samples": 21.6,
      "learning_rate": 4.976529111778872e-07,
      "loss": 0.0189,
      "positive_teacher_pairs": 27363.5,
      "query_h_norm_ratio": 0.415182489156723,
      "relation_loss": 0.0019151362939737737,
      "selected_pairs": 27363.5,
      "step": 28720,
      "student_distance_mean": 22.590970039367676,
      "teacher_distance_mean": 0.8525619328022003,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.0019151362939737737,
      "weighted_relation_loss": 0.0019151362939737737
    },
    {
      "action_query_batch_std": 0.34459096789360044,
      "action_query_norm": 40.66500587463379,
      "action_relation_loss": 0.0018246605060994626,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.8755846585594,
      "flow_matching_loss": 0.015251260530203582,
      "grad_norm": 0.08864916861057281,
      "h_stock_valid_norm": 97.96970596313477,
      "invalid_tail_samples": 22.4,
      "learning_rate": 4.899261456801862e-07,
      "loss": 0.0171,
      "positive_teacher_pairs": 27175.2,
      "query_h_norm_ratio": 0.4150779217481613,
      "relation_loss": 0.0018246605060994626,
      "selected_pairs": 27175.2,
      "step": 28730,
      "student_distance_mean": 22.635653495788574,
      "teacher_distance_mean": 0.8545255720615387,
      "valid_action_samples": 233.6,
      "weighted_auxiliary_loss": 0.0018246605060994626,
      "weighted_relation_loss": 0.0018246605060994626
    },
    {
      "action_query_batch_std": 0.34496279060840607,
      "action_query_norm": 40.63869400024414,
      "action_relation_loss": 0.0018974337144754827,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.884939195509823,
      "flow_matching_loss": 0.015151079278439283,
      "grad_norm": 0.08713995665311813,
      "h_stock_valid_norm": 98.0743782043457,
      "invalid_tail_samples": 24.9,
      "learning_rate": 4.822595395907126e-07,
      "loss": 0.017,
      "positive_teacher_pairs": 26594.9,
      "query_h_norm_ratio": 0.4143675059080124,
      "relation_loss": 0.0018974337144754827,
      "selected_pairs": 26594.9,
      "step": 28740,
      "student_distance_mean": 22.644382095336915,
      "teacher_distance_mean": 0.8541071772575378,
      "valid_action_samples": 231.1,
      "weighted_auxiliary_loss": 0.0018974337144754827,
      "weighted_relation_loss": 0.0018974337144754827
    },
    {
      "action_query_batch_std": 0.3449076503515244,
      "action_query_norm": 40.64532737731933,
      "action_relation_loss": 0.001967469393275678,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.894293732460245,
      "flow_matching_loss": 0.01503399247303605,
      "grad_norm": 0.0822703167796135,
      "h_stock_valid_norm": 98.02632675170898,
      "invalid_tail_samples": 21.7,
      "learning_rate": 4.7465310222510886e-07,
      "loss": 0.017,
      "positive_teacher_pairs": 27334.8,
      "query_h_norm_ratio": 0.41464069187641145,
      "relation_loss": 0.001967469393275678,
      "selected_pairs": 27334.8,
      "step": 28750,
      "student_distance_mean": 22.62463512420654,
      "teacher_distance_mean": 0.8548985064029694,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.001967469393275678,
      "weighted_relation_loss": 0.001967469393275678
    },
    {
      "action_query_batch_std": 0.34413479268550873,
      "action_query_norm": 40.643023300170896,
      "action_relation_loss": 0.001885429909452796,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.903648269410663,
      "flow_matching_loss": 0.0161546902731061,
      "grad_norm": 0.11939763277769089,
      "h_stock_valid_norm": 97.97553787231445,
      "invalid_tail_samples": 23.4,
      "learning_rate": 4.6710684282593175e-07,
      "loss": 0.018,
      "positive_teacher_pairs": 26942.0,
      "query_h_norm_ratio": 0.4148333102464676,
      "relation_loss": 0.001885429909452796,
      "selected_pairs": 26942.0,
      "step": 28760,
      "student_distance_mean": 22.61316833496094,
      "teacher_distance_mean": 0.8537512123584747,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.001885429909452796,
      "weighted_relation_loss": 0.001885429909452796
    },
    {
      "action_query_batch_std": 0.3416202127933502,
      "action_query_norm": 40.611328506469725,
      "action_relation_loss": 0.001896230154670775,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.913002806361085,
      "flow_matching_loss": 0.015268267411738635,
      "grad_norm": 0.11117240786552429,
      "h_stock_valid_norm": 97.96319580078125,
      "invalid_tail_samples": 24.6,
      "learning_rate": 4.5962077056257993e-07,
      "loss": 0.0172,
      "positive_teacher_pairs": 26663.3,
      "query_h_norm_ratio": 0.41456195116043093,
      "relation_loss": 0.001896230154670775,
      "selected_pairs": 26663.3,
      "step": 28770,
      "student_distance_mean": 22.4412561416626,
      "teacher_distance_mean": 0.84829261302948,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.001896230154670775,
      "weighted_relation_loss": 0.001896230154670775
    },
    {
      "action_query_batch_std": 0.3448005437850952,
      "action_query_norm": 40.663866424560545,
      "action_relation_loss": 0.0018641120404936373,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.922357343311507,
      "flow_matching_loss": 0.013924196362495422,
      "grad_norm": 0.0974721908569336,
      "h_stock_valid_norm": 98.15869750976563,
      "invalid_tail_samples": 23.6,
      "learning_rate": 4.521948945313492e-07,
      "loss": 0.0158,
      "positive_teacher_pairs": 26896.0,
      "query_h_norm_ratio": 0.4142685174942017,
      "relation_loss": 0.0018641120404936373,
      "selected_pairs": 26896.0,
      "step": 28780,
      "student_distance_mean": 22.700617408752443,
      "teacher_distance_mean": 0.85729221701622,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0018641120404936373,
      "weighted_relation_loss": 0.0018641120404936373
    },
    {
      "action_query_batch_std": 0.3439606040716171,
      "action_query_norm": 40.63515777587891,
      "action_relation_loss": 0.0018694004858843983,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.931711880261926,
      "flow_matching_loss": 0.016600980516523124,
      "grad_norm": 0.08024615049362183,
      "h_stock_valid_norm": 98.01500244140625,
      "invalid_tail_samples": 28.2,
      "learning_rate": 4.4482922375537195e-07,
      "loss": 0.0185,
      "positive_teacher_pairs": 25845.2,
      "query_h_norm_ratio": 0.41458675265312195,
      "relation_loss": 0.0018694004858843983,
      "selected_pairs": 25845.2,
      "step": 28790,
      "student_distance_mean": 22.599935722351074,
      "teacher_distance_mean": 0.8534460604190827,
      "valid_action_samples": 227.8,
      "weighted_auxiliary_loss": 0.0018694004858843983,
      "weighted_relation_loss": 0.0018694004858843983
    },
    {
      "action_query_batch_std": 0.3467907279729843,
      "action_query_norm": 40.679429626464845,
      "action_relation_loss": 0.0018411310971714555,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.941066417212348,
      "flow_matching_loss": 0.015422448795288802,
      "grad_norm": 0.07750359922647476,
      "h_stock_valid_norm": 98.0592269897461,
      "invalid_tail_samples": 25.9,
      "learning_rate": 4.375237671846333e-07,
      "loss": 0.0173,
      "positive_teacher_pairs": 26368.6,
      "query_h_norm_ratio": 0.41484657526016233,
      "relation_loss": 0.0018411310971714555,
      "selected_pairs": 26368.6,
      "step": 28800,
      "student_distance_mean": 22.724099540710448,
      "teacher_distance_mean": 0.8573513150215148,
      "valid_action_samples": 230.1,
      "weighted_auxiliary_loss": 0.0018411310971714555,
      "weighted_relation_loss": 0.0018411310971714555
    },
    {
      "action_query_batch_std": 0.3408410042524338,
      "action_query_norm": 40.63409309387207,
      "action_relation_loss": 0.0020129294716753066,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.95042095416277,
      "flow_matching_loss": 0.014793674740940333,
      "grad_norm": 0.07058468461036682,
      "h_stock_valid_norm": 97.94217529296876,
      "invalid_tail_samples": 26.4,
      "learning_rate": 4.302785336959547e-07,
      "loss": 0.0168,
      "positive_teacher_pairs": 26246.3,
      "query_h_norm_ratio": 0.4148805499076843,
      "relation_loss": 0.0020129294716753066,
      "selected_pairs": 26246.3,
      "step": 28810,
      "student_distance_mean": 22.43456039428711,
      "teacher_distance_mean": 0.8468885838985443,
      "valid_action_samples": 229.6,
      "weighted_auxiliary_loss": 0.0020129294716753066,
      "weighted_relation_loss": 0.0020129294716753066
    },
    {
      "action_query_batch_std": 0.3429940611124039,
      "action_query_norm": 40.65915985107422,
      "action_relation_loss": 0.0019279515952803194,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.95977549111319,
      "flow_matching_loss": 0.016088382340967655,
      "grad_norm": 0.11356905847787857,
      "h_stock_valid_norm": 97.93734130859374,
      "invalid_tail_samples": 25.3,
      "learning_rate": 4.2309353209297744e-07,
      "loss": 0.018,
      "positive_teacher_pairs": 26515.1,
      "query_h_norm_ratio": 0.415159210562706,
      "relation_loss": 0.0019279515952803194,
      "selected_pairs": 26515.1,
      "step": 28820,
      "student_distance_mean": 22.483333778381347,
      "teacher_distance_mean": 0.8493692994117736,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0019279515952803194,
      "weighted_relation_loss": 0.0019279515952803194
    },
    {
      "action_query_batch_std": 0.34410462379455564,
      "action_query_norm": 40.668283462524414,
      "action_relation_loss": 0.001945751893799752,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.96913002806361,
      "flow_matching_loss": 0.014560600370168686,
      "grad_norm": 0.08850464224815369,
      "h_stock_valid_norm": 98.01514587402343,
      "invalid_tail_samples": 20.3,
      "learning_rate": 4.159687711061566e-07,
      "loss": 0.0165,
      "positive_teacher_pairs": 27667.7,
      "query_h_norm_ratio": 0.4149208813905716,
      "relation_loss": 0.001945751893799752,
      "selected_pairs": 27667.7,
      "step": 28830,
      "student_distance_mean": 22.568101501464845,
      "teacher_distance_mean": 0.8518456935882568,
      "valid_action_samples": 235.7,
      "weighted_auxiliary_loss": 0.001945751893799752,
      "weighted_relation_loss": 0.001945751893799752
    },
    {
      "action_query_batch_std": 0.3451714128255844,
      "action_query_norm": 40.6544994354248,
      "action_relation_loss": 0.0019192687468603253,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.978484565014032,
      "flow_matching_loss": 0.016568890400230886,
      "grad_norm": 0.12760326266288757,
      "h_stock_valid_norm": 97.78883285522461,
      "invalid_tail_samples": 21.6,
      "learning_rate": 4.0890425939275055e-07,
      "loss": 0.0185,
      "positive_teacher_pairs": 27366.7,
      "query_h_norm_ratio": 0.41574046313762664,
      "relation_loss": 0.0019192687468603253,
      "selected_pairs": 27366.7,
      "step": 28840,
      "student_distance_mean": 22.5510046005249,
      "teacher_distance_mean": 0.8516728281974792,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.0019192687468603253,
      "weighted_relation_loss": 0.0019192687468603253
    },
    {
      "action_query_batch_std": 0.34515070021152494,
      "action_query_norm": 40.65451774597168,
      "action_relation_loss": 0.0018836261006072164,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.987839101964454,
      "flow_matching_loss": 0.014733590837568044,
      "grad_norm": 0.09194820374250412,
      "h_stock_valid_norm": 98.01902694702149,
      "invalid_tail_samples": 21.6,
      "learning_rate": 4.0190000553679827e-07,
      "loss": 0.0166,
      "positive_teacher_pairs": 27371.7,
      "query_h_norm_ratio": 0.41476339995861056,
      "relation_loss": 0.0018836261006072164,
      "selected_pairs": 27371.7,
      "step": 28850,
      "student_distance_mean": 22.59493293762207,
      "teacher_distance_mean": 0.8524950206279754,
      "valid_action_samples": 234.4,
      "weighted_auxiliary_loss": 0.0018836261006072164,
      "weighted_relation_loss": 0.0018836261006072164
    },
    {
      "action_query_batch_std": 0.34541028141975405,
      "action_query_norm": 40.640839385986325,
      "action_relation_loss": 0.0018629076657816768,
      "auxiliary_loss_weight": 1.0,
      "epoch": 26.997193638914872,
      "flow_matching_loss": 0.015833723079413175,
      "grad_norm": 0.08083994686603546,
      "h_stock_valid_norm": 98.11019744873047,
      "invalid_tail_samples": 24.3,
      "learning_rate": 3.9495601804913627e-07,
      "loss": 0.0177,
      "positive_teacher_pairs": 26738.6,
      "query_h_norm_ratio": 0.41423766016960145,
      "relation_loss": 0.0018629076657816768,
      "selected_pairs": 26738.6,
      "step": 28860,
      "student_distance_mean": 22.625975799560546,
      "teacher_distance_mean": 0.8561433792114258,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0018629076657816768,
      "weighted_relation_loss": 0.0018629076657816768
    },
    {
      "action_query_batch_std": 0.34460754990577697,
      "action_query_norm": 40.632373046875,
      "action_relation_loss": 0.0021587097900919614,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.006548175865294,
      "flow_matching_loss": 0.016028029564768075,
      "grad_norm": 0.1005183532834053,
      "h_stock_valid_norm": 97.85454711914062,
      "invalid_tail_samples": 21.6,
      "learning_rate": 3.880723053673652e-07,
      "loss": 0.0182,
      "positive_teacher_pairs": 24389.7,
      "query_h_norm_ratio": 0.41523620784282683,
      "relation_loss": 0.0021587097900919614,
      "selected_pairs": 24389.7,
      "step": 28870,
      "student_distance_mean": 22.65613079071045,
      "teacher_distance_mean": 0.8552198231220245,
      "valid_action_samples": 214.5,
      "weighted_auxiliary_loss": 0.0021587097900919614,
      "weighted_relation_loss": 0.0021587097900919614
    },
    {
      "action_query_batch_std": 0.34186090230941774,
      "action_query_norm": 40.62527313232422,
      "action_relation_loss": 0.0018808958353474735,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.015902712815716,
      "flow_matching_loss": 0.015948797203600408,
      "grad_norm": 0.0995945855975151,
      "h_stock_valid_norm": 98.06333389282227,
      "invalid_tail_samples": 24.8,
      "learning_rate": 3.812488758558386e-07,
      "loss": 0.0178,
