ForgeLoc-R1 / baselines /af_video-r1 /trainer_state.json
sdzt's picture
Add baselines/af_video-r1
86f65f2 verified
Raw History Blame Contribute Delete
474 kB
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 4.0,
"eval_steps": 500,
"global_step": 956,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"completion_length": 21.984375,
"epoch": 0.0041841004184100415,
"grad_norm": 4.002626895904541,
"kl": 0.0,
"learning_rate": 1e-06,
"loss": -0.0,
"reward": 0.28539182245731354,
"reward_std": 0.22613096237182617,
"reward_std_all": 0.2739841192960739,
"rewards-max/forensics_tempsamp_iou_reward": 0.6741725504398346,
"rewards/forensics_tempsamp_iou_reward": 0.1791418120265007,
"step": 1
},
{
"completion_length": 23.234375,
"epoch": 0.008368200836820083,
"grad_norm": 3.0689070224761963,
"kl": 0.00039768218994140625,
"learning_rate": 9.989539748953975e-07,
"loss": 0.0,
"reward": 0.3482043147087097,
"reward_std": 0.26919345557689667,
"reward_std_all": 0.27954597771167755,
"rewards-max/forensics_tempsamp_iou_reward": 0.8965044915676117,
"rewards/forensics_tempsamp_iou_reward": 0.22945428639650345,
"step": 2
},
{
"completion_length": 23.890625,
"epoch": 0.012552301255230125,
"grad_norm": 4.079133033752441,
"kl": 0.0012216567993164062,
"learning_rate": 9.97907949790795e-07,
"loss": 0.0,
"reward": 0.27900373935699463,
"reward_std": 0.18316103145480156,
"reward_std_all": 0.17254893109202385,
"rewards-max/forensics_tempsamp_iou_reward": 0.8209690153598785,
"rewards/forensics_tempsamp_iou_reward": 0.16025373339653015,
"step": 3
},
{
"completion_length": 20.53125,
"epoch": 0.016736401673640166,
"grad_norm": 3.8261938095092773,
"kl": 0.00220489501953125,
"learning_rate": 9.968619246861925e-07,
"loss": 0.0001,
"reward": 0.2939131259918213,
"reward_std": 0.19231359660625458,
"reward_std_all": 0.2519846111536026,
"rewards-max/forensics_tempsamp_iou_reward": 0.5918509364128113,
"rewards/forensics_tempsamp_iou_reward": 0.18766313046216965,
"step": 4
},
{
"completion_length": 20.734375,
"epoch": 0.02092050209205021,
"grad_norm": 2.9332010746002197,
"kl": 0.00167083740234375,
"learning_rate": 9.9581589958159e-07,
"loss": 0.0001,
"reward": 0.34954069554805756,
"reward_std": 0.1505582109093666,
"reward_std_all": 0.17685284465551376,
"rewards-max/forensics_tempsamp_iou_reward": 0.7182429134845734,
"rewards/forensics_tempsamp_iou_reward": 0.22454067319631577,
"step": 5
},
{
"completion_length": 20.5,
"epoch": 0.02510460251046025,
"grad_norm": 3.5646421909332275,
"kl": 0.00402069091796875,
"learning_rate": 9.947698744769873e-07,
"loss": 0.0002,
"reward": 0.33065760135650635,
"reward_std": 0.21486569941043854,
"reward_std_all": 0.21098600327968597,
"rewards-max/forensics_tempsamp_iou_reward": 0.7804670035839081,
"rewards/forensics_tempsamp_iou_reward": 0.19940760731697083,
"step": 6
},
{
"completion_length": 20.46875,
"epoch": 0.029288702928870293,
"grad_norm": 2.9464504718780518,
"kl": 0.01776123046875,
"learning_rate": 9.937238493723848e-07,
"loss": 0.0007,
"reward": 0.3994917720556259,
"reward_std": 0.20303894579410553,
"reward_std_all": 0.22648047655820847,
"rewards-max/forensics_tempsamp_iou_reward": 0.8951578140258789,
"rewards/forensics_tempsamp_iou_reward": 0.28074176609516144,
"step": 7
},
{
"completion_length": 19.875,
"epoch": 0.03347280334728033,
"grad_norm": 2.5315914154052734,
