{
"best_metric": 0.9155922909513518,
"best_model_checkpoint": "/shared/data/jatin2/sec-ner/multirun/onto_final/span-seq-new/run-242/checkpoints/checkpoint-64000",
"epoch": 68.30309498399146,
"global_step": 64000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.53,
"learning_rate": 9.982212735681253e-06,
"loss": 0.1396,
"step": 500
},
{
"epoch": 0.53,
"eval_loss": 0.06917473673820496,
"eval_micro_f1": 0.8507889064576218,
"eval_runtime": 47.2215,
"eval_samples_per_second": 180.596,
"step": 500
},
{
"epoch": 1.07,
"learning_rate": 9.964425471362506e-06,
"loss": 0.0645,
"step": 1000
},
{
"epoch": 1.07,
"eval_loss": 0.053893305361270905,
"eval_micro_f1": 0.8855092733841213,
"eval_runtime": 44.3745,
"eval_samples_per_second": 192.182,
"step": 1000
},
{
"epoch": 1.6,
"learning_rate": 9.946638207043758e-06,
"loss": 0.0518,
"step": 1500
},
{
"epoch": 1.6,
"eval_loss": 0.054491929709911346,
"eval_micro_f1": 0.8965486720665066,
"eval_runtime": 46.9823,
"eval_samples_per_second": 181.515,
"step": 1500
},
{
"epoch": 2.13,
"learning_rate": 9.92885094272501e-06,
"loss": 0.0488,
"step": 2000
},
{
"epoch": 2.13,
"eval_loss": 0.05046902969479561,
"eval_micro_f1": 0.89676840165434,
"eval_runtime": 47.1831,
"eval_samples_per_second": 180.743,
"step": 2000
},
{
"epoch": 2.67,
"learning_rate": 9.911063678406261e-06,
"loss": 0.041,
"step": 2500
},
{
"epoch": 2.67,
"eval_loss": 0.0494869165122509,
"eval_micro_f1": 0.9004850003980875,
"eval_runtime": 43.5595,
"eval_samples_per_second": 195.778,
"step": 2500
},
{
"epoch": 3.2,
"learning_rate": 9.893276414087513e-06,
"loss": 0.0391,
"step": 3000
},
{
"epoch": 3.2,
"eval_loss": 0.05162811279296875,
"eval_micro_f1": 0.9040532684815226,
"eval_runtime": 43.6064,
"eval_samples_per_second": 195.568,
"step": 3000
},
{
"epoch": 3.74,
"learning_rate": 9.875489149768766e-06,
"loss": 0.0351,
"step": 3500
},
{
"epoch": 3.74,
"eval_loss": 0.04937701299786568,
"eval_micro_f1": 0.902902185096518,
"eval_runtime": 43.5193,
"eval_samples_per_second": 195.959,
"step": 3500
},
{
"epoch": 4.27,
"learning_rate": 9.857701885450018e-06,
"loss": 0.0313,
"step": 4000
},
{
"epoch": 4.27,
"eval_loss": 0.053947556763887405,
"eval_micro_f1": 0.9051955003034208,
"eval_runtime": 45.8041,
"eval_samples_per_second": 186.184,
"step": 4000
},
{
"epoch": 4.8,
"learning_rate": 9.839914621131271e-06,
"loss": 0.0307,
"step": 4500
},
{
"epoch": 4.8,
"eval_loss": 0.05616093799471855,
"eval_micro_f1": 0.9047917034155855,
"eval_runtime": 46.5027,
"eval_samples_per_second": 183.387,
"step": 4500
},
{
"epoch": 5.34,
"learning_rate": 9.822127356812523e-06,
"loss": 0.0272,
"step": 5000
},
{
"epoch": 5.34,
"eval_loss": 0.055654264986515045,
"eval_micro_f1": 0.903533395213601,
"eval_runtime": 44.9896,
"eval_samples_per_second": 189.555,
"step": 5000
},
{
"epoch": 5.87,
"learning_rate": 9.804340092493775e-06,
"loss": 0.0253,
"step": 5500
},
{
"epoch": 5.87,
"eval_loss": 0.058256134390830994,
"eval_micro_f1": 0.9093266560322097,
"eval_runtime": 41.4006,
"eval_samples_per_second": 205.988,
"step": 5500
},
{
"epoch": 6.4,
"learning_rate": 9.786552828175028e-06,
"loss": 0.0226,
"step": 6000
},
{
"epoch": 6.4,
"eval_loss": 0.058179326355457306,
"eval_micro_f1": 0.9070450092847776,
"eval_runtime": 40.9904,
"eval_samples_per_second": 208.049,
"step": 6000
},
{
"epoch": 6.94,
"learning_rate": 9.76876556385628e-06,
"loss": 0.0222,
"step": 6500
},
{
"epoch": 6.94,
"eval_loss": 0.06260060518980026,
"eval_micro_f1": 0.9043354902350924,
"eval_runtime": 44.5657,
"eval_samples_per_second": 191.358,
"step": 6500
},
{
"epoch": 7.47,
"learning_rate": 9.750978299537532e-06,
"loss": 0.0203,
"step": 7000
},
{
"epoch": 7.47,
"eval_loss": 0.06296257674694061,
"eval_micro_f1": 0.9062177323843387,
"eval_runtime": 40.6652,
"eval_samples_per_second": 209.713,
"step": 7000
},
{
"epoch": 8.0,
"learning_rate": 9.733191035218784e-06,
"loss": 0.0195,
"step": 7500
},
{
"epoch": 8.0,
"eval_loss": 0.06152811273932457,
"eval_micro_f1": 0.9078007849337684,
"eval_runtime": 48.085,
"eval_samples_per_second": 177.353,
"step": 7500
},
{
"epoch": 8.54,
"learning_rate": 9.715403770900035e-06,
"loss": 0.0168,
"step": 8000
},
{
"epoch": 8.54,
"eval_loss": 0.061763305217027664,
"eval_micro_f1": 0.9065951258134094,
"eval_runtime": 44.3661,
"eval_samples_per_second": 192.219,
"step": 8000
},
{
"epoch": 9.07,
"learning_rate": 9.697616506581289e-06,
"loss": 0.0174,
"step": 8500