      "positive_teacher_pairs": 26616.7,
      "query_h_norm_ratio": 0.414279505610466,
      "relation_loss": 0.0018808958353474735,
      "selected_pairs": 26616.7,
      "step": 28880,
      "student_distance_mean": 22.318446159362793,
      "teacher_distance_mean": 0.8430625140666962,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0018808958353474735,
      "weighted_relation_loss": 0.0018808958353474735
    },
    {
      "action_query_batch_std": 0.3439019381999969,
      "action_query_norm": 40.64079360961914,
      "action_relation_loss": 0.0018402515561319888,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.02525724976614,
      "flow_matching_loss": 0.015665987599641086,
      "grad_norm": 0.104092538356781,
      "h_stock_valid_norm": 97.92191925048829,
      "invalid_tail_samples": 28.6,
      "learning_rate": 3.744857378056743e-07,
      "loss": 0.0175,
      "positive_teacher_pairs": 25756.0,
      "query_h_norm_ratio": 0.41503359377384186,
      "relation_loss": 0.0018402515561319888,
      "selected_pairs": 25756.0,
      "step": 28890,
      "student_distance_mean": 22.676917457580565,
      "teacher_distance_mean": 0.8566320896148681,
      "valid_action_samples": 227.4,
      "weighted_auxiliary_loss": 0.0018402515561319888,
      "weighted_relation_loss": 0.0018402515561319888
    },
    {
      "action_query_batch_std": 0.33854755759239197,
      "action_query_norm": 40.609423828125,
      "action_relation_loss": 0.001979296328499913,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.034611786716557,
      "flow_matching_loss": 0.015547097846865653,
      "grad_norm": 0.1310240626335144,
      "h_stock_valid_norm": 97.8894271850586,
      "invalid_tail_samples": 21.9,
      "learning_rate": 3.677828994347154e-07,
      "loss": 0.0175,
      "positive_teacher_pairs": 27298.1,
      "query_h_norm_ratio": 0.4148521155118942,
      "relation_loss": 0.001979296328499913,
      "selected_pairs": 27298.1,
      "step": 28900,
      "student_distance_mean": 22.1340913772583,
      "teacher_distance_mean": 0.8365091979503632,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.001979296328499913,
      "weighted_relation_loss": 0.001979296328499913
    },
    {
      "action_query_batch_std": 0.34357792139053345,
      "action_query_norm": 40.63214225769043,
      "action_relation_loss": 0.0019661158206872643,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.04396632366698,
      "flow_matching_loss": 0.015272471122443676,
      "grad_norm": 0.10781548172235489,
      "h_stock_valid_norm": 97.85607604980468,
      "invalid_tail_samples": 24.0,
      "learning_rate": 3.61140368887547e-07,
      "loss": 0.0172,
      "positive_teacher_pairs": 26803.4,
      "query_h_norm_ratio": 0.41522630453109743,
      "relation_loss": 0.0019661158206872643,
      "selected_pairs": 26803.4,
      "step": 28910,
      "student_distance_mean": 22.57343578338623,
      "teacher_distance_mean": 0.8535025835037231,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0019661158206872643,
      "weighted_relation_loss": 0.0019661158206872643
    },
    {
      "action_query_batch_std": 0.34367462396621706,
      "action_query_norm": 40.63922348022461,
      "action_relation_loss": 0.0018725581583566963,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.0533208606174,
      "flow_matching_loss": 0.01559868110343814,
      "grad_norm": 0.10463182628154755,
      "h_stock_valid_norm": 98.12180786132812,
      "invalid_tail_samples": 24.5,
      "learning_rate": 3.545581542354681e-07,
      "loss": 0.0175,
      "positive_teacher_pairs": 26695.0,
      "query_h_norm_ratio": 0.41417274475097654,
      "relation_loss": 0.0018725581583566963,
      "selected_pairs": 26695.0,
      "step": 28920,
      "student_distance_mean": 22.613194084167482,
      "teacher_distance_mean": 0.8545563578605652,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0018725581583566963,
      "weighted_relation_loss": 0.0018725581583566963
    },
    {
      "action_query_batch_std": 0.347243657708168,
      "action_query_norm": 40.65318336486816,
      "action_relation_loss": 0.001853154646232724,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.06267539756782,
      "flow_matching_loss": 0.015711168199777602,
      "grad_norm": 0.14855904877185822,
      "h_stock_valid_norm": 97.92442092895507,
      "invalid_tail_samples": 24.1,
      "learning_rate": 3.480362634764922e-07,
      "loss": 0.0176,
      "positive_teacher_pairs": 26777.4,
      "query_h_norm_ratio": 0.4151511549949646,
      "relation_loss": 0.001853154646232724,
      "selected_pairs": 26777.4,
      "step": 28930,
      "student_distance_mean": 22.68706111907959,
      "teacher_distance_mean": 0.856304270029068,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.001853154646232724,
      "weighted_relation_loss": 0.001853154646232724
    },
    {
      "action_query_batch_std": 0.34479987919330596,
      "action_query_norm": 40.656550598144534,
      "action_relation_loss": 0.0019218557863496244,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.07202993451824,
      "flow_matching_loss": 0.014479051157832146,
      "grad_norm": 0.11092808842658997,
      "h_stock_valid_norm": 98.12738647460938,
      "invalid_tail_samples": 23.9,
      "learning_rate": 3.4157470453533014e-07,
      "loss": 0.0164,
      "positive_teacher_pairs": 26833.5,
      "query_h_norm_ratio": 0.4143251031637192,
      "relation_loss": 0.0019218557863496244,
      "selected_pairs": 26833.5,
      "step": 28940,
      "student_distance_mean": 22.6735502243042,
      "teacher_distance_mean": 0.8565811932086944,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0019218557863496244,
      "weighted_relation_loss": 0.0019218557863496244
    },
    {
      "action_query_batch_std": 0.34565172493457796,
      "action_query_norm": 40.66336097717285,
      "action_relation_loss": 0.0018351394333876669,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.081384471468663,
      "flow_matching_loss": 0.01588859800249338,
      "grad_norm": 0.08019496500492096,
      "h_stock_valid_norm": 98.03194656372071,
      "invalid_tail_samples": 23.7,
      "learning_rate": 3.3517348526339034e-07,
      "loss": 0.0177,
      "positive_teacher_pairs": 26879.9,
      "query_h_norm_ratio": 0.4147986888885498,
      "relation_loss": 0.0018351394333876669,
      "selected_pairs": 26879.9,
      "step": 28950,
      "student_distance_mean": 22.64489116668701,
      "teacher_distance_mean": 0.8534647464752197,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0018351394333876669,
      "weighted_relation_loss": 0.0018351394333876669
    },
    {
      "action_query_batch_std": 0.34354802072048185,
      "action_query_norm": 40.64484214782715,
      "action_relation_loss": 0.001891432120464742,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.09073900841908,
      "flow_matching_loss": 0.016040339320898055,
      "grad_norm": 0.08834520727396011,
      "h_stock_valid_norm": 97.70837478637695,
      "invalid_tail_samples": 24.5,
      "learning_rate": 3.288326134387454e-07,
      "loss": 0.0179,
      "positive_teacher_pairs": 26700.1,
      "query_h_norm_ratio": 0.41598367094993594,
      "relation_loss": 0.001891432120464742,
      "selected_pairs": 26700.1,
      "step": 28960,
      "student_distance_mean": 22.54352149963379,
      "teacher_distance_mean": 0.8520882070064545,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.001891432120464742,
      "weighted_relation_loss": 0.001891432120464742
    },
    {
      "action_query_batch_std": 0.3464349746704102,
      "action_query_norm": 40.66092872619629,
      "action_relation_loss": 0.0018565622274763882,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.100093545369504,
      "flow_matching_loss": 0.016239600628614424,
      "grad_norm": 0.16345736384391785,
      "h_stock_valid_norm": 98.02452697753907,
      "invalid_tail_samples": 27.8,
      "learning_rate": 3.225520967661655e-07,
      "loss": 0.0181,
      "positive_teacher_pairs": 25933.8,
      "query_h_norm_ratio": 0.4148072749376297,
      "relation_loss": 0.0018565622274763882,
      "selected_pairs": 25933.8,
      "step": 28970,
      "student_distance_mean": 22.75191459655762,
      "teacher_distance_mean": 0.8593230485916138,
      "valid_action_samples": 228.2,
      "weighted_auxiliary_loss": 0.0018565622274763882,
      "weighted_relation_loss": 0.0018565622274763882
    },
    {
      "action_query_batch_std": 0.34467473328113557,
      "action_query_norm": 40.63914947509765,
      "action_relation_loss": 0.0019133068853989244,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.109448082319926,
      "flow_matching_loss": 0.014060433767735959,
      "grad_norm": 0.11201897263526917,
      "h_stock_valid_norm": 97.97116470336914,
      "invalid_tail_samples": 23.8,
      "learning_rate": 3.163319428770628e-07,
      "loss": 0.016,
      "positive_teacher_pairs": 26861.3,
      "query_h_norm_ratio": 0.4148118019104004,
      "relation_loss": 0.0019133068853989244,
      "selected_pairs": 26861.3,
      "step": 28980,
      "student_distance_mean": 22.63652095794678,
      "teacher_distance_mean": 0.8542381167411804,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0019133068853989244,
      "weighted_relation_loss": 0.0019133068853989244
    },
    {
      "action_query_batch_std": 0.3438397705554962,
      "action_query_norm": 40.62997322082519,
      "action_relation_loss": 0.0019361914717592299,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.118802619270348,
      "flow_matching_loss": 0.01668684035539627,
      "grad_norm": 0.09445277601480484,
      "h_stock_valid_norm": 97.89226608276367,
      "invalid_tail_samples": 28.0,
      "learning_rate": 3.1017215932951374e-07,
      "loss": 0.0186,
      "positive_teacher_pairs": 25896.5,
      "query_h_norm_ratio": 0.4150496631860733,
      "relation_loss": 0.0019361914717592299,
      "selected_pairs": 25896.5,
      "step": 28990,
      "student_distance_mean": 22.630508804321288,
      "teacher_distance_mean": 0.854913306236267,
      "valid_action_samples": 228.0,
      "weighted_auxiliary_loss": 0.0019361914717592299,
      "weighted_relation_loss": 0.0019361914717592299
    },
    {
      "action_query_batch_std": 0.34856729209423065,
      "action_query_norm": 40.68796081542969,
      "action_relation_loss": 0.0018278689356520772,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.128157156220766,
      "flow_matching_loss": 0.01677013598382473,
      "grad_norm": 0.09909837692975998,
      "h_stock_valid_norm": 97.96884460449219,
      "invalid_tail_samples": 23.1,
      "learning_rate": 3.040727536082366e-07,
      "loss": 0.0186,
      "positive_teacher_pairs": 27011.8,
      "query_h_norm_ratio": 0.4153183400630951,
      "relation_loss": 0.0018278689356520772,
      "selected_pairs": 27011.8,
      "step": 29000,
      "student_distance_mean": 22.88143844604492,
      "teacher_distance_mean": 0.8644352555274963,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0018278689356520772,
      "weighted_relation_loss": 0.0018278689356520772
    },
    {
      "action_query_batch_std": 0.34917200803756715,
      "action_query_norm": 40.693105697631836,
      "action_relation_loss": 0.0018571405671536922,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.137511693171188,
      "flow_matching_loss": 0.01360905459150672,
      "grad_norm": 0.09030556678771973,
      "h_stock_valid_norm": 97.91791915893555,
      "invalid_tail_samples": 24.5,
      "learning_rate": 2.980337331245864e-07,
      "loss": 0.0155,
      "positive_teacher_pairs": 26690.9,
      "query_h_norm_ratio": 0.41558708250522614,
      "relation_loss": 0.0018571405671536922,
      "selected_pairs": 26690.9,
      "step": 29010,
      "student_distance_mean": 22.917662048339842,
      "teacher_distance_mean": 0.8662551820278168,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0018571405671536922,
      "weighted_relation_loss": 0.0018571405671536922
    },
    {
      "action_query_batch_std": 0.34628286957740784,
      "action_query_norm": 40.6735294342041,
      "action_relation_loss": 0.0018796101096086204,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.14686623012161,
      "flow_matching_loss": 0.01563995238393545,
      "grad_norm": 0.09162434935569763,
      "h_stock_valid_norm": 98.05583038330079,
      "invalid_tail_samples": 25.7,
      "learning_rate": 2.9205510521653213e-07,
      "loss": 0.0175,
      "positive_teacher_pairs": 26415.2,
      "query_h_norm_ratio": 0.41480283737182616,
      "relation_loss": 0.0018796101096086204,
      "selected_pairs": 26415.2,
      "step": 29020,
      "student_distance_mean": 22.773803329467775,
      "teacher_distance_mean": 0.8598312318325043,
      "valid_action_samples": 230.3,
      "weighted_auxiliary_loss": 0.0018796101096086204,
      "weighted_relation_loss": 0.0018796101096086204
    },
    {
      "action_query_batch_std": 0.3470546960830688,
      "action_query_norm": 40.696499252319335,
      "action_relation_loss": 0.0019472264917567371,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.15622076707203,
      "flow_matching_loss": 0.015824859403073786,
      "grad_norm": 0.11352900415658951,
      "h_stock_valid_norm": 98.04772567749023,
      "invalid_tail_samples": 22.1,
      "learning_rate": 2.86136877148685e-07,
      "loss": 0.0178,
      "positive_teacher_pairs": 27251.0,
      "query_h_norm_ratio": 0.4150695651769638,
      "relation_loss": 0.0019472264917567371,
      "selected_pairs": 27251.0,
      "step": 29030,
      "student_distance_mean": 22.835115814208983,
      "teacher_distance_mean": 0.863145262002945,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0019472264917567371,
      "weighted_relation_loss": 0.0019472264917567371
    },
    {
      "action_query_batch_std": 0.3495421320199966,
      "action_query_norm": 40.695330429077146,
      "action_relation_loss": 0.0019135259557515384,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.16557530402245,
      "flow_matching_loss": 0.015503323916345835,
      "grad_norm": 0.11992103606462479,
      "h_stock_valid_norm": 98.15159072875977,
      "invalid_tail_samples": 22.1,
      "learning_rate": 2.8027905611223704e-07,
      "loss": 0.0174,
      "positive_teacher_pairs": 27244.7,
      "query_h_norm_ratio": 0.41461859941482543,
      "relation_loss": 0.0019135259557515384,
      "selected_pairs": 27244.7,
      "step": 29040,