"kl": 0.0151824951171875,
"learning_rate": 9.926778242677823e-07,
"loss": 0.0006,
"reward": 0.42974166572093964,
"reward_std": 0.2075742930173874,
"reward_std_all": 0.18582271039485931,
"rewards-max/forensics_tempsamp_iou_reward": 0.6181280612945557,
"rewards/forensics_tempsamp_iou_reward": 0.2547416314482689,
"step": 8
},
{
"completion_length": 20.15625,
"epoch": 0.03765690376569038,
"grad_norm": 3.144050359725952,
"kl": 0.012969970703125,
"learning_rate": 9.916317991631798e-07,
"loss": 0.0005,
"reward": 0.4936816692352295,
"reward_std": 0.2946309670805931,
"reward_std_all": 0.3069722652435303,
"rewards-max/forensics_tempsamp_iou_reward": 0.9581174254417419,
"rewards/forensics_tempsamp_iou_reward": 0.3311816453933716,
"step": 9
},
{
"completion_length": 20.40625,
"epoch": 0.04184100418410042,
"grad_norm": 3.048820972442627,
"kl": 0.0408935546875,
"learning_rate": 9.905857740585774e-07,
"loss": 0.0016,
"reward": 0.3776208311319351,
"reward_std": 0.1174989715218544,
"reward_std_all": 0.10746251791715622,
"rewards-max/forensics_tempsamp_iou_reward": 0.7676825523376465,
"rewards/forensics_tempsamp_iou_reward": 0.23387081176042557,
"step": 10
},
{
"completion_length": 19.96875,
"epoch": 0.04602510460251046,
"grad_norm": 3.1604318618774414,
"kl": 0.03955078125,
"learning_rate": 9.895397489539749e-07,
"loss": 0.0016,
"reward": 0.48226746916770935,
"reward_std": 0.15548361092805862,
"reward_std_all": 0.17289111018180847,
"rewards-max/forensics_tempsamp_iou_reward": 0.7029362320899963,
"rewards/forensics_tempsamp_iou_reward": 0.29476750642061234,
"step": 11
},
{
"completion_length": 19.34375,
"epoch": 0.0502092050209205,
"grad_norm": 2.8502626419067383,
"kl": 0.0711669921875,
"learning_rate": 9.884937238493724e-07,
"loss": 0.0028,
"reward": 0.4281023293733597,
"reward_std": 0.21333851665258408,
"reward_std_all": 0.2409406378865242,
"rewards-max/forensics_tempsamp_iou_reward": 0.7232193648815155,
"rewards/forensics_tempsamp_iou_reward": 0.25935229659080505,
"step": 12
},
{
"completion_length": 20.15625,
"epoch": 0.05439330543933055,
"grad_norm": 3.115363121032715,
"kl": 0.06103515625,
"learning_rate": 9.8744769874477e-07,
"loss": 0.0024,
"reward": 0.5466255992650986,
"reward_std": 0.1836034543812275,
"reward_std_all": 0.1855858489871025,
"rewards-max/forensics_tempsamp_iou_reward": 0.7762897312641144,
"rewards/forensics_tempsamp_iou_reward": 0.3341256007552147,
"step": 13
},
{
"completion_length": 23.328125,
"epoch": 0.058577405857740586,
"grad_norm": 2.787935972213745,
"kl": 0.0849609375,
"learning_rate": 9.864016736401674e-07,
"loss": 0.0034,
"reward": 0.5251421332359314,
"reward_std": 0.21701569855213165,
"reward_std_all": 0.1910037100315094,
"rewards-max/forensics_tempsamp_iou_reward": 0.7928536832332611,
"rewards/forensics_tempsamp_iou_reward": 0.3376421332359314,
"step": 14
},
{
"completion_length": 19.875,
"epoch": 0.06276150627615062,
"grad_norm": 2.9273109436035156,
"kl": 0.10693359375,
"learning_rate": 9.85355648535565e-07,
"loss": 0.0043,
"reward": 0.4844385087490082,
"reward_std": 0.12383098527789116,
"reward_std_all": 0.1293032243847847,
"rewards-max/forensics_tempsamp_iou_reward": 0.8170822560787201,
"rewards/forensics_tempsamp_iou_reward": 0.3031885027885437,
"step": 15
},
{
"completion_length": 21.9375,
"epoch": 0.06694560669456066,