},
{
"epoch": 9.07,
"eval_loss": 0.0693483054637909,
"eval_micro_f1": 0.9062695639396461,
"eval_runtime": 45.6716,
"eval_samples_per_second": 186.725,
"step": 8500
},
{
"epoch": 9.61,
"learning_rate": 9.67982924226254e-06,
"loss": 0.0143,
"step": 9000
},
{
"epoch": 9.61,
"eval_loss": 0.0685899630188942,
"eval_micro_f1": 0.9052481379046072,
"eval_runtime": 45.0647,
"eval_samples_per_second": 189.239,
"step": 9000
},
{
"epoch": 10.14,
"learning_rate": 9.662041977943794e-06,
"loss": 0.0148,
"step": 9500
},
{
"epoch": 10.14,
"eval_loss": 0.0761900469660759,
"eval_micro_f1": 0.9051335796257772,
"eval_runtime": 42.891,
"eval_samples_per_second": 198.829,
"step": 9500
},
{
"epoch": 10.67,
"learning_rate": 9.644254713625046e-06,
"loss": 0.013,
"step": 10000
},
{
"epoch": 10.67,
"eval_loss": 0.07245437055826187,
"eval_micro_f1": 0.9080864330041375,
"eval_runtime": 42.463,
"eval_samples_per_second": 200.834,
"step": 10000
},
{
"epoch": 11.21,
"learning_rate": 9.626467449306297e-06,
"loss": 0.0126,
"step": 10500
},
{
"epoch": 11.21,
"eval_loss": 0.07603873312473297,
"eval_micro_f1": 0.9079348015776946,
"eval_runtime": 48.8066,
"eval_samples_per_second": 174.731,
"step": 10500
},
{
"epoch": 11.74,
"learning_rate": 9.60868018498755e-06,
"loss": 0.0115,
"step": 11000
},
{
"epoch": 11.74,
"eval_loss": 0.07910393923521042,
"eval_micro_f1": 0.9053824863338777,
"eval_runtime": 50.5685,
"eval_samples_per_second": 168.643,
"step": 11000
},
{
"epoch": 12.27,
"learning_rate": 9.590892920668802e-06,
"loss": 0.0108,
"step": 11500
},
{
"epoch": 12.27,
"eval_loss": 0.08104216307401657,
"eval_micro_f1": 0.9075532481128166,
"eval_runtime": 43.7601,
"eval_samples_per_second": 194.881,
"step": 11500
},
{
"epoch": 12.81,
"learning_rate": 9.573105656350054e-06,
"loss": 0.0097,
"step": 12000
},
{
"epoch": 12.81,
"eval_loss": 0.08225009590387344,
"eval_micro_f1": 0.9074148650300137,
"eval_runtime": 44.9505,
"eval_samples_per_second": 189.72,
"step": 12000
},
{
"epoch": 13.34,
"learning_rate": 9.555318392031306e-06,
"loss": 0.0093,
"step": 12500
},
{
"epoch": 13.34,
"eval_loss": 0.08659771829843521,
"eval_micro_f1": 0.9080490638745694,
"eval_runtime": 46.346,
"eval_samples_per_second": 184.007,
"step": 12500
},
{
"epoch": 13.87,
"learning_rate": 9.537531127712558e-06,
"loss": 0.0091,
"step": 13000
},
{
"epoch": 13.87,
"eval_loss": 0.08333197236061096,
"eval_micro_f1": 0.9081934841988564,
"eval_runtime": 45.3598,
"eval_samples_per_second": 188.008,
"step": 13000
},
{
"epoch": 14.41,
"learning_rate": 9.519743863393811e-06,
"loss": 0.0079,
"step": 13500
},
{
"epoch": 14.41,
"eval_loss": 0.09330150485038757,
"eval_micro_f1": 0.9081930410263139,
"eval_runtime": 44.0003,
"eval_samples_per_second": 193.817,
"step": 13500
},
{
"epoch": 14.94,
"learning_rate": 9.501956599075063e-06,
"loss": 0.0081,
"step": 14000
},
{
"epoch": 14.94,
"eval_loss": 0.08975318819284439,
"eval_micro_f1": 0.909180342030044,
"eval_runtime": 43.5141,
"eval_samples_per_second": 195.982,
"step": 14000
},
{
"epoch": 15.47,
"learning_rate": 9.484169334756316e-06,
"loss": 0.0074,
"step": 14500
},
{
"epoch": 15.47,
"eval_loss": 0.09231750667095184,
"eval_micro_f1": 0.9108057122458268,
"eval_runtime": 44.0088,
"eval_samples_per_second": 193.779,
"step": 14500
},
{
"epoch": 16.01,
"learning_rate": 9.466382070437568e-06,
"loss": 0.0071,
"step": 15000
},
{
"epoch": 16.01,
"eval_loss": 0.0896356925368309,
"eval_micro_f1": 0.9083528770030768,
"eval_runtime": 45.1902,
"eval_samples_per_second": 188.713,
"step": 15000
},
{
"epoch": 16.54,
"learning_rate": 9.44859480611882e-06,
"loss": 0.0066,
"step": 15500
},
{
"epoch": 16.54,
"eval_loss": 0.09429354965686798,
"eval_micro_f1": 0.9097079569886002,
"eval_runtime": 43.4878,
"eval_samples_per_second": 196.101,
"step": 15500
},
{
"epoch": 17.08,
"learning_rate": 9.430807541800071e-06,
"loss": 0.0062,
"step": 16000
},
{
"epoch": 17.08,
"eval_loss": 0.09975157678127289,
"eval_micro_f1": 0.9111468663990563,
"eval_runtime": 43.9806,
"eval_samples_per_second": 193.904,
"step": 16000
},
{
"epoch": 17.61,
"learning_rate": 9.413020277481323e-06,
"loss": 0.0058,
"step": 16500
},
{
"epoch": 17.61,
"eval_loss": 0.095846988260746,
"eval_micro_f1": 0.9107440243457641,
"eval_runtime": 47.8669,
"eval_samples_per_second": 178.161,
"step": 16500
},
{
"epoch": 18.14,
"learning_rate": 9.395233013162577e-06,
"loss": 0.0061,
"step": 17000
},
{