      "student_distance_mean": 22.904652404785157,
      "teacher_distance_mean": 0.8641137659549714,
      "valid_action_samples": 233.9,
      "weighted_auxiliary_loss": 0.0019135259557515384,
      "weighted_relation_loss": 0.0019135259557515384
    },
    {
      "action_query_batch_std": 0.347022944688797,
      "action_query_norm": 40.66873054504394,
      "action_relation_loss": 0.0018463653977960348,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.174929840972872,
      "flow_matching_loss": 0.015405636001378297,
      "grad_norm": 0.11089178919792175,
      "h_stock_valid_norm": 97.91759948730468,
      "invalid_tail_samples": 23.7,
      "learning_rate": 2.7448164922500573e-07,
      "loss": 0.0173,
      "positive_teacher_pairs": 26881.1,
      "query_h_norm_ratio": 0.4153394937515259,
      "relation_loss": 0.0018463653977960348,
      "selected_pairs": 26881.1,
      "step": 29050,
      "student_distance_mean": 22.752031135559083,
      "teacher_distance_mean": 0.8603982567787171,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0018463653977960348,
      "weighted_relation_loss": 0.0018463653977960348
    },
    {
      "action_query_batch_std": 0.34686723053455354,
      "action_query_norm": 40.671691513061525,
      "action_relation_loss": 0.0018626237288117409,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.184284377923294,
      "flow_matching_loss": 0.015897920075803994,
      "grad_norm": 0.11868295073509216,
      "h_stock_valid_norm": 97.94962005615234,
      "invalid_tail_samples": 24.3,
      "learning_rate": 2.687446635313784e-07,
      "loss": 0.0178,
      "positive_teacher_pairs": 26736.4,
      "query_h_norm_ratio": 0.4152320623397827,
      "relation_loss": 0.0018626237288117409,
      "selected_pairs": 26736.4,
      "step": 29060,
      "student_distance_mean": 22.82668991088867,
      "teacher_distance_mean": 0.8619286835193634,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0018626237288117409,
      "weighted_relation_loss": 0.0018626237288117409
    },
    {
      "action_query_batch_std": 0.3486321419477463,
      "action_query_norm": 40.68278121948242,
      "action_relation_loss": 0.0018292788416147232,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.193638914873713,
      "flow_matching_loss": 0.015738558769226075,
      "grad_norm": 0.07909587025642395,
      "h_stock_valid_norm": 97.8567497253418,
      "invalid_tail_samples": 23.3,
      "learning_rate": 2.630681060023343e-07,
      "loss": 0.0176,
      "positive_teacher_pairs": 26973.5,
      "query_h_norm_ratio": 0.41574103236198423,
      "relation_loss": 0.0018292788416147232,
      "selected_pairs": 26973.5,
      "step": 29070,
      "student_distance_mean": 22.982150650024415,
      "teacher_distance_mean": 0.8680238962173462,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0018292788416147232,
      "weighted_relation_loss": 0.0018292788416147232
    },
    {
      "action_query_batch_std": 0.3505590885877609,
      "action_query_norm": 40.70806121826172,
      "action_relation_loss": 0.0019096489064395427,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.202993451824135,
      "flow_matching_loss": 0.016116888727992773,
      "grad_norm": 0.10822704434394836,
      "h_stock_valid_norm": 97.84561462402344,
      "invalid_tail_samples": 21.9,
      "learning_rate": 2.5745198353542833e-07,
      "loss": 0.018,
      "positive_teacher_pairs": 27291.9,
      "query_h_norm_ratio": 0.41604718565940857,
      "relation_loss": 0.0019096489064395427,
      "selected_pairs": 27291.9,
      "step": 29080,
      "student_distance_mean": 22.89402484893799,
      "teacher_distance_mean": 0.8654094934463501,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.0019096489064395427,
      "weighted_relation_loss": 0.0019096489064395427
    },
    {
      "action_query_batch_std": 0.3396812528371811,
      "action_query_norm": 40.61867218017578,
      "action_relation_loss": 0.0018895306740887464,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.212347988774557,
      "flow_matching_loss": 0.013606001157313585,
      "grad_norm": 0.09149152040481567,
      "h_stock_valid_norm": 97.90608825683594,
      "invalid_tail_samples": 23.9,
      "learning_rate": 2.518963029547794e-07,
      "loss": 0.0155,
      "positive_teacher_pairs": 26821.9,
      "query_h_norm_ratio": 0.41487516164779664,
      "relation_loss": 0.0018895306740887464,
      "selected_pairs": 26821.9,
      "step": 29090,
      "student_distance_mean": 22.196088600158692,
      "teacher_distance_mean": 0.8390437543392182,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0018895306740887464,
      "weighted_relation_loss": 0.0018895306740887464
    },
    {
      "action_query_batch_std": 0.3472948223352432,
      "action_query_norm": 40.68120155334473,
      "action_relation_loss": 0.0018573880661278963,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.221702525724975,
      "flow_matching_loss": 0.013761207927018404,
      "grad_norm": 0.10074593871831894,
      "h_stock_valid_norm": 97.91483917236329,
      "invalid_tail_samples": 24.5,
      "learning_rate": 2.464010710110598e-07,
      "loss": 0.0156,
      "positive_teacher_pairs": 26685.8,
      "query_h_norm_ratio": 0.4154788017272949,
      "relation_loss": 0.0018573880661278963,
      "selected_pairs": 26685.8,
      "step": 29100,
      "student_distance_mean": 22.85686511993408,
      "teacher_distance_mean": 0.8634848415851593,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.0018573880661278963,
      "weighted_relation_loss": 0.0018573880661278963
    },
    {
      "action_query_batch_std": 0.3409978747367859,
      "action_query_norm": 40.64611740112305,
      "action_relation_loss": 0.0018237916403450072,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.231057062675397,
      "flow_matching_loss": 0.015179847180843354,
      "grad_norm": 0.12238840758800507,
      "h_stock_valid_norm": 98.00913543701172,
      "invalid_tail_samples": 22.8,
      "learning_rate": 2.4096629438150054e-07,
      "loss": 0.017,
      "positive_teacher_pairs": 27080.2,
      "query_h_norm_ratio": 0.41472187936306,
      "relation_loss": 0.0018237916403450072,
      "selected_pairs": 27080.2,
      "step": 29110,
      "student_distance_mean": 22.43241195678711,
      "teacher_distance_mean": 0.8477736115455627,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0018237916403450072,
      "weighted_relation_loss": 0.0018237916403450072
    },
    {
      "action_query_batch_std": 0.34533463716506957,
      "action_query_norm": 40.67427215576172,
      "action_relation_loss": 0.0019125235616229475,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.24041159962582,
      "flow_matching_loss": 0.0161127969622612,
      "grad_norm": 0.1415853649377823,
      "h_stock_valid_norm": 97.8981330871582,
      "invalid_tail_samples": 24.6,
      "learning_rate": 2.3559197966985802e-07,
      "loss": 0.018,
      "positive_teacher_pairs": 26675.5,
      "query_h_norm_ratio": 0.4154775500297546,
      "relation_loss": 0.0019125235616229475,
      "selected_pairs": 26675.5,
      "step": 29120,
      "student_distance_mean": 22.578128623962403,
      "teacher_distance_mean": 0.8520877957344055,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0019125235616229475,
      "weighted_relation_loss": 0.0019125235616229475
    },
    {
      "action_query_batch_std": 0.34893330931663513,
      "action_query_norm": 40.69233512878418,
      "action_relation_loss": 0.0018833115813322366,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.24976613657624,
      "flow_matching_loss": 0.014709905441850423,
      "grad_norm": 0.08246158808469772,
      "h_stock_valid_norm": 98.02689361572266,
      "invalid_tail_samples": 22.0,
      "learning_rate": 2.3027813340644188e-07,
      "loss": 0.0166,
      "positive_teacher_pairs": 27268.6,
      "query_h_norm_ratio": 0.41511522233486176,
      "relation_loss": 0.0018833115813322366,
      "selected_pairs": 27268.6,
      "step": 29130,
      "student_distance_mean": 22.730307006835936,
      "teacher_distance_mean": 0.8581798732280731,
      "valid_action_samples": 234.0,
      "weighted_auxiliary_loss": 0.0018833115813322366,
      "weighted_relation_loss": 0.0018833115813322366
    },
    {
      "action_query_batch_std": 0.3465124428272247,
      "action_query_norm": 40.67977409362793,
      "action_relation_loss": 0.0019219961832277477,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.25912067352666,
      "flow_matching_loss": 0.01651056418195367,
      "grad_norm": 0.08889183402061462,
      "h_stock_valid_norm": 97.93893966674804,
      "invalid_tail_samples": 21.1,
      "learning_rate": 2.2502476204807055e-07,
      "loss": 0.0184,
      "positive_teacher_pairs": 27477.1,
      "query_h_norm_ratio": 0.4153610795736313,
      "relation_loss": 0.0019219961832277477,
      "selected_pairs": 27477.1,
      "step": 29140,
      "student_distance_mean": 22.697016716003418,
      "teacher_distance_mean": 0.8564531624317169,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.0019219961832277477,
      "weighted_relation_loss": 0.0019219961832277477
    },
    {
      "action_query_batch_std": 0.3390541046857834,
      "action_query_norm": 40.604949188232425,
      "action_relation_loss": 0.0018725877278484404,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.26847521047708,
      "flow_matching_loss": 0.014729502331465482,
      "grad_norm": 0.11859191954135895,
      "h_stock_valid_norm": 98.1622413635254,
      "invalid_tail_samples": 25.3,
      "learning_rate": 2.1983187197808786e-07,
      "loss": 0.0166,
      "positive_teacher_pairs": 26501.6,
      "query_h_norm_ratio": 0.41365564465522764,
      "relation_loss": 0.0018725877278484404,
      "selected_pairs": 26501.6,
      "step": 29150,
      "student_distance_mean": 22.253194427490236,
      "teacher_distance_mean": 0.8411959111690521,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.0018725877278484404,
      "weighted_relation_loss": 0.0018725877278484404
    },
    {
      "action_query_batch_std": 0.34415082931518554,
      "action_query_norm": 40.65971412658691,
      "action_relation_loss": 0.001980255590751767,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.277829747427504,
      "flow_matching_loss": 0.014222207292914391,
      "grad_norm": 0.07007003575563431,
      "h_stock_valid_norm": 97.97340927124023,
      "invalid_tail_samples": 24.6,
      "learning_rate": 2.1469946950634644e-07,
      "loss": 0.0162,
      "positive_teacher_pairs": 26661.5,
      "query_h_norm_ratio": 0.41501160264015197,
      "relation_loss": 0.001980255590751767,
      "selected_pairs": 26661.5,
      "step": 29160,
      "student_distance_mean": 22.593722343444824,
      "teacher_distance_mean": 0.8515038430690766,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.001980255590751767,
      "weighted_relation_loss": 0.001980255590751767
    },
    {
      "action_query_batch_std": 0.34411849081516266,
      "action_query_norm": 40.65220756530762,
      "action_relation_loss": 0.0019076082506217062,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.287184284377922,
      "flow_matching_loss": 0.016868685185909272,
      "grad_norm": 0.09193229675292969,
      "h_stock_valid_norm": 98.00992736816406,
      "invalid_tail_samples": 23.9,
      "learning_rate": 2.096275608691911e-07,
      "loss": 0.0188,
      "positive_teacher_pairs": 26831.2,
      "query_h_norm_ratio": 0.41477958858013153,
      "relation_loss": 0.0019076082506217062,
      "selected_pairs": 26831.2,
      "step": 29170,
      "student_distance_mean": 22.575246810913086,
      "teacher_distance_mean": 0.8531807482242584,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0019076082506217062,
      "weighted_relation_loss": 0.0019076082506217062
    },
    {
      "action_query_batch_std": 0.34715802073478697,
      "action_query_norm": 40.65991134643555,
      "action_relation_loss": 0.0019088752684183418,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.296538821328344,
      "flow_matching_loss": 0.01441800119355321,
      "grad_norm": 0.11997219920158386,
      "h_stock_valid_norm": 98.06072082519532,
      "invalid_tail_samples": 25.5,
      "learning_rate": 2.046161522294754e-07,
      "loss": 0.0163,
      "positive_teacher_pairs": 26462.4,
      "query_h_norm_ratio": 0.41464313566684724,
      "relation_loss": 0.0019088752684183418,
      "selected_pairs": 26462.4,
      "step": 29180,
      "student_distance_mean": 22.75815486907959,
      "teacher_distance_mean": 0.8605956912040711,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.0019088752684183418,
      "weighted_relation_loss": 0.0019088752684183418
    },
    {
      "action_query_batch_std": 0.34577700197696687,
      "action_query_norm": 40.66952972412109,
      "action_relation_loss": 0.0018966165254823864,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.305893358278766,
      "flow_matching_loss": 0.01376621499657631,
      "grad_norm": 0.09629211574792862,
      "h_stock_valid_norm": 97.97730026245117,
      "invalid_tail_samples": 22.8,
      "learning_rate": 1.9966524967653944e-07,
      "loss": 0.0157,
      "positive_teacher_pairs": 27080.7,
      "query_h_norm_ratio": 0.415091735124588,
      "relation_loss": 0.0018966165254823864,
      "selected_pairs": 27080.7,
      "step": 29190,
      "student_distance_mean": 22.66723766326904,
      "teacher_distance_mean": 0.8558631181716919,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.0018966165254823864,
      "weighted_relation_loss": 0.0018966165254823864
    },
    {
      "action_query_batch_std": 0.3452859163284302,
      "action_query_norm": 40.66131477355957,
      "action_relation_loss": 0.0018660909030586482,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.315247895229184,
      "flow_matching_loss": 0.015636601578444244,
      "grad_norm": 0.08484654873609543,
      "h_stock_valid_norm": 97.97376556396485,
      "invalid_tail_samples": 23.5,
      "learning_rate": 1.9477485922618222e-07,
      "loss": 0.0175,
      "positive_teacher_pairs": 26930.0,
      "query_h_norm_ratio": 0.4150238364934921,
      "relation_loss": 0.0018660909030586482,
      "selected_pairs": 26930.0,
      "step": 29200,
      "student_distance_mean": 22.65492343902588,
      "teacher_distance_mean": 0.8562844872474671,
      "valid_action_samples": 232.5,
      "weighted_auxiliary_loss": 0.0018660909030586482,