"grad_norm": 3.2158007621765137,
"kl": 0.078369140625,
"learning_rate": 9.843096234309622e-07,
"loss": 0.0031,
"reward": 0.4770485907793045,
"reward_std": 0.14747874438762665,
"reward_std_all": 0.15439076721668243,
"rewards-max/forensics_tempsamp_iou_reward": 0.7447517514228821,
"rewards/forensics_tempsamp_iou_reward": 0.30204860866069794,
"step": 16
},
{
"completion_length": 20.140625,
"epoch": 0.07112970711297072,
"grad_norm": 2.852602958679199,
"kl": 0.11181640625,
"learning_rate": 9.832635983263597e-07,
"loss": 0.0045,
"reward": 0.5448223352432251,
"reward_std": 0.14728567749261856,
"reward_std_all": 0.14770068228244781,
"rewards-max/forensics_tempsamp_iou_reward": 0.6963790357112885,
"rewards/forensics_tempsamp_iou_reward": 0.33857230842113495,
"step": 17
},
{
"completion_length": 20.59375,
"epoch": 0.07531380753138076,
"grad_norm": 2.828254222869873,
"kl": 0.099853515625,
"learning_rate": 9.822175732217572e-07,
"loss": 0.004,
"reward": 0.5693179368972778,
"reward_std": 0.10102610290050507,
"reward_std_all": 0.12296112626791,
"rewards-max/forensics_tempsamp_iou_reward": 0.8109112679958344,
"rewards/forensics_tempsamp_iou_reward": 0.31931792199611664,
"step": 18
},
{
"completion_length": 23.28125,
"epoch": 0.0794979079497908,
"grad_norm": 2.6202967166900635,
"kl": 0.0888671875,
"learning_rate": 9.811715481171548e-07,
"loss": 0.0036,
"reward": 0.3989129364490509,
"reward_std": 0.10542266070842743,
"reward_std_all": 0.10888835415244102,
"rewards-max/forensics_tempsamp_iou_reward": 0.5177173912525177,
"rewards/forensics_tempsamp_iou_reward": 0.23016294091939926,
"step": 19
},
{
"completion_length": 21.125,
"epoch": 0.08368200836820083,
"grad_norm": 3.672427177429199,
"kl": 0.11572265625,
"learning_rate": 9.801255230125523e-07,
"loss": 0.0046,
"reward": 0.4362296164035797,
"reward_std": 0.15003864839673042,
"reward_std_all": 0.1765109747648239,
"rewards-max/forensics_tempsamp_iou_reward": 0.8368471562862396,
"rewards/forensics_tempsamp_iou_reward": 0.29247958958148956,
"step": 20
},
{
"completion_length": 21.90625,
"epoch": 0.08786610878661087,
"grad_norm": 3.730201244354248,
"kl": 0.154052734375,
"learning_rate": 9.790794979079498e-07,
"loss": 0.0061,
"reward": 0.4817183017730713,
"reward_std": 0.06642618030309677,
"reward_std_all": 0.0641812365502119,
"rewards-max/forensics_tempsamp_iou_reward": 0.6459777355194092,
"rewards/forensics_tempsamp_iou_reward": 0.2942183017730713,
"step": 21
},
{
"completion_length": 19.9375,
"epoch": 0.09205020920502092,
"grad_norm": 3.454014301300049,
"kl": 0.15576171875,
"learning_rate": 9.780334728033473e-07,
"loss": 0.0062,
"reward": 0.5083868205547333,
"reward_std": 0.15881125628948212,
"reward_std_all": 0.15300720185041428,
"rewards-max/forensics_tempsamp_iou_reward": 0.7269566059112549,
"rewards/forensics_tempsamp_iou_reward": 0.30213677883148193,
"step": 22
},
{
"completion_length": 21.859375,
"epoch": 0.09623430962343096,
"grad_norm": 3.6931312084198,
"kl": 0.123046875,
"learning_rate": 9.769874476987448e-07,
"loss": 0.0049,
"reward": 0.5710337460041046,
"reward_std": 0.08614809066057205,
"reward_std_all": 0.11662689223885536,
"rewards-max/forensics_tempsamp_iou_reward": 0.725141704082489,
"rewards/forensics_tempsamp_iou_reward": 0.32728374004364014,
"step": 23
},
{
"completion_length": 22.296875,
"epoch": 0.100418410041841,