"epoch": 18.14,
"eval_loss": 0.10315552353858948,
"eval_micro_f1": 0.9109053307954449,
"eval_runtime": 43.2647,
"eval_samples_per_second": 197.112,
"step": 17000
},
{
"epoch": 18.68,
"learning_rate": 9.377445748843828e-06,
"loss": 0.0055,
"step": 17500
},
{
"epoch": 18.68,
"eval_loss": 0.09754090011119843,
"eval_micro_f1": 0.9107190825124066,
"eval_runtime": 44.536,
"eval_samples_per_second": 191.486,
"step": 17500
},
{
"epoch": 19.21,
"learning_rate": 9.35965848452508e-06,
"loss": 0.005,
"step": 18000
},
{
"epoch": 19.21,
"eval_loss": 0.10331171751022339,
"eval_micro_f1": 0.9097737632485202,
"eval_runtime": 43.1325,
"eval_samples_per_second": 197.716,
"step": 18000
},
{
"epoch": 19.74,
"learning_rate": 9.341871220206333e-06,
"loss": 0.0048,
"step": 18500
},
{
"epoch": 19.74,
"eval_loss": 0.10065481066703796,
"eval_micro_f1": 0.9080872448100569,
"eval_runtime": 43.9617,
"eval_samples_per_second": 193.987,
"step": 18500
},
{
"epoch": 20.28,
"learning_rate": 9.324083955887585e-06,
"loss": 0.0047,
"step": 19000
},
{
"epoch": 20.28,
"eval_loss": 0.10182579606771469,
"eval_micro_f1": 0.9104290068594819,
"eval_runtime": 44.6644,
"eval_samples_per_second": 190.935,
"step": 19000
},
{
"epoch": 20.81,
"learning_rate": 9.306296691568839e-06,
"loss": 0.0042,
"step": 19500
},
{
"epoch": 20.81,
"eval_loss": 0.10571350157260895,
"eval_micro_f1": 0.9054650215580867,
"eval_runtime": 45.7704,
"eval_samples_per_second": 186.321,
"step": 19500
},
{
"epoch": 21.34,
"learning_rate": 9.28850942725009e-06,
"loss": 0.0041,
"step": 20000
},
{
"epoch": 21.34,
"eval_loss": 0.10765829682350159,
"eval_micro_f1": 0.9095137983101853,
"eval_runtime": 45.6344,
"eval_samples_per_second": 186.876,
"step": 20000
},
{
"epoch": 21.88,
"learning_rate": 9.270722162931342e-06,
"loss": 0.0041,
"step": 20500
},
{
"epoch": 21.88,
"eval_loss": 0.10483796149492264,
"eval_micro_f1": 0.9063044541713539,
"eval_runtime": 44.5327,
"eval_samples_per_second": 191.5,
"step": 20500
},
{
"epoch": 22.41,
"learning_rate": 9.252934898612594e-06,
"loss": 0.0035,
"step": 21000
},
{
"epoch": 22.41,
"eval_loss": 0.1031440868973732,
"eval_micro_f1": 0.909774435590189,
"eval_runtime": 49.1331,
"eval_samples_per_second": 173.569,
"step": 21000
},
{
"epoch": 22.95,
"learning_rate": 9.235147634293846e-06,
"loss": 0.0041,
"step": 21500
},
{
"epoch": 22.95,
"eval_loss": 0.10528811067342758,
"eval_micro_f1": 0.9088622372935843,
"eval_runtime": 48.2005,
"eval_samples_per_second": 176.928,
"step": 21500
},
{
"epoch": 23.48,
"learning_rate": 9.217360369975099e-06,
"loss": 0.0034,
"step": 22000
},
{
"epoch": 23.48,
"eval_loss": 0.10731495916843414,
"eval_micro_f1": 0.9071338492567241,
"eval_runtime": 43.0621,
"eval_samples_per_second": 198.04,
"step": 22000
},
{
"epoch": 24.01,
"learning_rate": 9.19957310565635e-06,
"loss": 0.0034,
"step": 22500
},
{
"epoch": 24.01,
"eval_loss": 0.10347431898117065,
"eval_micro_f1": 0.9092945123778954,
"eval_runtime": 45.5983,
"eval_samples_per_second": 187.024,
"step": 22500
},
{
"epoch": 24.55,
"learning_rate": 9.181785841337602e-06,
"loss": 0.0031,
"step": 23000
},
{
"epoch": 24.55,
"eval_loss": 0.10810932517051697,
"eval_micro_f1": 0.9092624197676561,
"eval_runtime": 44.3094,
"eval_samples_per_second": 192.465,
"step": 23000
},
{
"epoch": 25.08,
"learning_rate": 9.163998577018856e-06,
"loss": 0.0033,
"step": 23500
},
{
"epoch": 25.08,
"eval_loss": 0.10781945288181305,
"eval_micro_f1": 0.9100247798558581,
"eval_runtime": 42.7061,
"eval_samples_per_second": 199.691,
"step": 23500
},
{
"epoch": 25.61,
"learning_rate": 9.146211312700108e-06,
"loss": 0.0031,
"step": 24000
},
{
"epoch": 25.61,
"eval_loss": 0.11176478862762451,
"eval_micro_f1": 0.9099009895989358,
"eval_runtime": 46.3293,
"eval_samples_per_second": 184.074,
"step": 24000
},
{
"epoch": 26.15,
"learning_rate": 9.128424048381361e-06,
"loss": 0.003,
"step": 24500
},
{
"epoch": 26.15,
"eval_loss": 0.11290744692087173,
"eval_micro_f1": 0.911982082366697,
"eval_runtime": 44.2113,
"eval_samples_per_second": 192.892,
"step": 24500
},
{
"epoch": 26.68,
"learning_rate": 9.110636784062613e-06,
"loss": 0.0026,
"step": 25000
},
{
"epoch": 26.68,
"eval_loss": 0.11656230688095093,
"eval_micro_f1": 0.9108759969894135,
"eval_runtime": 43.1371,
"eval_samples_per_second": 197.695,
"step": 25000
},
{
"epoch": 27.21,
"learning_rate": 9.092849519743864e-06,
"loss": 0.0029,
"step": 25500
},
{
"epoch": 27.21,