      "weighted_relation_loss": 0.0018660909030586482
    },
    {
      "action_query_batch_std": 0.3462629348039627,
      "action_query_norm": 40.6757396697998,
      "action_relation_loss": 0.001961381791625172,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.324602432179606,
      "flow_matching_loss": 0.014584483299404382,
      "grad_norm": 0.08812137693166733,
      "h_stock_valid_norm": 98.05711669921875,
      "invalid_tail_samples": 22.2,
      "learning_rate": 1.899449868207004e-07,
      "loss": 0.0165,
      "positive_teacher_pairs": 27219.7,
      "query_h_norm_ratio": 0.4148189127445221,
      "relation_loss": 0.001961381791625172,
      "selected_pairs": 27219.7,
      "step": 29210,
      "student_distance_mean": 22.675248527526854,
      "teacher_distance_mean": 0.8564445734024048,
      "valid_action_samples": 233.8,
      "weighted_auxiliary_loss": 0.001961381791625172,
      "weighted_relation_loss": 0.001961381791625172
    },
    {
      "action_query_batch_std": 0.34405345618724825,
      "action_query_norm": 40.66012306213379,
      "action_relation_loss": 0.0018982224282808602,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.33395696913003,
      "flow_matching_loss": 0.014772637840360403,
      "grad_norm": 0.12768732011318207,
      "h_stock_valid_norm": 97.96184158325195,
      "invalid_tail_samples": 23.1,
      "learning_rate": 1.851756383288439e-07,
      "loss": 0.0167,
      "positive_teacher_pairs": 27020.5,
      "query_h_norm_ratio": 0.4150661021471024,
      "relation_loss": 0.0018982224282808602,
      "selected_pairs": 27020.5,
      "step": 29220,
      "student_distance_mean": 22.576612854003905,
      "teacher_distance_mean": 0.8518693149089813,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0018982224282808602,
      "weighted_relation_loss": 0.0018982224282808602
    },
    {
      "action_query_batch_std": 0.34831074476242063,
      "action_query_norm": 40.687330627441405,
      "action_relation_loss": 0.0018685352755710483,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.34331150608045,
      "flow_matching_loss": 0.014269799739122391,
      "grad_norm": 0.12730716168880463,
      "h_stock_valid_norm": 97.88787612915038,
      "invalid_tail_samples": 24.8,
      "learning_rate": 1.8046681954581035e-07,
      "loss": 0.0161,
      "positive_teacher_pairs": 26618.3,
      "query_h_norm_ratio": 0.4156564801931381,
      "relation_loss": 0.0018685352755710483,
      "selected_pairs": 26618.3,
      "step": 29230,
      "student_distance_mean": 22.780019187927245,
      "teacher_distance_mean": 0.8604408800601959,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0018685352755710483,
      "weighted_relation_loss": 0.0018685352755710483
    },
    {
      "action_query_batch_std": 0.3436389923095703,
      "action_query_norm": 40.654354858398435,
      "action_relation_loss": 0.00188288907520473,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.35266604303087,
      "flow_matching_loss": 0.014554287679493428,
      "grad_norm": 0.07211238145828247,
      "h_stock_valid_norm": 97.7963966369629,
      "invalid_tail_samples": 24.2,
      "learning_rate": 1.7581853619327294e-07,
      "loss": 0.0164,
      "positive_teacher_pairs": 26754.3,
      "query_h_norm_ratio": 0.4157053142786026,
      "relation_loss": 0.00188288907520473,
      "selected_pairs": 26754.3,
      "step": 29240,
      "student_distance_mean": 22.70914783477783,
      "teacher_distance_mean": 0.8573582291603088,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.00188288907520473,
      "weighted_relation_loss": 0.00188288907520473
    },
    {
      "action_query_batch_std": 0.34308737218379975,
      "action_query_norm": 40.65414619445801,
      "action_relation_loss": 0.0019028794253244997,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.36202057998129,
      "flow_matching_loss": 0.01445101834833622,
      "grad_norm": 0.07579538971185684,
      "h_stock_valid_norm": 97.81366882324218,
      "invalid_tail_samples": 22.7,
      "learning_rate": 1.7123079391932472e-07,
      "loss": 0.0164,
      "positive_teacher_pairs": 27107.3,
      "query_h_norm_ratio": 0.41563278436660767,
      "relation_loss": 0.0019028794253244997,
      "selected_pairs": 27107.3,
      "step": 29250,
      "student_distance_mean": 22.479141807556154,
      "teacher_distance_mean": 0.8487218499183655,
      "valid_action_samples": 233.3,
      "weighted_auxiliary_loss": 0.0019028794253244997,
      "weighted_relation_loss": 0.0019028794253244997
    },
    {
      "action_query_batch_std": 0.3472332537174225,
      "action_query_norm": 40.65368385314942,
      "action_relation_loss": 0.0018830609973520041,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.371375116931713,
      "flow_matching_loss": 0.014727553352713585,
      "grad_norm": 0.1131291314959526,
      "h_stock_valid_norm": 97.94776611328125,
      "invalid_tail_samples": 23.4,
      "learning_rate": 1.6670359829850657e-07,
      "loss": 0.0166,
      "positive_teacher_pairs": 26942.3,
      "query_h_norm_ratio": 0.41505615711212157,
      "relation_loss": 0.0018830609973520041,
      "selected_pairs": 26942.3,
      "step": 29260,
      "student_distance_mean": 22.86010456085205,
      "teacher_distance_mean": 0.8628111481666565,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0018830609973520041,
      "weighted_relation_loss": 0.0018830609973520041
    },
    {
      "action_query_batch_std": 0.3436510652303696,
      "action_query_norm": 40.652780532836914,
      "action_relation_loss": 0.0018747822963632644,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.38072965388213,
      "flow_matching_loss": 0.014257754199206829,
      "grad_norm": 0.07244822382926941,
      "h_stock_valid_norm": 97.96731643676758,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.6223695483179037e-07,
      "loss": 0.0161,
      "positive_teacher_pairs": 26880.1,
      "query_h_norm_ratio": 0.4149647504091263,
      "relation_loss": 0.0018747822963632644,
      "selected_pairs": 26880.1,
      "step": 29270,
      "student_distance_mean": 22.54087390899658,
      "teacher_distance_mean": 0.8519785165786743,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0018747822963632644,
      "weighted_relation_loss": 0.0018747822963632644
    },
    {
      "action_query_batch_std": 0.34231407642364503,
      "action_query_norm": 40.623675155639646,
      "action_relation_loss": 0.0019463393138721586,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.390084190832553,
      "flow_matching_loss": 0.016363874170929195,
      "grad_norm": 0.11192338913679123,
      "h_stock_valid_norm": 97.90382080078125,
      "invalid_tail_samples": 23.3,
      "learning_rate": 1.5783086894656795e-07,
      "loss": 0.0183,
      "positive_teacher_pairs": 26976.1,
      "query_h_norm_ratio": 0.4149360477924347,
      "relation_loss": 0.0019463393138721586,
      "selected_pairs": 26976.1,
      "step": 29280,
      "student_distance_mean": 22.562565422058107,
      "teacher_distance_mean": 0.8515621483325958,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0019463393138721586,
      "weighted_relation_loss": 0.0019463393138721586
    },
    {
      "action_query_batch_std": 0.3503440052270889,
      "action_query_norm": 40.69161682128906,
      "action_relation_loss": 0.0018011744949035346,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.399438727782975,
      "flow_matching_loss": 0.015720237977802753,
      "grad_norm": 0.09547322243452072,
      "h_stock_valid_norm": 98.10037155151367,
      "invalid_tail_samples": 27.4,
      "learning_rate": 1.5348534599665121e-07,
      "loss": 0.0175,
      "positive_teacher_pairs": 26027.0,
      "query_h_norm_ratio": 0.41479878723621366,
      "relation_loss": 0.0018011744949035346,
      "selected_pairs": 26027.0,
      "step": 29290,
      "student_distance_mean": 23.051828384399414,
      "teacher_distance_mean": 0.8710382163524628,
      "valid_action_samples": 228.6,
      "weighted_auxiliary_loss": 0.0018011744949035346,
      "weighted_relation_loss": 0.0018011744949035346
    },
    {
      "action_query_batch_std": 0.34615626335144045,
      "action_query_norm": 40.648846817016604,
      "action_relation_loss": 0.001883204048499465,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.408793264733397,
      "flow_matching_loss": 0.014463899284601211,
      "grad_norm": 0.10128185153007507,
      "h_stock_valid_norm": 97.97330245971679,
      "invalid_tail_samples": 23.0,
      "learning_rate": 1.4920039126225527e-07,
      "loss": 0.0163,
      "positive_teacher_pairs": 27033.8,
      "query_h_norm_ratio": 0.41489960849285124,
      "relation_loss": 0.001883204048499465,
      "selected_pairs": 27033.8,
      "step": 29300,
      "student_distance_mean": 22.60948600769043,
      "teacher_distance_mean": 0.8530788421630859,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.001883204048499465,
      "weighted_relation_loss": 0.001883204048499465
    },
    {
      "action_query_batch_std": 0.34499778151512145,
      "action_query_norm": 40.67483978271484,
      "action_relation_loss": 0.0019191202358342708,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.418147801683816,
      "flow_matching_loss": 0.015069871209561825,
      "grad_norm": 0.07576362043619156,
      "h_stock_valid_norm": 97.91285095214843,
      "invalid_tail_samples": 25.8,
      "learning_rate": 1.449760099500097e-07,
      "loss": 0.017,
      "positive_teacher_pairs": 26390.8,
      "query_h_norm_ratio": 0.4154191553592682,
      "relation_loss": 0.0019191202358342708,
      "selected_pairs": 26390.8,
      "step": 29310,
      "student_distance_mean": 22.683979988098145,
      "teacher_distance_mean": 0.8567094385623932,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.0019191202358342708,
      "weighted_relation_loss": 0.0019191202358342708
    },
    {
      "action_query_batch_std": 0.34298027455806734,
      "action_query_norm": 40.63434066772461,
      "action_relation_loss": 0.0019140788936056197,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.427502338634238,
      "flow_matching_loss": 0.016049053985625505,
      "grad_norm": 0.0705486536026001,
      "h_stock_valid_norm": 97.88547592163086,
      "invalid_tail_samples": 22.3,
      "learning_rate": 1.4081220719293075e-07,
      "loss": 0.018,
      "positive_teacher_pairs": 27198.1,
      "query_h_norm_ratio": 0.41512286365032197,
      "relation_loss": 0.0019140788936056197,
      "selected_pairs": 27198.1,
      "step": 29320,
      "student_distance_mean": 22.529771614074708,
      "teacher_distance_mean": 0.8508605480194091,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0019140788936056197,
      "weighted_relation_loss": 0.0019140788936056197
    },
    {
      "action_query_batch_std": 0.34639482796192167,
      "action_query_norm": 40.66377754211426,
      "action_relation_loss": 0.001911115401890129,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.43685687558466,
      "flow_matching_loss": 0.014045700524002313,
      "grad_norm": 0.10904230922460556,
      "h_stock_valid_norm": 97.77324447631835,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.3670898805043797e-07,
      "loss": 0.016,
      "positive_teacher_pairs": 26898.8,
      "query_h_norm_ratio": 0.4159016817808151,
      "relation_loss": 0.001911115401890129,
      "selected_pairs": 26898.8,
      "step": 29330,
      "student_distance_mean": 22.710323143005372,
      "teacher_distance_mean": 0.857498687505722,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.001911115401890129,
      "weighted_relation_loss": 0.001911115401890129
    },
    {
      "action_query_batch_std": 0.3399424910545349,
      "action_query_norm": 40.619275665283205,
      "action_relation_loss": 0.0019441307988017798,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.446211412535078,
      "flow_matching_loss": 0.015274824760854244,
      "grad_norm": 0.08579704165458679,
      "h_stock_valid_norm": 98.12149810791016,
      "invalid_tail_samples": 23.4,
      "learning_rate": 1.326663575083209e-07,
      "loss": 0.0172,
      "positive_teacher_pairs": 26940.5,
      "query_h_norm_ratio": 0.41397200524806976,
      "relation_loss": 0.0019441307988017798,
      "selected_pairs": 26940.5,
      "step": 29340,
      "student_distance_mean": 22.40209217071533,
      "teacher_distance_mean": 0.8461718201637268,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0019441307988017798,
      "weighted_relation_loss": 0.0019441307988017798
    },
    {
      "action_query_batch_std": 0.34753248989582064,
      "action_query_norm": 40.66921615600586,
      "action_relation_loss": 0.0019615036668255924,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.4555659494855,
      "flow_matching_loss": 0.016663917619735003,
      "grad_norm": 0.15675929188728333,
      "h_stock_valid_norm": 97.95839767456054,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.2868432047876688e-07,
      "loss": 0.0186,
      "positive_teacher_pairs": 26902.1,
      "query_h_norm_ratio": 0.4151718318462372,
      "relation_loss": 0.0019615036668255924,
      "selected_pairs": 26902.1,
      "step": 29350,
      "student_distance_mean": 22.738264274597167,
      "teacher_distance_mean": 0.859024441242218,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0019615036668255924,
      "weighted_relation_loss": 0.0019615036668255924
    },
    {
      "action_query_batch_std": 0.3469414979219437,
      "action_query_norm": 40.676482391357425,
      "action_relation_loss": 0.0018434613128192722,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.464920486435922,
      "flow_matching_loss": 0.016151506174355745,
      "grad_norm": 0.08532185852527618,
      "h_stock_valid_norm": 97.89861373901367,
      "invalid_tail_samples": 22.6,
      "learning_rate": 1.2476288180032213e-07,
      "loss": 0.018,
      "positive_teacher_pairs": 27131.4,
      "query_h_norm_ratio": 0.4154975563287735,
      "relation_loss": 0.0018434613128192722,
      "selected_pairs": 27131.4,
      "step": 29360,
      "student_distance_mean": 22.8145450592041,
      "teacher_distance_mean": 0.861458820104599,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.0018434613128192722,
      "weighted_relation_loss": 0.0018434613128192722
    },
    {
      "action_query_batch_std": 0.3427894562482834,
      "action_query_norm": 40.64257621765137,