"grad_norm": 3.52529239654541,
"kl": 0.11865234375,
"learning_rate": 9.759414225941423e-07,
"loss": 0.0048,
"reward": 0.6978468596935272,
"reward_std": 0.1488838978111744,
"reward_std_all": 0.18451911211013794,
"rewards-max/forensics_tempsamp_iou_reward": 0.7951519787311554,
"rewards/forensics_tempsamp_iou_reward": 0.435346856713295,
"step": 24
},
{
"completion_length": 20.171875,
"epoch": 0.10460251046025104,
"grad_norm": 3.1676409244537354,
"kl": 0.14404296875,
"learning_rate": 9.748953974895398e-07,
"loss": 0.0058,
"reward": 0.502629742026329,
"reward_std": 0.06766787730157375,
"reward_std_all": 0.06529182195663452,
"rewards-max/forensics_tempsamp_iou_reward": 0.8744308054447174,
"rewards/forensics_tempsamp_iou_reward": 0.3338797241449356,
"step": 25
},
{
"completion_length": 20.34375,
"epoch": 0.1087866108786611,
"grad_norm": 3.0699055194854736,
"kl": 0.10693359375,
"learning_rate": 9.738493723849371e-07,
"loss": 0.0043,
"reward": 0.6422387361526489,
"reward_std": 0.0897088460624218,
"reward_std_all": 0.11040650308132172,
"rewards-max/forensics_tempsamp_iou_reward": 0.922232449054718,
"rewards/forensics_tempsamp_iou_reward": 0.4109887033700943,
"step": 26
},
{
"completion_length": 20.59375,
"epoch": 0.11297071129707113,
"grad_norm": 2.519780397415161,
"kl": 0.12890625,
"learning_rate": 9.728033472803347e-07,
"loss": 0.0051,
"reward": 0.5828422605991364,
"reward_std": 0.09332774952054024,
"reward_std_all": 0.09598035737872124,
"rewards-max/forensics_tempsamp_iou_reward": 0.8354969024658203,
"rewards/forensics_tempsamp_iou_reward": 0.3703422248363495,
"step": 27
},
{
"completion_length": 21.65625,
"epoch": 0.11715481171548117,
"grad_norm": 1.9970245361328125,
"kl": 0.140625,
"learning_rate": 9.717573221757322e-07,
"loss": 0.0056,
"reward": 0.7342773079872131,
"reward_std": 0.08063654974102974,
"reward_std_all": 0.09899628907442093,
"rewards-max/forensics_tempsamp_iou_reward": 0.8297416567802429,
"rewards/forensics_tempsamp_iou_reward": 0.45927731692790985,
"step": 28
},
{
"completion_length": 19.96875,
"epoch": 0.12133891213389121,
"grad_norm": 2.158614158630371,
"kl": 0.094970703125,
"learning_rate": 9.707112970711297e-07,
"loss": 0.0038,
"reward": 0.45566099882125854,
"reward_std": 0.06186266243457794,
"reward_std_all": 0.08661800622940063,
"rewards-max/forensics_tempsamp_iou_reward": 0.6258953958749771,
"rewards/forensics_tempsamp_iou_reward": 0.26191096007823944,
"step": 29
},
{
"completion_length": 20.953125,
"epoch": 0.12552301255230125,
"grad_norm": 2.213346242904663,
"kl": 0.14501953125,
"learning_rate": 9.696652719665272e-07,
"loss": 0.0058,
"reward": 0.5066787749528885,
"reward_std": 0.06552152708172798,
"reward_std_all": 0.08695441484451294,
"rewards-max/forensics_tempsamp_iou_reward": 0.6378747522830963,
"rewards/forensics_tempsamp_iou_reward": 0.30667877197265625,
"step": 30
},
{
"completion_length": 21.421875,
"epoch": 0.1297071129707113,
"grad_norm": 2.7957892417907715,
"kl": 0.103271484375,
"learning_rate": 9.686192468619247e-07,
"loss": 0.0041,
"reward": 0.5303078889846802,
"reward_std": 0.09799080714583397,
"reward_std_all": 0.08879905194044113,
"rewards-max/forensics_tempsamp_iou_reward": 0.8672501444816589,
"rewards/forensics_tempsamp_iou_reward": 0.3240578770637512,
"step": 31
},
{
"completion_length": 22.5625,