"eval_loss": 0.10980264842510223,
"eval_micro_f1": 0.9106142743330552,
"eval_runtime": 46.7351,
"eval_samples_per_second": 182.475,
"step": 25500
},
{
"epoch": 27.75,
"learning_rate": 9.075062255425116e-06,
"loss": 0.0025,
"step": 26000
},
{
"epoch": 27.75,
"eval_loss": 0.10632364451885223,
"eval_micro_f1": 0.9063570438525452,
"eval_runtime": 45.3683,
"eval_samples_per_second": 187.973,
"step": 26000
},
{
"epoch": 28.28,
"learning_rate": 9.057274991106368e-06,
"loss": 0.0027,
"step": 26500
},
{
"epoch": 28.28,
"eval_loss": 0.12237359583377838,
"eval_micro_f1": 0.9086129300351484,
"eval_runtime": 41.7327,
"eval_samples_per_second": 204.348,
"step": 26500
},
{
"epoch": 28.82,
"learning_rate": 9.039487726787621e-06,
"loss": 0.0025,
"step": 27000
},
{
"epoch": 28.82,
"eval_loss": 0.11551995575428009,
"eval_micro_f1": 0.906537398131734,
"eval_runtime": 46.1505,
"eval_samples_per_second": 184.787,
"step": 27000
},
{
"epoch": 29.35,
"learning_rate": 9.021700462468873e-06,
"loss": 0.0025,
"step": 27500
},
{
"epoch": 29.35,
"eval_loss": 0.11351078003644943,
"eval_micro_f1": 0.9115163487633205,
"eval_runtime": 43.724,
"eval_samples_per_second": 195.042,
"step": 27500
},
{
"epoch": 29.88,
"learning_rate": 9.003913198150125e-06,
"loss": 0.0022,
"step": 28000
},
{
"epoch": 29.88,
"eval_loss": 0.1204790323972702,
"eval_micro_f1": 0.9119454532298145,
"eval_runtime": 46.9877,
"eval_samples_per_second": 181.494,
"step": 28000
},
{
"epoch": 30.42,
"learning_rate": 8.986125933831378e-06,
"loss": 0.0021,
"step": 28500
},
{
"epoch": 30.42,
"eval_loss": 0.11966580152511597,
"eval_micro_f1": 0.9103319052815836,
"eval_runtime": 48.1213,
"eval_samples_per_second": 177.219,
"step": 28500
},
{
"epoch": 30.95,
"learning_rate": 8.96833866951263e-06,
"loss": 0.0022,
"step": 29000
},
{
"epoch": 30.95,
"eval_loss": 0.11313212662935257,
"eval_micro_f1": 0.9111858322728692,
"eval_runtime": 44.8208,
"eval_samples_per_second": 190.269,
"step": 29000
},
{
"epoch": 31.48,
"learning_rate": 8.950551405193882e-06,
"loss": 0.0022,
"step": 29500
},
{
"epoch": 31.48,
"eval_loss": 0.11699814349412918,
"eval_micro_f1": 0.911924058067997,
"eval_runtime": 46.7711,
"eval_samples_per_second": 182.335,
"step": 29500
},
{
"epoch": 32.02,
"learning_rate": 8.932764140875133e-06,
"loss": 0.0023,
"step": 30000
},
{
"epoch": 32.02,
"eval_loss": 0.12167295068502426,
"eval_micro_f1": 0.9091890428139211,
"eval_runtime": 44.7146,
"eval_samples_per_second": 190.721,
"step": 30000
},
{
"epoch": 32.55,
"learning_rate": 8.914976876556385e-06,
"loss": 0.0021,
"step": 30500
},
{
"epoch": 32.55,
"eval_loss": 0.12321187555789948,
"eval_micro_f1": 0.9111778679481384,
"eval_runtime": 43.986,
"eval_samples_per_second": 193.88,
"step": 30500
},
{
"epoch": 33.08,
"learning_rate": 8.897189612237639e-06,
"loss": 0.0019,
"step": 31000
},
{
"epoch": 33.08,
"eval_loss": 0.11899471282958984,
"eval_micro_f1": 0.9125106607200316,
"eval_runtime": 45.7124,
"eval_samples_per_second": 186.558,
"step": 31000
},
{
"epoch": 33.62,
"learning_rate": 8.87940234791889e-06,
"loss": 0.0019,
"step": 31500
},
{
"epoch": 33.62,
"eval_loss": 0.117021843791008,
"eval_micro_f1": 0.9110102865490342,
"eval_runtime": 46.4773,
"eval_samples_per_second": 183.487,
"step": 31500
},
{
"epoch": 34.15,
"learning_rate": 8.861615083600144e-06,
"loss": 0.0018,
"step": 32000
},
{
"epoch": 34.15,
"eval_loss": 0.12451651692390442,
"eval_micro_f1": 0.9086342224198825,
"eval_runtime": 44.3046,
"eval_samples_per_second": 192.486,
"step": 32000
},
{
"epoch": 34.69,
"learning_rate": 8.843827819281396e-06,
"loss": 0.0016,
"step": 32500
},
{
"epoch": 34.69,
"eval_loss": 0.12626737356185913,
"eval_micro_f1": 0.910206457469414,
"eval_runtime": 46.4789,
"eval_samples_per_second": 183.481,
"step": 32500
},
{
"epoch": 35.22,
"learning_rate": 8.826040554962647e-06,
"loss": 0.0017,
"step": 33000
},
{
"epoch": 35.22,
"eval_loss": 0.11690003424882889,
"eval_micro_f1": 0.9091967920457599,
"eval_runtime": 43.9637,
"eval_samples_per_second": 193.978,
"step": 33000
},
{
"epoch": 35.75,
"learning_rate": 8.8082532906439e-06,
"loss": 0.0019,
"step": 33500
},
{
"epoch": 35.75,
"eval_loss": 0.11754824221134186,
"eval_micro_f1": 0.9126588526141316,
"eval_runtime": 47.8135,
"eval_samples_per_second": 178.36,
"step": 33500
},
{
"epoch": 36.29,
"learning_rate": 8.790466026325152e-06,
"loss": 0.0018,
"step": 34000
},
{
"epoch": 36.29,