      "action_relation_loss": 0.00196800654521212,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.474275023386344,
      "flow_matching_loss": 0.014945304114371538,
      "grad_norm": 0.13183170557022095,
      "h_stock_valid_norm": 97.93910217285156,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.209020462379029e-07,
      "loss": 0.0169,
      "positive_teacher_pairs": 26826.4,
      "query_h_norm_ratio": 0.414984455704689,
      "relation_loss": 0.00196800654521212,
      "selected_pairs": 26826.4,
      "step": 29370,
      "student_distance_mean": 22.58943290710449,
      "teacher_distance_mean": 0.8529768109321594,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.00196800654521212,
      "weighted_relation_loss": 0.00196800654521212
    },
    {
      "action_query_batch_std": 0.3482159286737442,
      "action_query_norm": 40.67287483215332,
      "action_relation_loss": 0.00189761808142066,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.483629560336762,
      "flow_matching_loss": 0.014542557112872601,
      "grad_norm": 0.09207186102867126,
      "h_stock_valid_norm": 97.97195739746094,
      "invalid_tail_samples": 21.4,
      "learning_rate": 1.1710181848278435e-07,
      "loss": 0.0164,
      "positive_teacher_pairs": 27408.3,
      "query_h_norm_ratio": 0.4151515781879425,
      "relation_loss": 0.00189761808142066,
      "selected_pairs": 27408.3,
      "step": 29380,
      "student_distance_mean": 22.760995292663573,
      "teacher_distance_mean": 0.8587408721446991,
      "valid_action_samples": 234.6,
      "weighted_auxiliary_loss": 0.00189761808142066,
      "weighted_relation_loss": 0.00189761808142066
    },
    {
      "action_query_batch_std": 0.3436916023492813,
      "action_query_norm": 40.63163032531738,
      "action_relation_loss": 0.0018803074373863638,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.492984097287184,
      "flow_matching_loss": 0.014176939427852631,
      "grad_norm": 0.13963009417057037,
      "h_stock_valid_norm": 97.96538696289062,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.133622031526116e-07,
      "loss": 0.0161,
      "positive_teacher_pairs": 26802.8,
      "query_h_norm_ratio": 0.41475614309310915,
      "relation_loss": 0.0018803074373863638,
      "selected_pairs": 26802.8,
      "step": 29390,
      "student_distance_mean": 22.583003807067872,
      "teacher_distance_mean": 0.8533502757549286,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0018803074373863638,
      "weighted_relation_loss": 0.0018803074373863638
    },
    {
      "action_query_batch_std": 0.34913446605205534,
      "action_query_norm": 40.68053359985352,
      "action_relation_loss": 0.001909118506591767,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.502338634237606,
      "flow_matching_loss": 0.014202397782355547,
      "grad_norm": 0.10457102954387665,
      "h_stock_valid_norm": 97.89719314575196,
      "invalid_tail_samples": 23.9,
      "learning_rate": 1.0968320479136095e-07,
      "loss": 0.0161,
      "positive_teacher_pairs": 26826.5,
      "query_h_norm_ratio": 0.41554802656173706,
      "relation_loss": 0.001909118506591767,
      "selected_pairs": 26826.5,
      "step": 29400,
      "student_distance_mean": 22.805739974975587,
      "teacher_distance_mean": 0.86236732006073,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.001909118506591767,
      "weighted_relation_loss": 0.001909118506591767
    },
    {
      "action_query_batch_std": 0.34400720298290255,
      "action_query_norm": 40.64277687072754,
      "action_relation_loss": 0.0019415586022660137,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.511693171188025,
      "flow_matching_loss": 0.015488557144999504,
      "grad_norm": 0.12286809086799622,
      "h_stock_valid_norm": 98.10943222045898,
      "invalid_tail_samples": 23.7,
      "learning_rate": 1.0606482786936767e-07,
      "loss": 0.0174,
      "positive_teacher_pairs": 26880.6,
      "query_h_norm_ratio": 0.41426109373569486,
      "relation_loss": 0.0019415586022660137,
      "selected_pairs": 26880.6,
      "step": 29410,
      "student_distance_mean": 22.580808258056642,
      "teacher_distance_mean": 0.8527771949768066,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0019415586022660137,
      "weighted_relation_loss": 0.0019415586022660137
    },
    {
      "action_query_batch_std": 0.3445529967546463,
      "action_query_norm": 40.672664642333984,
      "action_relation_loss": 0.0018951500416733325,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.521047708138447,
      "flow_matching_loss": 0.016306945867836476,
      "grad_norm": 0.0720420703291893,
      "h_stock_valid_norm": 97.86284484863282,
      "invalid_tail_samples": 26.6,
      "learning_rate": 1.0250707678329808e-07,
      "loss": 0.0182,
      "positive_teacher_pairs": 26209.2,
      "query_h_norm_ratio": 0.4156105726957321,
      "relation_loss": 0.0018951500416733325,
      "selected_pairs": 26209.2,
      "step": 29420,
      "student_distance_mean": 22.668224334716797,
      "teacher_distance_mean": 0.8558191239833832,
      "valid_action_samples": 229.4,
      "weighted_auxiliary_loss": 0.0018951500416733325,
      "weighted_relation_loss": 0.0018951500416733325
    },
    {
      "action_query_batch_std": 0.3436621636152267,
      "action_query_norm": 40.647134399414064,
      "action_relation_loss": 0.0019209865597076714,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.53040224508887,
      "flow_matching_loss": 0.015794390719383954,
      "grad_norm": 0.095148965716362,
      "h_stock_valid_norm": 98.00468826293945,
      "invalid_tail_samples": 24.0,
      "learning_rate": 9.900995585615525e-08,
      "loss": 0.0177,
      "positive_teacher_pairs": 26810.7,
      "query_h_norm_ratio": 0.41475037634372713,
      "relation_loss": 0.0019209865597076714,
      "selected_pairs": 26810.7,
      "step": 29430,
      "student_distance_mean": 22.601186180114745,
      "teacher_distance_mean": 0.8537345647811889,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0019209865597076714,
      "weighted_relation_loss": 0.0019209865597076714
    },
    {
      "action_query_batch_std": 0.34488971829414367,
      "action_query_norm": 40.66242370605469,
      "action_relation_loss": 0.001860419160220772,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.53975678203929,
      "flow_matching_loss": 0.01622067978605628,
      "grad_norm": 0.08303358405828476,
      "h_stock_valid_norm": 97.9116943359375,
      "invalid_tail_samples": 25.5,
      "learning_rate": 9.55734693372734e-08,
      "loss": 0.0181,
      "positive_teacher_pairs": 26461.8,
      "query_h_norm_ratio": 0.41529797911643984,
      "relation_loss": 0.001860419160220772,
      "selected_pairs": 26461.8,
      "step": 29440,
      "student_distance_mean": 22.672687339782716,
      "teacher_distance_mean": 0.8558795809745788,
      "valid_action_samples": 230.5,
      "weighted_auxiliary_loss": 0.001860419160220772,
      "weighted_relation_loss": 0.001860419160220772
    },
    {
      "action_query_batch_std": 0.34422576129436494,
      "action_query_norm": 40.63670272827149,
      "action_relation_loss": 0.0019491795101203024,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.54911131898971,
      "flow_matching_loss": 0.014627442508935929,
      "grad_norm": 0.09627814590930939,
      "h_stock_valid_norm": 97.93863449096679,
      "invalid_tail_samples": 24.1,
      "learning_rate": 9.219762140231236e-08,
      "loss": 0.0166,
      "positive_teacher_pairs": 26781.2,
      "query_h_norm_ratio": 0.4149222493171692,
      "relation_loss": 0.0019491795101203024,
      "selected_pairs": 26781.2,
      "step": 29450,
      "student_distance_mean": 22.547056198120117,
      "teacher_distance_mean": 0.8515130281448364,
      "valid_action_samples": 231.9,
      "weighted_auxiliary_loss": 0.0019491795101203024,
      "weighted_relation_loss": 0.0019491795101203024
    },
    {
      "action_query_batch_std": 0.33889256715774535,
      "action_query_norm": 40.61437339782715,
      "action_relation_loss": 0.00195836138445884,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.55846585594013,
      "flow_matching_loss": 0.015133362542837859,
      "grad_norm": 0.14883717894554138,
      "h_stock_valid_norm": 97.90145568847656,
      "invalid_tail_samples": 22.9,
      "learning_rate": 8.888241615322978e-08,
      "loss": 0.0171,
      "positive_teacher_pairs": 27061.8,
      "query_h_norm_ratio": 0.4148519515991211,
      "relation_loss": 0.00195836138445884,
      "selected_pairs": 27061.8,
      "step": 29460,
      "student_distance_mean": 22.21526527404785,
      "teacher_distance_mean": 0.837987071275711,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.00195836138445884,
      "weighted_relation_loss": 0.00195836138445884
    },
    {
      "action_query_batch_std": 0.34451437294483184,
      "action_query_norm": 40.64777946472168,
      "action_relation_loss": 0.0018971132347360253,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.567820392890553,
      "flow_matching_loss": 0.015634803008288144,
      "grad_norm": 0.09464804083108902,
      "h_stock_valid_norm": 97.92290802001953,
      "invalid_tail_samples": 25.0,
      "learning_rate": 8.562785761833114e-08,
      "loss": 0.0175,
      "positive_teacher_pairs": 26577.7,
      "query_h_norm_ratio": 0.4151025414466858,
      "relation_loss": 0.0018971132347360253,
      "selected_pairs": 26577.7,
      "step": 29470,
      "student_distance_mean": 22.55251235961914,
      "teacher_distance_mean": 0.8519108235836029,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0018971132347360253,
      "weighted_relation_loss": 0.0018971132347360253
    },
    {
      "action_query_batch_std": 0.34456769526004793,
      "action_query_norm": 40.66160659790039,
      "action_relation_loss": 0.0018816502182744444,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.57717492984097,
      "flow_matching_loss": 0.013889002054929734,
      "grad_norm": 0.08356713503599167,
      "h_stock_valid_norm": 97.75729217529297,
      "invalid_tail_samples": 21.7,
      "learning_rate": 8.243394975219753e-08,
      "loss": 0.0158,
      "positive_teacher_pairs": 27336.7,
      "query_h_norm_ratio": 0.4159461289644241,
      "relation_loss": 0.0018816502182744444,
      "selected_pairs": 27336.7,
      "step": 29480,
      "student_distance_mean": 22.543397521972658,
      "teacher_distance_mean": 0.8514366269111633,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0018816502182744444,
      "weighted_relation_loss": 0.0018816502182744444
    },
    {
      "action_query_batch_std": 0.3408293157815933,
      "action_query_norm": 40.654687118530276,
      "action_relation_loss": 0.0019298817845992745,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.586529466791394,
      "flow_matching_loss": 0.016235123481601478,
      "grad_norm": 0.11878552287817001,
      "h_stock_valid_norm": 97.92191085815429,
      "invalid_tail_samples": 21.7,
      "learning_rate": 7.930069643573568e-08,
      "loss": 0.0182,
      "positive_teacher_pairs": 27339.0,
      "query_h_norm_ratio": 0.41517542004585267,
      "relation_loss": 0.0019298817845992745,
      "selected_pairs": 27339.0,
      "step": 29490,
      "student_distance_mean": 22.343710899353027,
      "teacher_distance_mean": 0.8442297041416168,
      "valid_action_samples": 234.3,
      "weighted_auxiliary_loss": 0.0019298817845992745,
      "weighted_relation_loss": 0.0019298817845992745
    },
    {
      "action_query_batch_std": 0.34144236147403717,
      "action_query_norm": 40.61867408752441,
      "action_relation_loss": 0.0019786521326750515,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.595884003741816,
      "flow_matching_loss": 0.015551713295280933,
      "grad_norm": 0.09820044040679932,
      "h_stock_valid_norm": 97.8517692565918,
      "invalid_tail_samples": 24.3,
      "learning_rate": 7.622810147614456e-08,
      "loss": 0.0175,
      "positive_teacher_pairs": 26743.2,
      "query_h_norm_ratio": 0.4151056528091431,
      "relation_loss": 0.0019786521326750515,
      "selected_pairs": 26743.2,
      "step": 29500,
      "student_distance_mean": 22.46250801086426,
      "teacher_distance_mean": 0.849542647600174,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0019786521326750515,
      "weighted_relation_loss": 0.0019786521326750515
    },
    {
      "action_query_batch_std": 0.3413820803165436,
      "action_query_norm": 40.64550247192383,
      "action_relation_loss": 0.0019299813313409687,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.605238540692234,
      "flow_matching_loss": 0.01486161993816495,
      "grad_norm": 0.0976485162973404,
      "h_stock_valid_norm": 97.97163391113281,
      "invalid_tail_samples": 23.0,
      "learning_rate": 7.321616860690995e-08,
      "loss": 0.0168,
      "positive_teacher_pairs": 27046.2,
      "query_h_norm_ratio": 0.41487115025520327,
      "relation_loss": 0.0019299813313409687,
      "selected_pairs": 27046.2,
      "step": 29510,
      "student_distance_mean": 22.361762619018556,
      "teacher_distance_mean": 0.8433170437812805,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.0019299813313409687,
      "weighted_relation_loss": 0.0019299813313409687
    },
    {
      "action_query_batch_std": 0.3396588534116745,
      "action_query_norm": 40.59709129333496,
      "action_relation_loss": 0.001934686943423003,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.614593077642656,
      "flow_matching_loss": 0.01644692812114954,
      "grad_norm": 0.10232068598270416,
      "h_stock_valid_norm": 98.02604370117187,
      "invalid_tail_samples": 24.5,
      "learning_rate": 7.026490148782095e-08,
      "loss": 0.0184,
      "positive_teacher_pairs": 26688.2,
      "query_h_norm_ratio": 0.4141474485397339,
      "relation_loss": 0.001934686943423003,
      "selected_pairs": 26688.2,
      "step": 29520,
      "student_distance_mean": 22.37163562774658,
      "teacher_distance_mean": 0.8443889021873474,
      "valid_action_samples": 231.5,
      "weighted_auxiliary_loss": 0.001934686943423003,
      "weighted_relation_loss": 0.001934686943423003
    },
    {
      "action_query_batch_std": 0.34351706802845,
      "action_query_norm": 40.65930404663086,
      "action_relation_loss": 0.0019292107550427317,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.623947614593078,
      "flow_matching_loss": 0.015268615167587995,
      "grad_norm": 0.0890459343791008,