"epoch": 0.13389121338912133,
"grad_norm": 3.2633309364318848,
"kl": 0.115478515625,
"learning_rate": 9.675732217573222e-07,
"loss": 0.0046,
"reward": 0.6072786152362823,
"reward_std": 0.16296730190515518,
"reward_std_all": 0.1455642506480217,
"rewards-max/forensics_tempsamp_iou_reward": 0.8572685718536377,
"rewards/forensics_tempsamp_iou_reward": 0.35727857053279877,
"step": 32
},
{
"completion_length": 21.328125,
"epoch": 0.13807531380753138,
"grad_norm": 1.9105486869812012,
"kl": 0.08056640625,
"learning_rate": 9.665271966527197e-07,
"loss": 0.0032,
"reward": 0.48113588988780975,
"reward_std": 0.09388535842299461,
"reward_std_all": 0.08686132729053497,
"rewards-max/forensics_tempsamp_iou_reward": 0.6988419890403748,
"rewards/forensics_tempsamp_iou_reward": 0.28738586604595184,
"step": 33
},
{
"completion_length": 22.265625,
"epoch": 0.14225941422594143,
"grad_norm": 3.043543815612793,
"kl": 0.11962890625,
"learning_rate": 9.654811715481172e-07,
"loss": 0.0048,
"reward": 0.4665468782186508,
"reward_std": 0.12931611388921738,
"reward_std_all": 0.1472834274172783,
"rewards-max/forensics_tempsamp_iou_reward": 0.8738099634647369,
"rewards/forensics_tempsamp_iou_reward": 0.28529684990644455,
"step": 34
},
{
"completion_length": 19.59375,
"epoch": 0.14644351464435146,
"grad_norm": 4.505025863647461,
"kl": 0.085693359375,
"learning_rate": 9.644351464435145e-07,
"loss": 0.0034,
"reward": 0.41333964467048645,
"reward_std": 0.09820215404033661,
"reward_std_all": 0.10587918385863304,
"rewards-max/forensics_tempsamp_iou_reward": 0.825681209564209,
"rewards/forensics_tempsamp_iou_reward": 0.24458961933851242,
"step": 35
},
{
"completion_length": 19.375,
"epoch": 0.1506276150627615,
"grad_norm": 2.7706079483032227,
"kl": 0.092529296875,
"learning_rate": 9.63389121338912e-07,
"loss": 0.0037,
"reward": 0.5252922177314758,
"reward_std": 0.09615125507116318,
"reward_std_all": 0.08963784575462341,
"rewards-max/forensics_tempsamp_iou_reward": 0.9081679880619049,
"rewards/forensics_tempsamp_iou_reward": 0.31279221177101135,
"step": 36
},
{
"completion_length": 19.265625,
"epoch": 0.15481171548117154,
"grad_norm": 3.5488247871398926,
"kl": 0.124267578125,
"learning_rate": 9.623430962343096e-07,
"loss": 0.005,
"reward": 0.6461915969848633,
"reward_std": 0.07091345638036728,
"reward_std_all": 0.11846798285841942,
"rewards-max/forensics_tempsamp_iou_reward": 0.665217399597168,
"rewards/forensics_tempsamp_iou_reward": 0.37744152545928955,
"step": 37
},
{
"completion_length": 19.75,
"epoch": 0.1589958158995816,
"grad_norm": 3.2155003547668457,
"kl": 0.10498046875,
"learning_rate": 9.61297071129707e-07,
"loss": 0.0042,
"reward": 0.588426947593689,
"reward_std": 0.1311771497130394,
"reward_std_all": 0.1558668315410614,
"rewards-max/forensics_tempsamp_iou_reward": 0.9008106291294098,
"rewards/forensics_tempsamp_iou_reward": 0.35092693567276,
"step": 38
},
{
"completion_length": 19.71875,
"epoch": 0.16317991631799164,
"grad_norm": 3.3254075050354004,
"kl": 0.13818359375,
"learning_rate": 9.602510460251046e-07,
"loss": 0.0055,
"reward": 0.5944716334342957,
"reward_std": 0.09832578524947166,
"reward_std_all": 0.09869503974914551,
"rewards-max/forensics_tempsamp_iou_reward": 0.8230858743190765,
"rewards/forensics_tempsamp_iou_reward": 0.3569715917110443,
"step": 39
},
{
"completion_length": 19.25,