"eval_loss": 0.11732104420661926,
"eval_micro_f1": 0.9114286989309831,
"eval_runtime": 45.4705,
"eval_samples_per_second": 187.55,
"step": 34000
},
{
"epoch": 36.82,
"learning_rate": 8.772678762006404e-06,
"loss": 0.0016,
"step": 34500
},
{
"epoch": 36.82,
"eval_loss": 0.12185899168252945,
"eval_micro_f1": 0.9141495859796149,
"eval_runtime": 45.7362,
"eval_samples_per_second": 186.461,
"step": 34500
},
{
"epoch": 37.35,
"learning_rate": 8.754891497687656e-06,
"loss": 0.0015,
"step": 35000
},
{
"epoch": 37.35,
"eval_loss": 0.12255921214818954,
"eval_micro_f1": 0.9103342776301292,
"eval_runtime": 49.0843,
"eval_samples_per_second": 173.742,
"step": 35000
},
{
"epoch": 37.89,
"learning_rate": 8.737104233368908e-06,
"loss": 0.0015,
"step": 35500
},
{
"epoch": 37.89,
"eval_loss": 0.12213725596666336,
"eval_micro_f1": 0.9106478363078391,
"eval_runtime": 44.8035,
"eval_samples_per_second": 190.342,
"step": 35500
},
{
"epoch": 38.42,
"learning_rate": 8.719316969050161e-06,
"loss": 0.0013,
"step": 36000
},
{
"epoch": 38.42,
"eval_loss": 0.1252118945121765,
"eval_micro_f1": 0.9098604227327022,
"eval_runtime": 47.4461,
"eval_samples_per_second": 179.741,
"step": 36000
},
{
"epoch": 38.95,
"learning_rate": 8.701529704731413e-06,
"loss": 0.0017,
"step": 36500
},
{
"epoch": 38.95,
"eval_loss": 0.12273646146059036,
"eval_micro_f1": 0.9102719163570082,
"eval_runtime": 44.0448,
"eval_samples_per_second": 193.621,
"step": 36500
},
{
"epoch": 39.49,
"learning_rate": 8.683742440412666e-06,
"loss": 0.0014,
"step": 37000
},
{
"epoch": 39.49,
"eval_loss": 0.12856261432170868,
"eval_micro_f1": 0.909180342030044,
"eval_runtime": 44.4228,
"eval_samples_per_second": 191.973,
"step": 37000
},
{
"epoch": 40.02,
"learning_rate": 8.665955176093918e-06,
"loss": 0.0017,
"step": 37500
},
{
"epoch": 40.02,
"eval_loss": 0.1222791001200676,
"eval_micro_f1": 0.9106700863547029,
"eval_runtime": 44.0883,
"eval_samples_per_second": 193.43,
"step": 37500
},
{
"epoch": 40.55,
"learning_rate": 8.64816791177517e-06,
"loss": 0.0014,
"step": 38000
},
{
"epoch": 40.55,
"eval_loss": 0.1245175153017044,
"eval_micro_f1": 0.9115401880944589,
"eval_runtime": 43.2066,
"eval_samples_per_second": 197.377,
"step": 38000
},
{
"epoch": 41.09,
"learning_rate": 8.630380647456423e-06,
"loss": 0.0013,
"step": 38500
},
{
"epoch": 41.09,
"eval_loss": 0.1244526281952858,
"eval_micro_f1": 0.90980110397621,
"eval_runtime": 45.3014,
"eval_samples_per_second": 188.25,
"step": 38500
},
{
"epoch": 41.62,
"learning_rate": 8.612593383137675e-06,
"loss": 0.0013,
"step": 39000
},
{
"epoch": 41.62,
"eval_loss": 0.13098452985286713,
"eval_micro_f1": 0.9096126250379681,
"eval_runtime": 44.747,
"eval_samples_per_second": 190.583,
"step": 39000
},
{
"epoch": 42.16,
"learning_rate": 8.594806118818927e-06,
"loss": 0.0013,
"step": 39500
},
{
"epoch": 42.16,
"eval_loss": 0.12692411243915558,
"eval_micro_f1": 0.9110701267996445,
"eval_runtime": 51.5945,
"eval_samples_per_second": 165.289,
"step": 39500
},
{
"epoch": 42.69,
"learning_rate": 8.577018854500178e-06,
"loss": 0.0012,
"step": 40000
},
{
"epoch": 42.69,
"eval_loss": 0.12520797550678253,
"eval_micro_f1": 0.9075154884107766,
"eval_runtime": 46.8501,
"eval_samples_per_second": 182.027,
"step": 40000
},
{
"epoch": 43.22,
"learning_rate": 8.55923159018143e-06,
"loss": 0.0016,
"step": 40500
},
{
"epoch": 43.22,
"eval_loss": 0.12235218286514282,
"eval_micro_f1": 0.9129032253064053,
"eval_runtime": 47.9517,
"eval_samples_per_second": 177.846,
"step": 40500
},
{
"epoch": 43.76,
"learning_rate": 8.541444325862683e-06,
"loss": 0.0015,
"step": 41000
},
{
"epoch": 43.76,
"eval_loss": 0.12375035136938095,
"eval_micro_f1": 0.9107487754959827,
"eval_runtime": 44.7648,
"eval_samples_per_second": 190.507,
"step": 41000
},
{
"epoch": 44.29,
"learning_rate": 8.523657061543935e-06,
"loss": 0.0014,
"step": 41500
},
{
"epoch": 44.29,
"eval_loss": 0.12095450609922409,
"eval_micro_f1": 0.9114660219578029,
"eval_runtime": 44.8399,
"eval_samples_per_second": 190.188,
"step": 41500
},
{
"epoch": 44.82,
"learning_rate": 8.505869797225187e-06,
"loss": 0.0016,
"step": 42000
},
{
"epoch": 44.82,
"eval_loss": 0.11838436126708984,
"eval_micro_f1": 0.9107334520938881,
"eval_runtime": 47.7032,
"eval_samples_per_second": 178.772,
"step": 42000
},
{
"epoch": 45.36,
"learning_rate": 8.48808253290644e-06,
"loss": 0.0012,
"step": 42500
},
{
"epoch": 45.36,
"eval_loss": 0.12828320264816284,