      "h_stock_valid_norm": 97.81620483398437,
      "invalid_tail_samples": 21.9,
      "learning_rate": 6.737430370494236e-08,
      "loss": 0.0172,
      "positive_teacher_pairs": 27292.7,
      "query_h_norm_ratio": 0.41567343175411225,
      "relation_loss": 0.0019292107550427317,
      "selected_pairs": 27292.7,
      "step": 29530,
      "student_distance_mean": 22.573793601989745,
      "teacher_distance_mean": 0.8519434452056884,
      "valid_action_samples": 234.1,
      "weighted_auxiliary_loss": 0.0019292107550427317,
      "weighted_relation_loss": 0.0019292107550427317
    },
    {
      "action_query_batch_std": 0.3476811170578003,
      "action_query_norm": 40.67724533081055,
      "action_relation_loss": 0.0018742718617431819,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.6333021515435,
      "flow_matching_loss": 0.015893175080418587,
      "grad_norm": 0.13273902237415314,
      "h_stock_valid_norm": 97.97667922973633,
      "invalid_tail_samples": 25.6,
      "learning_rate": 6.454437877062569e-08,
      "loss": 0.0178,
      "positive_teacher_pairs": 26433.9,
      "query_h_norm_ratio": 0.41517719626426697,
      "relation_loss": 0.0018742718617431819,
      "selected_pairs": 26433.9,
      "step": 29540,
      "student_distance_mean": 22.92485580444336,
      "teacher_distance_mean": 0.8651415288448334,
      "valid_action_samples": 230.4,
      "weighted_auxiliary_loss": 0.0018742718617431819,
      "weighted_relation_loss": 0.0018742718617431819
    },
    {
      "action_query_batch_std": 0.34413357079029083,
      "action_query_norm": 40.64608535766602,
      "action_relation_loss": 0.0018965233583003282,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.64265668849392,
      "flow_matching_loss": 0.014999378751963378,
      "grad_norm": 0.08326408267021179,
      "h_stock_valid_norm": 97.93367309570313,
      "invalid_tail_samples": 23.2,
      "learning_rate": 6.177513012349256e-08,
      "loss": 0.0169,
      "positive_teacher_pairs": 26987.9,
      "query_h_norm_ratio": 0.415038651227951,
      "relation_loss": 0.0018965233583003282,
      "selected_pairs": 26987.9,
      "step": 29550,
      "student_distance_mean": 22.637903022766114,
      "teacher_distance_mean": 0.8556109368801117,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0018965233583003282,
      "weighted_relation_loss": 0.0018965233583003282
    },
    {
      "action_query_batch_std": 0.3427737891674042,
      "action_query_norm": 40.62767105102539,
      "action_relation_loss": 0.0019084763480350375,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.65201122544434,
      "flow_matching_loss": 0.014588380232453346,
      "grad_norm": 0.1734941154718399,
      "h_stock_valid_norm": 97.80147323608398,
      "invalid_tail_samples": 22.3,
      "learning_rate": 5.9066561128445775e-08,
      "loss": 0.0165,
      "positive_teacher_pairs": 27199.1,
      "query_h_norm_ratio": 0.4154116988182068,
      "relation_loss": 0.0019084763480350375,
      "selected_pairs": 27199.1,
      "step": 29560,
      "student_distance_mean": 22.458441543579102,
      "teacher_distance_mean": 0.8478014767169952,
      "valid_action_samples": 233.7,
      "weighted_auxiliary_loss": 0.0019084763480350375,
      "weighted_relation_loss": 0.0019084763480350375
    },
    {
      "action_query_batch_std": 0.34642029404640196,
      "action_query_norm": 40.6670166015625,
      "action_relation_loss": 0.0018735321937128901,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.661365762394762,
      "flow_matching_loss": 0.015947872307151556,
      "grad_norm": 0.09984904527664185,
      "h_stock_valid_norm": 97.9313949584961,
      "invalid_tail_samples": 23.6,
      "learning_rate": 5.6418675076641556e-08,
      "loss": 0.0178,
      "positive_teacher_pairs": 26895.2,
      "query_h_norm_ratio": 0.41526433229446413,
      "relation_loss": 0.0018735321937128901,
      "selected_pairs": 26895.2,
      "step": 29570,
      "student_distance_mean": 22.692755699157715,
      "teacher_distance_mean": 0.8569369673728943,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.0018735321937128901,
      "weighted_relation_loss": 0.0018735321937128901
    },
    {
      "action_query_batch_std": 0.3504944801330566,
      "action_query_norm": 40.7140510559082,
      "action_relation_loss": 0.0018525893450714649,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.67072029934518,
      "flow_matching_loss": 0.015707601513713598,
      "grad_norm": 0.13590295612812042,
      "h_stock_valid_norm": 98.03883056640625,
      "invalid_tail_samples": 21.1,
      "learning_rate": 5.383147518552845e-08,
      "loss": 0.0176,
      "positive_teacher_pairs": 27481.5,
      "query_h_norm_ratio": 0.415287122130394,
      "relation_loss": 0.0018525893450714649,
      "selected_pairs": 27481.5,
      "step": 29580,
      "student_distance_mean": 22.836252975463868,
      "teacher_distance_mean": 0.8627241432666779,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.0018525893450714649,
      "weighted_relation_loss": 0.0018525893450714649
    },
    {
      "action_query_batch_std": 0.3484870731830597,
      "action_query_norm": 40.67891159057617,
      "action_relation_loss": 0.0018222335958853364,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.680074836295603,
      "flow_matching_loss": 0.01524875983595848,
      "grad_norm": 0.08319798111915588,
      "h_stock_valid_norm": 98.02233505249023,
      "invalid_tail_samples": 25.0,
      "learning_rate": 5.1304964598786196e-08,
      "loss": 0.0171,
      "positive_teacher_pairs": 26572.2,
      "query_h_norm_ratio": 0.414999321103096,
      "relation_loss": 0.0018222335958853364,
      "selected_pairs": 26572.2,
      "step": 29590,
      "student_distance_mean": 22.89048042297363,
      "teacher_distance_mean": 0.8636237859725953,
      "valid_action_samples": 231.0,
      "weighted_auxiliary_loss": 0.0018222335958853364,
      "weighted_relation_loss": 0.0018222335958853364
    },
    {
      "action_query_batch_std": 0.3416209578514099,
      "action_query_norm": 40.647560119628906,
      "action_relation_loss": 0.0018605856574140489,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.689429373246025,
      "flow_matching_loss": 0.014837147481739522,
      "grad_norm": 0.12262097746133804,
      "h_stock_valid_norm": 97.97297897338868,
      "invalid_tail_samples": 25.4,
      "learning_rate": 4.883914638636467e-08,
      "loss": 0.0167,
      "positive_teacher_pairs": 26477.0,
      "query_h_norm_ratio": 0.41488643288612365,
      "relation_loss": 0.0018605856574140489,
      "selected_pairs": 26477.0,
      "step": 29600,
      "student_distance_mean": 22.458487701416015,
      "teacher_distance_mean": 0.8489549458026886,
      "valid_action_samples": 230.6,
      "weighted_auxiliary_loss": 0.0018605856574140489,
      "weighted_relation_loss": 0.0018605856574140489
    },
    {
      "action_query_batch_std": 0.3446433514356613,
      "action_query_norm": 40.65571174621582,
      "action_relation_loss": 0.001840895030181855,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.698783910196447,
      "flow_matching_loss": 0.01456485940143466,
      "grad_norm": 0.07192174345254898,
      "h_stock_valid_norm": 97.93967132568359,
      "invalid_tail_samples": 22.8,
      "learning_rate": 4.643402354446713e-08,
      "loss": 0.0164,
      "positive_teacher_pairs": 27080.7,
      "query_h_norm_ratio": 0.415111181139946,
      "relation_loss": 0.001840895030181855,
      "selected_pairs": 27080.7,
      "step": 29610,
      "student_distance_mean": 22.624752044677734,
      "teacher_distance_mean": 0.8545102655887604,
      "valid_action_samples": 233.2,
      "weighted_auxiliary_loss": 0.001840895030181855,
      "weighted_relation_loss": 0.001840895030181855
    },
    {
      "action_query_batch_std": 0.3479538679122925,
      "action_query_norm": 40.683675384521486,
      "action_relation_loss": 0.0018972587422467768,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.708138447146865,
      "flow_matching_loss": 0.015926542412489652,
      "grad_norm": 0.10251117497682571,
      "h_stock_valid_norm": 97.94406661987304,
      "invalid_tail_samples": 23.1,
      "learning_rate": 4.4089598995544766e-08,
      "loss": 0.0178,
      "positive_teacher_pairs": 27015.6,
      "query_h_norm_ratio": 0.41537941694259645,
      "relation_loss": 0.0018972587422467768,
      "selected_pairs": 27015.6,
      "step": 29620,
      "student_distance_mean": 22.76917209625244,
      "teacher_distance_mean": 0.8605853974819183,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.0018972587422467768,
      "weighted_relation_loss": 0.0018972587422467768
    },
    {
      "action_query_batch_std": 0.34540699124336244,
      "action_query_norm": 40.6515453338623,
      "action_relation_loss": 0.001899091841187328,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.717492984097287,
      "flow_matching_loss": 0.013878767937421798,
      "grad_norm": 0.07887306809425354,
      "h_stock_valid_norm": 97.84138946533203,
      "invalid_tail_samples": 23.7,
      "learning_rate": 4.180587558829663e-08,
      "loss": 0.0158,
      "positive_teacher_pairs": 26874.9,
      "query_h_norm_ratio": 0.4154874473810196,
      "relation_loss": 0.001899091841187328,
      "selected_pairs": 26874.9,
      "step": 29630,
      "student_distance_mean": 22.678337860107423,
      "teacher_distance_mean": 0.8570815920829773,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.001899091841187328,
      "weighted_relation_loss": 0.001899091841187328
    },
    {
      "action_query_batch_std": 0.34321253597736356,
      "action_query_norm": 40.639597702026364,
      "action_relation_loss": 0.0019170716288499535,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.72684752104771,
      "flow_matching_loss": 0.016370092425495385,
      "grad_norm": 0.1352366954088211,
      "h_stock_valid_norm": 97.7402359008789,
      "invalid_tail_samples": 23.9,
      "learning_rate": 3.958285609765855e-08,
      "loss": 0.0183,
      "positive_teacher_pairs": 26825.2,
      "query_h_norm_ratio": 0.4157953649759293,
      "relation_loss": 0.0019170716288499535,
      "selected_pairs": 26825.2,
      "step": 29640,
      "student_distance_mean": 22.582297706604002,
      "teacher_distance_mean": 0.8526329517364502,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.0019170716288499535,
      "weighted_relation_loss": 0.0019170716288499535
    },
    {
      "action_query_batch_std": 0.3447964549064636,
      "action_query_norm": 40.66926155090332,
      "action_relation_loss": 0.0019148939405567943,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.736202057998128,
      "flow_matching_loss": 0.014921601209789515,
      "grad_norm": 0.19796009361743927,
      "h_stock_valid_norm": 97.85824508666992,
      "invalid_tail_samples": 24.3,
      "learning_rate": 3.74205432248087e-08,
      "loss": 0.0168,
      "positive_teacher_pairs": 26736.3,
      "query_h_norm_ratio": 0.4155951738357544,
      "relation_loss": 0.0019148939405567943,
      "selected_pairs": 26736.3,
      "step": 29650,
      "student_distance_mean": 22.60104465484619,
      "teacher_distance_mean": 0.8519802629947663,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0019148939405567943,
      "weighted_relation_loss": 0.0019148939405567943
    },
    {
      "action_query_batch_std": 0.3387616366147995,
      "action_query_norm": 40.609005737304685,
      "action_relation_loss": 0.0019481114693917334,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.74555659494855,
      "flow_matching_loss": 0.01506903050467372,
      "grad_norm": 0.0971812978386879,
      "h_stock_valid_norm": 97.85083923339843,
      "invalid_tail_samples": 24.2,
      "learning_rate": 3.531893959716204e-08,
      "loss": 0.017,
      "positive_teacher_pairs": 26763.2,
      "query_h_norm_ratio": 0.41501171290874483,
      "relation_loss": 0.0019481114693917334,
      "selected_pairs": 26763.2,
      "step": 29660,
      "student_distance_mean": 22.17511978149414,
      "teacher_distance_mean": 0.8367544889450074,
      "valid_action_samples": 231.8,
      "weighted_auxiliary_loss": 0.0019481114693917334,
      "weighted_relation_loss": 0.0019481114693917334
    },
    {
      "action_query_batch_std": 0.3462922275066376,
      "action_query_norm": 40.663143157958984,
      "action_relation_loss": 0.001938595506362617,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.75491113189897,
      "flow_matching_loss": 0.015968295000493526,
      "grad_norm": 0.07949867844581604,
      "h_stock_valid_norm": 98.07138442993164,
      "invalid_tail_samples": 23.9,
      "learning_rate": 3.327804776837029e-08,
      "loss": 0.0179,
      "positive_teacher_pairs": 26829.4,
      "query_h_norm_ratio": 0.4146290898323059,
      "relation_loss": 0.001938595506362617,
      "selected_pairs": 26829.4,
      "step": 29670,
      "student_distance_mean": 22.771009635925292,
      "teacher_distance_mean": 0.8606666564941406,
      "valid_action_samples": 232.1,
      "weighted_auxiliary_loss": 0.001938595506362617,
      "weighted_relation_loss": 0.001938595506362617
    },
    {
      "action_query_batch_std": 0.34246321618556974,
      "action_query_norm": 40.63968353271484,
      "action_relation_loss": 0.001924604782834649,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.764265668849394,
      "flow_matching_loss": 0.015238958410918713,
      "grad_norm": 0.07936130464076996,
      "h_stock_valid_norm": 97.97270202636719,
      "invalid_tail_samples": 22.6,
      "learning_rate": 3.129787021829977e-08,
      "loss": 0.0172,
      "positive_teacher_pairs": 27131.7,
      "query_h_norm_ratio": 0.41480990052223204,
      "relation_loss": 0.001924604782834649,
      "selected_pairs": 27131.7,
      "step": 29680,
      "student_distance_mean": 22.46383571624756,
      "teacher_distance_mean": 0.8477837264537811,
      "valid_action_samples": 233.4,
      "weighted_auxiliary_loss": 0.001924604782834649,
      "weighted_relation_loss": 0.001924604782834649
    },
    {
      "action_query_batch_std": 0.3410226494073868,
      "action_query_norm": 40.64545211791992,
      "action_relation_loss": 0.0019982542493380607,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.773620205799812,
      "flow_matching_loss": 0.014968566037714482,
      "grad_norm": 0.080302394926548,
      "h_stock_valid_norm": 97.90881271362305,
      "invalid_tail_samples": 23.2,