"epoch": 0.16736401673640167,
"grad_norm": 2.9132349491119385,
"kl": 0.149658203125,
"learning_rate": 9.592050209205021e-07,
"loss": 0.006,
"reward": 0.5858767628669739,
"reward_std": 0.09860999509692192,
"reward_std_all": 0.09570219367742538,
"rewards-max/forensics_tempsamp_iou_reward": 0.6457344442605972,
"rewards/forensics_tempsamp_iou_reward": 0.3483767509460449,
"step": 40
},
{
"completion_length": 20.296875,
"epoch": 0.17154811715481172,
"grad_norm": 2.6258935928344727,
"kl": 0.13720703125,
"learning_rate": 9.581589958158996e-07,
"loss": 0.0055,
"reward": 0.39064235985279083,
"reward_std": 0.10514245927333832,
"reward_std_all": 0.12689443677663803,
"rewards-max/forensics_tempsamp_iou_reward": 0.7308205664157867,
"rewards/forensics_tempsamp_iou_reward": 0.24689233303070068,
"step": 41
},
{
"completion_length": 20.515625,
"epoch": 0.17573221757322174,
"grad_norm": 2.6272406578063965,
"kl": 0.16357421875,
"learning_rate": 9.571129707112971e-07,
"loss": 0.0065,
"reward": 0.3817840814590454,
"reward_std": 0.07638367265462875,
"reward_std_all": 0.08244697377085686,
"rewards-max/forensics_tempsamp_iou_reward": 0.49710923433303833,
"rewards/forensics_tempsamp_iou_reward": 0.20053409039974213,
"step": 42
},
{
"completion_length": 19.328125,
"epoch": 0.1799163179916318,
"grad_norm": 2.7698416709899902,
"kl": 0.14111328125,
"learning_rate": 9.560669456066946e-07,
"loss": 0.0056,
"reward": 0.3923322856426239,
"reward_std": 0.13843297585844994,
"reward_std_all": 0.13809527456760406,
"rewards-max/forensics_tempsamp_iou_reward": 0.8166723251342773,
"rewards/forensics_tempsamp_iou_reward": 0.2610822767019272,
"step": 43
},
{
"completion_length": 19.46875,
"epoch": 0.18410041841004185,
"grad_norm": 3.2652626037597656,
"kl": 0.177734375,
"learning_rate": 9.55020920502092e-07,
"loss": 0.0071,
"reward": 0.5747921466827393,
"reward_std": 0.09316155686974525,
"reward_std_all": 0.08351097628474236,
"rewards-max/forensics_tempsamp_iou_reward": 0.768417090177536,
"rewards/forensics_tempsamp_iou_reward": 0.35604216158390045,
"step": 44
},
{
"completion_length": 19.328125,
"epoch": 0.18828451882845187,
"grad_norm": 2.882613182067871,
"kl": 0.12890625,
"learning_rate": 9.539748953974895e-07,
"loss": 0.0052,
"reward": 0.4682290256023407,
"reward_std": 0.11332917958498001,
"reward_std_all": 0.12170657515525818,
"rewards-max/forensics_tempsamp_iou_reward": 0.7037151753902435,
"rewards/forensics_tempsamp_iou_reward": 0.2994789779186249,
"step": 45
},
{
"completion_length": 19.421875,
"epoch": 0.19246861924686193,
"grad_norm": 3.04581880569458,
"kl": 0.1396484375,
"learning_rate": 9.52928870292887e-07,
"loss": 0.0056,
"reward": 0.396413654088974,
"reward_std": 0.11112528294324875,
"reward_std_all": 0.12239022552967072,
"rewards-max/forensics_tempsamp_iou_reward": 0.807574063539505,
"rewards/forensics_tempsamp_iou_reward": 0.25266364216804504,
"step": 46
},
{
"completion_length": 19.703125,
"epoch": 0.19665271966527198,
"grad_norm": 3.104187488555908,
"kl": 0.169921875,
"learning_rate": 9.518828451882844e-07,
"loss": 0.0068,
"reward": 0.41276146471500397,
"reward_std": 0.11429809033870697,
"reward_std_all": 0.12308334559202194,
"rewards-max/forensics_tempsamp_iou_reward": 0.8038274049758911,
"rewards/forensics_tempsamp_iou_reward": 0.269011452794075,
"step": 47
},
{