"eval_micro_f1": 0.9126065495223747,
"eval_runtime": 50.6843,
"eval_samples_per_second": 168.257,
"step": 42500
},
{
"epoch": 45.89,
"learning_rate": 8.470295268587692e-06,
"loss": 0.0015,
"step": 43000
},
{
"epoch": 45.89,
"eval_loss": 0.12744639813899994,
"eval_micro_f1": 0.9121225772376408,
"eval_runtime": 45.1765,
"eval_samples_per_second": 188.771,
"step": 43000
},
{
"epoch": 46.42,
"learning_rate": 8.452508004268944e-06,
"loss": 0.0011,
"step": 43500
},
{
"epoch": 46.42,
"eval_loss": 0.12377772480249405,
"eval_micro_f1": 0.9112910476330811,
"eval_runtime": 47.3938,
"eval_samples_per_second": 179.939,
"step": 43500
},
{
"epoch": 46.96,
"learning_rate": 8.434720739950195e-06,
"loss": 0.0014,
"step": 44000
},
{
"epoch": 46.96,
"eval_loss": 0.13046270608901978,
"eval_micro_f1": 0.9105807544365998,
"eval_runtime": 46.593,
"eval_samples_per_second": 183.032,
"step": 44000
},
{
"epoch": 47.49,
"learning_rate": 8.416933475631447e-06,
"loss": 0.0011,
"step": 44500
},
{
"epoch": 47.49,
"eval_loss": 0.13093392550945282,
"eval_micro_f1": 0.9136928713361216,
"eval_runtime": 43.4412,
"eval_samples_per_second": 196.311,
"step": 44500
},
{
"epoch": 48.03,
"learning_rate": 8.3991462113127e-06,
"loss": 0.0011,
"step": 45000
},
{
"epoch": 48.03,
"eval_loss": 0.12840227782726288,
"eval_micro_f1": 0.9109625663449327,
"eval_runtime": 43.0046,
"eval_samples_per_second": 198.304,
"step": 45000
},
{
"epoch": 48.56,
"learning_rate": 8.381358946993952e-06,
"loss": 0.0011,
"step": 45500
},
{
"epoch": 48.56,
"eval_loss": 0.12799955904483795,
"eval_micro_f1": 0.9129850340012522,
"eval_runtime": 43.8275,
"eval_samples_per_second": 194.581,
"step": 45500
},
{
"epoch": 49.09,
"learning_rate": 8.363571682675206e-06,
"loss": 0.0011,
"step": 46000
},
{
"epoch": 49.09,
"eval_loss": 0.1340486705303192,
"eval_micro_f1": 0.9129616344945723,
"eval_runtime": 43.6109,
"eval_samples_per_second": 195.547,
"step": 46000
},
{
"epoch": 49.63,
"learning_rate": 8.345784418356458e-06,
"loss": 0.0011,
"step": 46500
},
{
"epoch": 49.63,
"eval_loss": 0.12872454524040222,
"eval_micro_f1": 0.9137078646684711,
"eval_runtime": 47.0326,
"eval_samples_per_second": 181.321,
"step": 46500
},
{
"epoch": 50.16,
"learning_rate": 8.32799715403771e-06,
"loss": 0.0011,
"step": 47000
},
{
"epoch": 50.16,
"eval_loss": 0.13874606788158417,
"eval_micro_f1": 0.9128863905076278,
"eval_runtime": 45.1282,
"eval_samples_per_second": 188.973,
"step": 47000
},
{
"epoch": 50.69,
"learning_rate": 8.310209889718963e-06,
"loss": 0.001,
"step": 47500
},
{
"epoch": 50.69,
"eval_loss": 0.12773732841014862,
"eval_micro_f1": 0.9103907193416637,
"eval_runtime": 47.3484,
"eval_samples_per_second": 180.112,
"step": 47500
},
{
"epoch": 51.23,
"learning_rate": 8.292422625400214e-06,
"loss": 0.001,
"step": 48000
},
{
"epoch": 51.23,
"eval_loss": 0.13285644352436066,
"eval_micro_f1": 0.9120158168810779,
"eval_runtime": 47.3847,
"eval_samples_per_second": 179.974,
"step": 48000
},
{
"epoch": 51.76,
"learning_rate": 8.274635361081466e-06,
"loss": 0.0011,
"step": 48500
},
{
"epoch": 51.76,
"eval_loss": 0.12811879813671112,
"eval_micro_f1": 0.9127023265411526,
"eval_runtime": 43.9686,
"eval_samples_per_second": 193.957,
"step": 48500
},
{
"epoch": 52.29,
"learning_rate": 8.256848096762718e-06,
"loss": 0.0009,
"step": 49000
},
{
"epoch": 52.29,
"eval_loss": 0.13596977293491364,
"eval_micro_f1": 0.9087556769744087,
"eval_runtime": 52.7392,
"eval_samples_per_second": 161.701,
"step": 49000
},
{
"epoch": 52.83,
"learning_rate": 8.23906083244397e-06,
"loss": 0.0013,
"step": 49500
},
{
"epoch": 52.83,
"eval_loss": 0.1310349702835083,
"eval_micro_f1": 0.9129323100907124,
"eval_runtime": 44.2565,
"eval_samples_per_second": 192.695,
"step": 49500
},
{
"epoch": 53.36,
"learning_rate": 8.221273568125223e-06,
"loss": 0.0011,
"step": 50000
},
{
"epoch": 53.36,
"eval_loss": 0.12716206908226013,
"eval_micro_f1": 0.9134063168007321,
"eval_runtime": 42.457,
"eval_samples_per_second": 200.862,
"step": 50000
},
{
"epoch": 53.9,
"learning_rate": 8.203486303806475e-06,
"loss": 0.001,
"step": 50500
},
{
"epoch": 53.9,
"eval_loss": 0.12756939232349396,
"eval_micro_f1": 0.9134550670220388,
"eval_runtime": 43.3886,
"eval_samples_per_second": 196.549,
"step": 50500
},
{
"epoch": 54.43,
"learning_rate": 8.185699039487728e-06,
"loss": 0.0011,
"step": 51000
},
{
"epoch": 54.43,
"eval_loss": 0.1292908936738968,