      "learning_rate": 2.9378409353059133e-08,
      "loss": 0.017,
      "positive_teacher_pairs": 26986.1,
      "query_h_norm_ratio": 0.4151369571685791,
      "relation_loss": 0.0019982542493380607,
      "selected_pairs": 26986.1,
      "step": 29690,
      "student_distance_mean": 22.468463325500487,
      "teacher_distance_mean": 0.8483591854572297,
      "valid_action_samples": 232.8,
      "weighted_auxiliary_loss": 0.0019982542493380607,
      "weighted_relation_loss": 0.0019982542493380607
    },
    {
      "action_query_batch_std": 0.3421488344669342,
      "action_query_norm": 40.64851722717285,
      "action_relation_loss": 0.0018582467804662884,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.782974742750234,
      "flow_matching_loss": 0.016407498624175787,
      "grad_norm": 0.1036340519785881,
      "h_stock_valid_norm": 97.86314239501954,
      "invalid_tail_samples": 22.5,
      "learning_rate": 2.7519667504971593e-08,
      "loss": 0.0183,
      "positive_teacher_pairs": 27147.4,
      "query_h_norm_ratio": 0.4153625875711441,
      "relation_loss": 0.0018582467804662884,
      "selected_pairs": 27147.4,
      "step": 29700,
      "student_distance_mean": 22.389090728759765,
      "teacher_distance_mean": 0.8459338366985321,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0018582467804662884,
      "weighted_relation_loss": 0.0018582467804662884
    },
    {
      "action_query_batch_std": 0.34883522391319277,
      "action_query_norm": 40.67596702575683,
      "action_relation_loss": 0.0018249746062792838,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.792329279700656,
      "flow_matching_loss": 0.015479640010744334,
      "grad_norm": 0.12327123433351517,
      "h_stock_valid_norm": 97.89321060180664,
      "invalid_tail_samples": 24.8,
      "learning_rate": 2.572164693258605e-08,
      "loss": 0.0173,
      "positive_teacher_pairs": 26616.7,
      "query_h_norm_ratio": 0.4155139267444611,
      "relation_loss": 0.0018249746062792838,
      "selected_pairs": 26616.7,
      "step": 29710,
      "student_distance_mean": 22.807992935180664,
      "teacher_distance_mean": 0.8612178981304168,
      "valid_action_samples": 231.2,
      "weighted_auxiliary_loss": 0.0018249746062792838,
      "weighted_relation_loss": 0.0018249746062792838
    },
    {
      "action_query_batch_std": 0.34127688109874726,
      "action_query_norm": 40.63690414428711,
      "action_relation_loss": 0.001940997294150293,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.801683816651074,
      "flow_matching_loss": 0.014669070485979319,
      "grad_norm": 0.09841445833444595,
      "h_stock_valid_norm": 97.91863174438477,
      "invalid_tail_samples": 25.8,
      "learning_rate": 2.3984349820665996e-08,
      "loss": 0.0166,
      "positive_teacher_pairs": 26397.5,
      "query_h_norm_ratio": 0.4150126546621323,
      "relation_loss": 0.001940997294150293,
      "selected_pairs": 26397.5,
      "step": 29720,
      "student_distance_mean": 22.543696022033693,
      "teacher_distance_mean": 0.853032261133194,
      "valid_action_samples": 230.2,
      "weighted_auxiliary_loss": 0.001940997294150293,
      "weighted_relation_loss": 0.001940997294150293
    },
    {
      "action_query_batch_std": 0.3473324865102768,
      "action_query_norm": 40.66519775390625,
      "action_relation_loss": 0.0019036230398342014,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.811038353601496,
      "flow_matching_loss": 0.016228125151246785,
      "grad_norm": 0.07595117390155792,
      "h_stock_valid_norm": 98.01487426757812,
      "invalid_tail_samples": 24.3,
      "learning_rate": 2.2307778280189485e-08,
      "loss": 0.0181,
      "positive_teacher_pairs": 26730.5,
      "query_h_norm_ratio": 0.4148900717496872,
      "relation_loss": 0.0019036230398342014,
      "selected_pairs": 26730.5,
      "step": 29730,
      "student_distance_mean": 22.658201217651367,
      "teacher_distance_mean": 0.8557639122009277,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.0019036230398342014,
      "weighted_relation_loss": 0.0019036230398342014
    },
    {
      "action_query_batch_std": 0.34132593870162964,
      "action_query_norm": 40.63542137145996,
      "action_relation_loss": 0.0018966531846672297,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.82039289055192,
      "flow_matching_loss": 0.014915014524012804,
      "grad_norm": 0.1099834144115448,
      "h_stock_valid_norm": 97.93335494995117,
      "invalid_tail_samples": 23.4,
      "learning_rate": 2.0691934348354704e-08,
      "loss": 0.0168,
      "positive_teacher_pairs": 26941.9,
      "query_h_norm_ratio": 0.41493003368377684,
      "relation_loss": 0.0018966531846672297,
      "selected_pairs": 26941.9,
      "step": 29740,
      "student_distance_mean": 22.472478675842286,
      "teacher_distance_mean": 0.8490984976291657,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0018966531846672297,
      "weighted_relation_loss": 0.0018966531846672297
    },
    {
      "action_query_batch_std": 0.3458348959684372,
      "action_query_norm": 40.659071350097655,
      "action_relation_loss": 0.00193228431744501,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.82974742750234,
      "flow_matching_loss": 0.014316396415233612,
      "grad_norm": 0.09639168530702591,
      "h_stock_valid_norm": 97.92762298583985,
      "invalid_tail_samples": 23.0,
      "learning_rate": 1.9136819988557763e-08,
      "loss": 0.0162,
      "positive_teacher_pairs": 27041.0,
      "query_h_norm_ratio": 0.41519728899002073,
      "relation_loss": 0.00193228431744501,
      "selected_pairs": 27041.0,
      "step": 29750,
      "student_distance_mean": 22.74539909362793,
      "teacher_distance_mean": 0.858797574043274,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.00193228431744501,
      "weighted_relation_loss": 0.00193228431744501
    },
    {
      "action_query_batch_std": 0.3433999627828598,
      "action_query_norm": 40.651140975952146,
      "action_relation_loss": 0.0019174031098373235,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.83910196445276,
      "flow_matching_loss": 0.015391266159713268,
      "grad_norm": 0.0989057868719101,
      "h_stock_valid_norm": 98.03929061889649,
      "invalid_tail_samples": 23.8,
      "learning_rate": 1.7642437090414908e-08,
      "loss": 0.0173,
      "positive_teacher_pairs": 26848.6,
      "query_h_norm_ratio": 0.4146437168121338,
      "relation_loss": 0.0019174031098373235,
      "selected_pairs": 26848.6,
      "step": 29760,
      "student_distance_mean": 22.56676902770996,
      "teacher_distance_mean": 0.8514629423618316,
      "valid_action_samples": 232.2,
      "weighted_auxiliary_loss": 0.0019174031098373235,
      "weighted_relation_loss": 0.0019174031098373235
    },
    {
      "action_query_batch_std": 0.34562754034996035,
      "action_query_norm": 40.65265464782715,
      "action_relation_loss": 0.001946813939139247,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.84845650140318,
      "flow_matching_loss": 0.015184861328452825,
      "grad_norm": 0.08314301073551178,
      "h_stock_valid_norm": 97.96039123535157,
      "invalid_tail_samples": 23.0,
      "learning_rate": 1.6208787469734755e-08,
      "loss": 0.0171,
      "positive_teacher_pairs": 27035.4,
      "query_h_norm_ratio": 0.4149922728538513,
      "relation_loss": 0.001946813939139247,
      "selected_pairs": 27035.4,
      "step": 29770,
      "student_distance_mean": 22.63758430480957,
      "teacher_distance_mean": 0.854973566532135,
      "valid_action_samples": 233.0,
      "weighted_auxiliary_loss": 0.001946813939139247,
      "weighted_relation_loss": 0.001946813939139247
    },
    {
      "action_query_batch_std": 0.3474689245223999,
      "action_query_norm": 40.66626625061035,
      "action_relation_loss": 0.001929922727867961,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.857811038353603,
      "flow_matching_loss": 0.014074531756341458,
      "grad_norm": 0.08499447256326675,
      "h_stock_valid_norm": 98.04492721557617,
      "invalid_tail_samples": 25.3,
      "learning_rate": 1.483587286854604e-08,
      "loss": 0.016,
      "positive_teacher_pairs": 26507.4,
      "query_h_norm_ratio": 0.41477417647838594,
      "relation_loss": 0.001929922727867961,
      "selected_pairs": 26507.4,
      "step": 29780,
      "student_distance_mean": 22.68186511993408,
      "teacher_distance_mean": 0.8578691482543945,
      "valid_action_samples": 230.7,
      "weighted_auxiliary_loss": 0.001929922727867961,
      "weighted_relation_loss": 0.001929922727867961
    },
    {
      "action_query_batch_std": 0.34515107572078707,
      "action_query_norm": 40.65579261779785,
      "action_relation_loss": 0.001879626512527466,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.86716557530402,
      "flow_matching_loss": 0.015283119957894086,
      "grad_norm": 0.09486162662506104,
      "h_stock_valid_norm": 97.89126281738281,
      "invalid_tail_samples": 23.6,
      "learning_rate": 1.3523694955064336e-08,
      "loss": 0.0172,
      "positive_teacher_pairs": 26895.5,
      "query_h_norm_ratio": 0.4153187543153763,
      "relation_loss": 0.001879626512527466,
      "selected_pairs": 26895.5,
      "step": 29790,
      "student_distance_mean": 22.65323600769043,
      "teacher_distance_mean": 0.8555575609207153,
      "valid_action_samples": 232.4,
      "weighted_auxiliary_loss": 0.001879626512527466,
      "weighted_relation_loss": 0.001879626512527466
    },
    {
      "action_query_batch_std": 0.3468663215637207,
      "action_query_norm": 40.682673645019534,
      "action_relation_loss": 0.0018887158599682151,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.876520112254443,
      "flow_matching_loss": 0.014939128141850232,
      "grad_norm": 0.0728188306093216,
      "h_stock_valid_norm": 97.96989517211914,
      "invalid_tail_samples": 21.1,
      "learning_rate": 1.2272255323708681e-08,
      "loss": 0.0168,
      "positive_teacher_pairs": 27479.5,
      "query_h_norm_ratio": 0.4152620941400528,
      "relation_loss": 0.0018887158599682151,
      "selected_pairs": 27479.5,
      "step": 29800,
      "student_distance_mean": 22.80662612915039,
      "teacher_distance_mean": 0.8608908712863922,
      "valid_action_samples": 234.9,
      "weighted_auxiliary_loss": 0.0018887158599682151,
      "weighted_relation_loss": 0.0018887158599682151
    },
    {
      "action_query_batch_std": 0.3447429805994034,
      "action_query_norm": 40.66202735900879,
      "action_relation_loss": 0.0018730093957856297,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.885874649204865,
      "flow_matching_loss": 0.01643076166510582,
      "grad_norm": 0.085627980530262,
      "h_stock_valid_norm": 98.0986946105957,
      "invalid_tail_samples": 22.9,
      "learning_rate": 1.1081555495096042e-08,
      "loss": 0.0183,
      "positive_teacher_pairs": 27056.0,
      "query_h_norm_ratio": 0.4145024836063385,
      "relation_loss": 0.0018730093957856297,
      "selected_pairs": 27056.0,
      "step": 29810,
      "student_distance_mean": 22.64178924560547,
      "teacher_distance_mean": 0.8541914939880371,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0018730093957856297,
      "weighted_relation_loss": 0.0018730093957856297
    },
    {
      "action_query_batch_std": 0.3458325654268265,
      "action_query_norm": 40.66653213500977,
      "action_relation_loss": 0.0018114032689481973,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.895229186155284,
      "flow_matching_loss": 0.01542565757408738,
      "grad_norm": 0.11809809505939484,
      "h_stock_valid_norm": 97.87941360473633,
      "invalid_tail_samples": 24.7,
      "learning_rate": 9.951596916041306e-09,
      "loss": 0.0172,
      "positive_teacher_pairs": 26649.5,
      "query_h_norm_ratio": 0.4154771000146866,
      "relation_loss": 0.0018114032689481973,
      "selected_pairs": 26649.5,
      "step": 29820,
      "student_distance_mean": 22.650362968444824,
      "teacher_distance_mean": 0.855498480796814,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0018114032689481973,
      "weighted_relation_loss": 0.0018114032689481973
    },
    {
      "action_query_batch_std": 0.34288076162338255,
      "action_query_norm": 40.63332176208496,
      "action_relation_loss": 0.0018575912807136774,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.904583723105706,
      "flow_matching_loss": 0.015166407637298106,
      "grad_norm": 0.09898068010807037,
      "h_stock_valid_norm": 98.06187744140625,
      "invalid_tail_samples": 23.4,
      "learning_rate": 8.882380959551739e-09,
      "loss": 0.017,
      "positive_teacher_pairs": 26952.6,
      "query_h_norm_ratio": 0.41436807811260223,
      "relation_loss": 0.0018575912807136774,
      "selected_pairs": 26952.6,
      "step": 29830,
      "student_distance_mean": 22.401179122924805,
      "teacher_distance_mean": 0.8454032599925995,
      "valid_action_samples": 232.6,
      "weighted_auxiliary_loss": 0.0018575912807136774,
      "weighted_relation_loss": 0.0018575912807136774
    },
    {
      "action_query_batch_std": 0.34331752359867096,
      "action_query_norm": 40.644076919555665,
      "action_relation_loss": 0.0018277716822922229,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.913938260056128,
      "flow_matching_loss": 0.014594803471118212,
      "grad_norm": 0.0764366164803505,
      "h_stock_valid_norm": 97.9291389465332,
      "invalid_tail_samples": 23.7,
      "learning_rate": 7.873908924821428e-09,
      "loss": 0.0164,
      "positive_teacher_pairs": 26874.9,
      "query_h_norm_ratio": 0.4150385230779648,
      "relation_loss": 0.0018277716822922229,
      "selected_pairs": 26874.9,
      "step": 29840,
      "student_distance_mean": 22.62922191619873,
      "teacher_distance_mean": 0.854653924703598,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0018277716822922229,
      "weighted_relation_loss": 0.0018277716822922229
    },
    {
      "action_query_batch_std": 0.34608903229236604,
      "action_query_norm": 40.6559383392334,
      "action_relation_loss": 0.0018723787972703577,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.92329279700655,
      "flow_matching_loss": 0.016256843507289887,
      "grad_norm": 0.1130795106291771,
      "h_stock_valid_norm": 97.8930435180664,
      "invalid_tail_samples": 22.9,
      "learning_rate": 6.926182037242379e-09,
      "loss": 0.0181,