"eval_micro_f1": 0.9132248077873131,
"eval_runtime": 45.3297,
"eval_samples_per_second": 188.133,
"step": 51000
},
{
"epoch": 54.96,
"learning_rate": 8.16791177516898e-06,
"loss": 0.0011,
"step": 51500
},
{
"epoch": 54.96,
"eval_loss": 0.12562933564186096,
"eval_micro_f1": 0.9106839512210302,
"eval_runtime": 45.7875,
"eval_samples_per_second": 186.252,
"step": 51500
},
{
"epoch": 55.5,
"learning_rate": 8.150124510850232e-06,
"loss": 0.001,
"step": 52000
},
{
"epoch": 55.5,
"eval_loss": 0.1305917203426361,
"eval_micro_f1": 0.9108813387417151,
"eval_runtime": 48.2614,
"eval_samples_per_second": 176.705,
"step": 52000
},
{
"epoch": 56.03,
"learning_rate": 8.132337246531485e-06,
"loss": 0.0012,
"step": 52500
},
{
"epoch": 56.03,
"eval_loss": 0.1318601816892624,
"eval_micro_f1": 0.9126969909039713,
"eval_runtime": 43.4459,
"eval_samples_per_second": 196.29,
"step": 52500
},
{
"epoch": 56.56,
"learning_rate": 8.114549982212737e-06,
"loss": 0.0008,
"step": 53000
},
{
"epoch": 56.56,
"eval_loss": 0.12605132162570953,
"eval_micro_f1": 0.9133497863520843,
"eval_runtime": 48.0012,
"eval_samples_per_second": 177.662,
"step": 53000
},
{
"epoch": 57.1,
"learning_rate": 8.096762717893989e-06,
"loss": 0.001,
"step": 53500
},
{
"epoch": 57.1,
"eval_loss": 0.13017183542251587,
"eval_micro_f1": 0.9122241613548477,
"eval_runtime": 44.1304,
"eval_samples_per_second": 193.246,
"step": 53500
},
{
"epoch": 57.63,
"learning_rate": 8.07897545357524e-06,
"loss": 0.0009,
"step": 54000
},
{
"epoch": 57.63,
"eval_loss": 0.13118210434913635,
"eval_micro_f1": 0.9133766519758647,
"eval_runtime": 48.1075,
"eval_samples_per_second": 177.27,
"step": 54000
},
{
"epoch": 58.16,
"learning_rate": 8.061188189256492e-06,
"loss": 0.001,
"step": 54500
},
{
"epoch": 58.16,
"eval_loss": 0.12910974025726318,
"eval_micro_f1": 0.9112526915315292,
"eval_runtime": 45.5581,
"eval_samples_per_second": 187.19,
"step": 54500
},
{
"epoch": 58.7,
"learning_rate": 8.043400924937745e-06,
"loss": 0.0008,
"step": 55000
},
{
"epoch": 58.7,
"eval_loss": 0.1358337253332138,
"eval_micro_f1": 0.9123137672256638,
"eval_runtime": 42.3059,
"eval_samples_per_second": 201.58,
"step": 55000
},
{
"epoch": 59.23,
"learning_rate": 8.025613660618997e-06,
"loss": 0.0009,
"step": 55500
},
{
"epoch": 59.23,
"eval_loss": 0.12926098704338074,
"eval_micro_f1": 0.9130766806912558,
"eval_runtime": 45.3862,
"eval_samples_per_second": 187.898,
"step": 55500
},
{
"epoch": 59.77,
"learning_rate": 8.00782639630025e-06,
"loss": 0.0007,
"step": 56000
},
{
"epoch": 59.77,
"eval_loss": 0.1348552107810974,
"eval_micro_f1": 0.9118770497852874,
"eval_runtime": 49.7406,
"eval_samples_per_second": 171.449,
"step": 56000
},
{
"epoch": 60.3,
"learning_rate": 7.990039131981502e-06,
"loss": 0.0011,
"step": 56500
},
{
"epoch": 60.3,
"eval_loss": 0.1303446888923645,
"eval_micro_f1": 0.9138566670657963,
"eval_runtime": 47.8045,
"eval_samples_per_second": 178.393,
"step": 56500
},
{
"epoch": 60.83,
"learning_rate": 7.972251867662754e-06,
"loss": 0.0008,
"step": 57000
},
{
"epoch": 60.83,
"eval_loss": 0.13565008342266083,
"eval_micro_f1": 0.9121833313413373,
"eval_runtime": 42.1628,
"eval_samples_per_second": 202.263,
"step": 57000
},
{
"epoch": 61.37,
"learning_rate": 7.954464603344006e-06,
"loss": 0.0011,
"step": 57500
},
{
"epoch": 61.37,
"eval_loss": 0.13657349348068237,
"eval_micro_f1": 0.9119885392080814,
"eval_runtime": 52.1051,
"eval_samples_per_second": 163.669,
"step": 57500
},
{
"epoch": 61.9,
"learning_rate": 7.936677339025257e-06,
"loss": 0.0008,
"step": 58000
},
{
"epoch": 61.9,
"eval_loss": 0.1361093819141388,
"eval_micro_f1": 0.9129787516991338,
"eval_runtime": 42.9942,
"eval_samples_per_second": 198.352,
"step": 58000
},
{
"epoch": 62.43,
"learning_rate": 7.918890074706511e-06,
"loss": 0.0009,
"step": 58500
},
{
"epoch": 62.43,
"eval_loss": 0.1298229992389679,
"eval_micro_f1": 0.9118557389697329,
"eval_runtime": 43.9332,
"eval_samples_per_second": 194.113,
"step": 58500
},
{
"epoch": 62.97,
"learning_rate": 7.901102810387763e-06,
"loss": 0.0008,
"step": 59000
},
{
"epoch": 62.97,
"eval_loss": 0.13627828657627106,
"eval_micro_f1": 0.9103615100496123,
"eval_runtime": 43.9707,
"eval_samples_per_second": 193.947,
"step": 59000
},
{
"epoch": 63.5,
"learning_rate": 7.883315546069014e-06,
"loss": 0.0008,
"step": 59500
},
{
"epoch": 63.5,
"eval_loss": 0.13557809591293335,
"eval_micro_f1": 0.9111331644223297,
"eval_runtime": 47.707,
"eval_samples_per_second": 178.758,
"step": 59500
},
{
"epoch": 64.03,
"learning_rate": 7.865528281750268e-06,
"loss": 0.0009,
"step": 60000
},
{
"epoch": 64.03,
"eval_loss": 0.1375960111618042,
"eval_micro_f1": 0.9128287992126606,
"eval_runtime": 47.5815,
"eval_samples_per_second": 179.229,
"step": 60000
},
{
"epoch": 64.57,
"learning_rate": 7.84774101743152e-06,
"loss": 0.0008,
"step": 60500
},
{
"epoch": 64.57,
"eval_loss": 0.13889284431934357,
"eval_micro_f1": 0.9124568554773578,
"eval_runtime": 42.2394,
"eval_samples_per_second": 201.897,
"step": 60500
},
{
"epoch": 65.1,
"learning_rate": 7.829953753112773e-06,
"loss": 0.001,
"step": 61000
},
{
"epoch": 65.1,
"eval_loss": 0.1284203678369522,
"eval_micro_f1": 0.9096126250379681,
"eval_runtime": 42.4537,
"eval_samples_per_second": 200.878,
"step": 61000
},
{
"epoch": 65.64,
"learning_rate": 7.812166488794025e-06,
"loss": 0.0008,
"step": 61500
},
{
"epoch": 65.64,
"eval_loss": 0.13290786743164062,
"eval_micro_f1": 0.9081056869159901,
"eval_runtime": 46.7022,
"eval_samples_per_second": 182.604,
"step": 61500
},
{
"epoch": 66.17,
"learning_rate": 7.794379224475276e-06,
"loss": 0.0008,
"step": 62000
},
{
"epoch": 66.17,
"eval_loss": 0.13682569563388824,
"eval_micro_f1": 0.9096936027968556,
"eval_runtime": 46.6469,
"eval_samples_per_second": 182.82,
"step": 62000
},
{
"epoch": 66.7,
"learning_rate": 7.776591960156528e-06,
"loss": 0.0009,
"step": 62500
},
{
"epoch": 66.7,
"eval_loss": 0.13292589783668518,
"eval_micro_f1": 0.9111568394391284,
"eval_runtime": 43.1464,
"eval_samples_per_second": 197.653,
"step": 62500
},
{
"epoch": 67.24,
"learning_rate": 7.75880469583778e-06,
"loss": 0.0008,
"step": 63000
},
{
"epoch": 67.24,
"eval_loss": 0.1360688954591751,
"eval_micro_f1": 0.9103516144234715,
"eval_runtime": 46.5325,
"eval_samples_per_second": 183.27,
"step": 63000
},
{
"epoch": 67.77,
"learning_rate": 7.741017431519033e-06,
"loss": 0.001,
"step": 63500
},
{
"epoch": 67.77,
"eval_loss": 0.12998247146606445,
"eval_micro_f1": 0.9142185658923968,
"eval_runtime": 46.3744,
"eval_samples_per_second": 183.894,
"step": 63500
},
{
"epoch": 68.3,
"learning_rate": 7.723230167200285e-06,
"loss": 0.0007,
"step": 64000
},
{
"epoch": 68.3,
"eval_loss": 0.13076944649219513,
"eval_micro_f1": 0.9155922909513518,
"eval_runtime": 44.839,
"eval_samples_per_second": 190.192,
"step": 64000
}
],
"max_steps": 281100,
"num_train_epochs": 300,
"total_flos": 289346678008515360,
"trial_name": null,
"trial_params": null
}
{
"best_metric": 0.9155922909513518,
"best_model_checkpoint": "/shared/data/jatin2/sec-ner/multirun/onto_final/span-seq-new/run-242/checkpoints/checkpoint-64000",
"epoch": 68.30309498399146,
"global_step": 64000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.53,
"learning_rate": 9.982212735681253e-06,
"loss": 0.1396,
"step": 500
},
{
"epoch": 0.53,
"eval_loss": 0.06917473673820496,
"eval_micro_f1": 0.8507889064576218,
"eval_runtime": 47.2215,
"eval_samples_per_second": 180.596,
"step": 500
},
{
"epoch": 1.07,
"learning_rate": 9.964425471362506e-06,
"loss": 0.0645,
"step": 1000
},
{
"epoch": 1.07,
"eval_loss": 0.053893305361270905,
"eval_micro_f1": 0.8855092733841213,
"eval_runtime": 44.3745,
"eval_samples_per_second": 192.182,
"step": 1000
},
{
"epoch": 1.6,
"learning_rate": 9.946638207043758e-06,
"loss": 0.0518,
"step": 1500
},
{
"epoch": 1.6,
"eval_loss": 0.054491929709911346,
"eval_micro_f1": 0.8965486720665066,
"eval_runtime": 46.9823,
"eval_samples_per_second": 181.515,
"step": 1500
},
{
"epoch": 2.13,
"learning_rate": 9.92885094272501e-06,
"loss": 0.0488,
"step": 2000
},
{
"epoch": 2.13,
"eval_loss": 0.05046902969479561,
"eval_micro_f1": 0.89676840165434,
"eval_runtime": 47.1831,
"eval_samples_per_second": 180.743,
"step": 2000
},
{
"epoch": 2.67,
"learning_rate": 9.911063678406261e-06,
"loss": 0.041,
"step": 2500
},
{
"epoch": 2.67,
"eval_loss": 0.0494869165122509,
"eval_micro_f1": 0.9004850003980875,
"eval_runtime": 43.5595,
"eval_samples_per_second": 195.778,
"step": 2500
},
{
"epoch": 3.2,
"learning_rate": 9.893276414087513e-06,
"loss": 0.0391,
"step": 3000
},
{
"epoch": 3.2,
"eval_loss": 0.05162811279296875,
"eval_micro_f1": 0.9040532684815226,
"eval_runtime": 43.6064,
"eval_samples_per_second": 195.568,
"step": 3000
},
{
"epoch": 3.74,
"learning_rate": 9.875489149768766e-06,
"loss": 0.0351,
"step": 3500
},
{