      "positive_teacher_pairs": 27057.3,
      "query_h_norm_ratio": 0.4153110980987549,
      "relation_loss": 0.0018723787972703577,
      "selected_pairs": 27057.3,
      "step": 29850,
      "student_distance_mean": 22.595057487487793,
      "teacher_distance_mean": 0.8540675580501557,
      "valid_action_samples": 233.1,
      "weighted_auxiliary_loss": 0.0018723787972703577,
      "weighted_relation_loss": 0.0018723787972703577
    },
    {
      "action_query_batch_std": 0.34733450412750244,
      "action_query_norm": 40.69544296264648,
      "action_relation_loss": 0.0019442847231402993,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.932647333956968,
      "flow_matching_loss": 0.014809890650212765,
      "grad_norm": 0.10375767201185226,
      "h_stock_valid_norm": 97.8969711303711,
      "invalid_tail_samples": 21.0,
      "learning_rate": 6.039201448387876e-09,
      "loss": 0.0168,
      "positive_teacher_pairs": 27502.6,
      "query_h_norm_ratio": 0.4156990647315979,
      "relation_loss": 0.0019442847231402993,
      "selected_pairs": 27502.6,
      "step": 29860,
      "student_distance_mean": 22.750358963012694,
      "teacher_distance_mean": 0.8592040240764618,
      "valid_action_samples": 235.0,
      "weighted_auxiliary_loss": 0.0019442847231402993,
      "weighted_relation_loss": 0.0019442847231402993
    },
    {
      "action_query_batch_std": 0.3419712662696838,
      "action_query_norm": 40.63553733825684,
      "action_relation_loss": 0.001962076302152127,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.94200187090739,
      "flow_matching_loss": 0.01739204963669181,
      "grad_norm": 0.11103624105453491,
      "h_stock_valid_norm": 97.85300827026367,
      "invalid_tail_samples": 20.4,
      "learning_rate": 5.212968236029125e-09,
      "loss": 0.0194,
      "positive_teacher_pairs": 27643.9,
      "query_h_norm_ratio": 0.4152727276086807,
      "relation_loss": 0.001962076302152127,
      "selected_pairs": 27643.9,
      "step": 29870,
      "student_distance_mean": 22.42533435821533,
      "teacher_distance_mean": 0.8466277778148651,
      "valid_action_samples": 235.6,
      "weighted_auxiliary_loss": 0.001962076302152127,
      "weighted_relation_loss": 0.001962076302152127
    },
    {
      "action_query_batch_std": 0.3518016993999481,
      "action_query_norm": 40.70191612243652,
      "action_relation_loss": 0.0018776617012917995,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.951356407857812,
      "flow_matching_loss": 0.015760946460068224,
      "grad_norm": 0.10635631531476974,
      "h_stock_valid_norm": 97.98810577392578,
      "invalid_tail_samples": 23.7,
      "learning_rate": 4.447483404118602e-09,
      "loss": 0.0176,
      "positive_teacher_pairs": 26883.5,
      "query_h_norm_ratio": 0.4153783440589905,
      "relation_loss": 0.0018776617012917995,
      "selected_pairs": 26883.5,
      "step": 29880,
      "student_distance_mean": 23.049982261657714,
      "teacher_distance_mean": 0.8721532940864563,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0018776617012917995,
      "weighted_relation_loss": 0.0018776617012917995
    },
    {
      "action_query_batch_std": 0.3419428765773773,
      "action_query_norm": 40.614508056640624,
      "action_relation_loss": 0.0020870744716376065,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.96071094480823,
      "flow_matching_loss": 0.016750903334468603,
      "grad_norm": 0.10365702956914902,
      "h_stock_valid_norm": 97.94017791748047,
      "invalid_tail_samples": 26.7,
      "learning_rate": 3.742747882784503e-09,
      "loss": 0.0188,
      "positive_teacher_pairs": 26183.9,
      "query_h_norm_ratio": 0.41468835771083834,
      "relation_loss": 0.0020870744716376065,
      "selected_pairs": 26183.9,
      "step": 29890,
      "student_distance_mean": 22.580184745788575,
      "teacher_distance_mean": 0.8529690742492676,
      "valid_action_samples": 229.3,
      "weighted_auxiliary_loss": 0.0020870744716376065,
      "weighted_relation_loss": 0.0020870744716376065
    },
    {
      "action_query_batch_std": 0.34684050381183623,
      "action_query_norm": 40.66546440124512,
      "action_relation_loss": 0.0018911549355834722,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.970065481758652,
      "flow_matching_loss": 0.014647528156638145,
      "grad_norm": 0.08835961669683456,
      "h_stock_valid_norm": 98.01978073120117,
      "invalid_tail_samples": 23.3,
      "learning_rate": 3.0987625283473987e-09,
      "loss": 0.0165,
      "positive_teacher_pairs": 26968.3,
      "query_h_norm_ratio": 0.4148729920387268,
      "relation_loss": 0.0018911549355834722,
      "selected_pairs": 26968.3,
      "step": 29900,
      "student_distance_mean": 22.8602575302124,
      "teacher_distance_mean": 0.8643620312213898,
      "valid_action_samples": 232.7,
      "weighted_auxiliary_loss": 0.0018911549355834722,
      "weighted_relation_loss": 0.0018911549355834722
    },
    {
      "action_query_batch_std": 0.3449513912200928,
      "action_query_norm": 40.65163612365723,
      "action_relation_loss": 0.0019148209481500088,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.979420018709074,
      "flow_matching_loss": 0.015378609579056502,
      "grad_norm": 0.171420618891716,
      "h_stock_valid_norm": 97.84396133422851,
      "invalid_tail_samples": 24.7,
      "learning_rate": 2.515528123320232e-09,
      "loss": 0.0173,
      "positive_teacher_pairs": 26639.2,
      "query_h_norm_ratio": 0.4154774099588394,
      "relation_loss": 0.0019148209481500088,
      "selected_pairs": 26639.2,
      "step": 29910,
      "student_distance_mean": 22.701839447021484,
      "teacher_distance_mean": 0.8590325891971589,
      "valid_action_samples": 231.3,
      "weighted_auxiliary_loss": 0.0019148209481500088,
      "weighted_relation_loss": 0.0019148209481500088
    },
    {
      "action_query_batch_std": 0.3444600135087967,
      "action_query_norm": 40.65915069580078,
      "action_relation_loss": 0.001873454824090004,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.988774555659496,
      "flow_matching_loss": 0.014725546631962062,
      "grad_norm": 0.07826880365610123,
      "h_stock_valid_norm": 98.0759147644043,
      "invalid_tail_samples": 24.3,
      "learning_rate": 1.9930453763750133e-09,
      "loss": 0.0166,
      "positive_teacher_pairs": 26734.2,
      "query_h_norm_ratio": 0.414568755030632,
      "relation_loss": 0.001873454824090004,
      "selected_pairs": 26734.2,
      "step": 29920,
      "student_distance_mean": 22.655963516235353,
      "teacher_distance_mean": 0.8551995635032654,
      "valid_action_samples": 231.7,
      "weighted_auxiliary_loss": 0.001873454824090004,
      "weighted_relation_loss": 0.001873454824090004
    },
    {
      "action_query_batch_std": 0.3452404379844666,
      "action_query_norm": 40.64454879760742,
      "action_relation_loss": 0.0019100556150078774,
      "auxiliary_loss_weight": 1.0,
      "epoch": 27.998129092609915,
      "flow_matching_loss": 0.015586388669908047,
      "grad_norm": 0.10332226008176804,
      "h_stock_valid_norm": 97.9190574645996,
      "invalid_tail_samples": 24.6,
      "learning_rate": 1.5313149223872281e-09,
      "loss": 0.0175,
      "positive_teacher_pairs": 26664.9,
      "query_h_norm_ratio": 0.4150860756635666,
      "relation_loss": 0.0019100556150078774,
      "selected_pairs": 26664.9,
      "step": 29930,
      "student_distance_mean": 22.623486137390138,
      "teacher_distance_mean": 0.8543819487094879,
      "valid_action_samples": 231.4,
      "weighted_auxiliary_loss": 0.0019100556150078774,
      "weighted_relation_loss": 0.0019100556150078774
    },
    {
      "action_query_batch_std": 0.3450734496116638,
      "action_query_norm": 40.67513809204102,
      "action_relation_loss": 0.0019132938352413476,
      "auxiliary_loss_weight": 1.0,
      "epoch": 28.007483629560337,
      "flow_matching_loss": 0.014045293489471079,
      "grad_norm": 0.11602833122015,
      "h_stock_valid_norm": 97.8552146911621,
      "invalid_tail_samples": 20.6,
      "learning_rate": 1.1303373224025305e-09,
      "loss": 0.016,
      "positive_teacher_pairs": 24608.7,
      "query_h_norm_ratio": 0.415667662024498,
      "relation_loss": 0.0019132938352413476,
      "selected_pairs": 24608.7,
      "step": 29940,
      "student_distance_mean": 22.69094753265381,
      "teacher_distance_mean": 0.8572041511535644,
      "valid_action_samples": 215.5,
      "weighted_auxiliary_loss": 0.0019132938352413476,
      "weighted_relation_loss": 0.0019132938352413476
    },
    {
      "action_query_batch_std": 0.3472134619951248,
      "action_query_norm": 40.65249824523926,
      "action_relation_loss": 0.0017799591063521803,
      "auxiliary_loss_weight": 1.0,
      "epoch": 28.01683816651076,
      "flow_matching_loss": 0.014684921130537987,
      "grad_norm": 0.0743785873055458,
      "h_stock_valid_norm": 98.02662200927735,
      "invalid_tail_samples": 25.1,
      "learning_rate": 7.901130636367438e-10,
      "loss": 0.0165,
      "positive_teacher_pairs": 26558.4,
      "query_h_norm_ratio": 0.41470994651317594,
      "relation_loss": 0.0017799591063521803,
      "selected_pairs": 26558.4,
      "step": 29950,
      "student_distance_mean": 22.79310894012451,
      "teacher_distance_mean": 0.8593349933624268,
      "valid_action_samples": 230.9,
      "weighted_auxiliary_loss": 0.0017799591063521803,
      "weighted_relation_loss": 0.0017799591063521803
    },
    {
      "action_query_batch_std": 0.34184554815292356,
      "action_query_norm": 40.629756927490234,
      "action_relation_loss": 0.0019110745517536998,
      "auxiliary_loss_weight": 1.0,
      "epoch": 28.026192703461177,
      "flow_matching_loss": 0.015910063218325377,
      "grad_norm": 0.06901122629642487,
      "h_stock_valid_norm": 97.99076766967774,
      "invalid_tail_samples": 26.5,
      "learning_rate": 5.106425595036158e-10,
      "loss": 0.0178,
      "positive_teacher_pairs": 26233.1,
      "query_h_norm_ratio": 0.4146295964717865,
      "relation_loss": 0.0019110745517536998,
      "selected_pairs": 26233.1,
      "step": 29960,
      "student_distance_mean": 22.375971603393555,
      "teacher_distance_mean": 0.8441360831260681,
      "valid_action_samples": 229.5,
      "weighted_auxiliary_loss": 0.0019110745517536998,
      "weighted_relation_loss": 0.0019110745517536998
    },
    {
      "action_query_batch_std": 0.3426850765943527,
      "action_query_norm": 40.656005859375,
      "action_relation_loss": 0.0018833858193829656,
      "auxiliary_loss_weight": 1.0,
      "epoch": 28.0355472404116,
      "flow_matching_loss": 0.014569173287600278,
      "grad_norm": 0.10439544916152954,
      "h_stock_valid_norm": 97.88118133544921,
      "invalid_tail_samples": 23.7,
      "learning_rate": 2.9192614958706286e-10,
      "loss": 0.0165,
      "positive_teacher_pairs": 26875.3,
      "query_h_norm_ratio": 0.41536247730255127,
      "relation_loss": 0.0018833858193829656,
      "selected_pairs": 26875.3,
      "step": 29970,
      "student_distance_mean": 22.61865863800049,
      "teacher_distance_mean": 0.8534406065940857,
      "valid_action_samples": 232.3,
      "weighted_auxiliary_loss": 0.0018833858193829656,
      "weighted_relation_loss": 0.0018833858193829656
    },
    {
      "action_query_batch_std": 0.34902789890766145,
      "action_query_norm": 40.70429573059082,
      "action_relation_loss": 0.0019043078646063805,
      "auxiliary_loss_weight": 1.0,
      "epoch": 28.04490177736202,
      "flow_matching_loss": 0.015668292809277774,
      "grad_norm": 0.11769145727157593,
      "h_stock_valid_norm": 97.84872131347656,
      "invalid_tail_samples": 24.0,
      "learning_rate": 1.3396409964117064e-10,
      "loss": 0.0176,
      "positive_teacher_pairs": 26808.6,
      "query_h_norm_ratio": 0.41599610149860383,
      "relation_loss": 0.0019043078646063805,
      "selected_pairs": 26808.6,
      "step": 29980,
      "student_distance_mean": 22.981830406188966,
      "teacher_distance_mean": 0.868180763721466,
      "valid_action_samples": 232.0,
      "weighted_auxiliary_loss": 0.0019043078646063805,
      "weighted_relation_loss": 0.0019043078646063805
    },
    {
      "action_query_batch_std": 0.34618253707885743,
      "action_query_norm": 40.66046257019043,
      "action_relation_loss": 0.0018781326478347182,
      "auxiliary_loss_weight": 1.0,
      "epoch": 28.054256314312443,
      "flow_matching_loss": 0.01418017726391554,
      "grad_norm": 0.11643343418836594,
      "h_stock_valid_norm": 97.97476043701172,
      "invalid_tail_samples": 22.5,
      "learning_rate": 3.6756601606846574e-11,
      "loss": 0.0161,
      "positive_teacher_pairs": 27151.3,
      "query_h_norm_ratio": 0.4150122910737991,
      "relation_loss": 0.0018781326478347182,
      "selected_pairs": 27151.3,
      "step": 29990,
      "student_distance_mean": 22.67154712677002,
      "teacher_distance_mean": 0.8565080225467682,
      "valid_action_samples": 233.5,
      "weighted_auxiliary_loss": 0.0018781326478347182,
      "weighted_relation_loss": 0.0018781326478347182
    },
    {
      "action_query_batch_std": 0.3445057809352875,
      "action_query_norm": 40.63103904724121,
      "action_relation_loss": 0.002035879495088011,
      "auxiliary_loss_weight": 1.0,
      "epoch": 28.06361085126286,
      "flow_matching_loss": 0.015804056636989118,
      "grad_norm": 0.10823199898004532,
      "h_stock_valid_norm": 97.95393447875976,
      "invalid_tail_samples": 23.1,
      "learning_rate": 3.0377360626943076e-13,
      "loss": 0.0178,
      "positive_teacher_pairs": 27014.0,
      "query_h_norm_ratio": 0.4147998869419098,
      "relation_loss": 0.002035879495088011,
      "selected_pairs": 27014.0,
      "step": 30000,
      "student_distance_mean": 22.56671714782715,
      "teacher_distance_mean": 0.8537414073944092,
      "valid_action_samples": 232.9,
      "weighted_auxiliary_loss": 0.002035879495088011,
      "weighted_relation_loss": 0.002035879495088011
    },
    {
      "epoch": 28.06361085126286,
      "step": 30000,
      "total_flos": 0.0,
      "train_loss": 0.044805257050693034,
      "train_runtime": 62779.2152,
      "train_samples_per_second": 122.333,
      "train_steps_per_second": 0.478
    }
  ],
  "logging_steps": 10,
  "max_steps": 30000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 29,
  "save_steps": 10000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}