"completion_length": 19.6875,
"epoch": 0.200836820083682,
"grad_norm": 4.872344970703125,
"kl": 0.32470703125,
"learning_rate": 9.508368200836819e-07,
"loss": 0.013,
"reward": 0.3616201877593994,
"reward_std": 0.14493507519364357,
"reward_std_all": 0.17492150515317917,
"rewards-max/forensics_tempsamp_iou_reward": 0.629606693983078,
"rewards/forensics_tempsamp_iou_reward": 0.22412020713090897,
"step": 48
},
{
"completion_length": 19.25,
"epoch": 0.20502092050209206,
"grad_norm": 2.6950430870056152,
"kl": 0.14013671875,
"learning_rate": 9.497907949790794e-07,
"loss": 0.0056,
"reward": 0.5935229659080505,
"reward_std": 0.10658245719969273,
"reward_std_all": 0.10621866583824158,
"rewards-max/forensics_tempsamp_iou_reward": 0.8391513526439667,
"rewards/forensics_tempsamp_iou_reward": 0.3872729539871216,
"step": 49
},
{
"completion_length": 19.6875,
"epoch": 0.20920502092050208,
"grad_norm": 3.0384693145751953,
"kl": 0.16162109375,
"learning_rate": 9.487447698744769e-07,
"loss": 0.0065,
"reward": 0.4618772268295288,
"reward_std": 0.10610320419073105,
"reward_std_all": 0.10689757019281387,
"rewards-max/forensics_tempsamp_iou_reward": 0.7903801798820496,
"rewards/forensics_tempsamp_iou_reward": 0.2743772119283676,
"step": 50
},
{
"completion_length": 19.453125,
"epoch": 0.21338912133891214,
"grad_norm": 2.5915472507476807,
"kl": 0.17578125,
"learning_rate": 9.476987447698744e-07,
"loss": 0.007,
"reward": 0.5972027480602264,
"reward_std": 0.13585232570767403,
"reward_std_all": 0.15704957395792007,
"rewards-max/forensics_tempsamp_iou_reward": 0.8957147300243378,
"rewards/forensics_tempsamp_iou_reward": 0.3972027003765106,
"step": 51
},
{
"completion_length": 19.609375,
"epoch": 0.2175732217573222,
"grad_norm": 3.0597615242004395,
"kl": 0.178955078125,
"learning_rate": 9.466527196652719e-07,
"loss": 0.0072,
"reward": 0.4948941469192505,
"reward_std": 0.09020435810089111,
"reward_std_all": 0.08365194499492645,
"rewards-max/forensics_tempsamp_iou_reward": 0.7833373546600342,
"rewards/forensics_tempsamp_iou_reward": 0.29489411413669586,
"step": 52
},
{
"completion_length": 19.40625,
"epoch": 0.2217573221757322,
"grad_norm": 2.5452880859375,
"kl": 0.160400390625,
"learning_rate": 9.456066945606695e-07,
"loss": 0.0064,
"reward": 0.48835210502147675,
"reward_std": 0.06942499801516533,
"reward_std_all": 0.08204708062112331,
"rewards-max/forensics_tempsamp_iou_reward": 0.795153796672821,
"rewards/forensics_tempsamp_iou_reward": 0.29460208117961884,
"step": 53
},
{
"completion_length": 19.125,
"epoch": 0.22594142259414227,
"grad_norm": 2.8443920612335205,
"kl": 0.158203125,
"learning_rate": 9.44560669456067e-07,
"loss": 0.0063,
"reward": 0.6090547442436218,
"reward_std": 0.09924175962805748,
"reward_std_all": 0.10903307422995567,
"rewards-max/forensics_tempsamp_iou_reward": 0.9791119992733002,
"rewards/forensics_tempsamp_iou_reward": 0.3903047740459442,
"step": 54
},
{
"completion_length": 19.796875,
"epoch": 0.2301255230125523,
"grad_norm": 3.01182222366333,
"kl": 0.132080078125,
"learning_rate": 9.435146443514645e-07,
"loss": 0.0053,
"reward": 0.5780399143695831,
"reward_std": 0.11745812371373177,
"reward_std_all": 0.1141616515815258,
"rewards-max/forensics_tempsamp_iou_reward": 0.8660606741905212,
"rewards/forensics_tempsamp_iou_reward": 0.3530399203300476,
"step": 55
},
{
"completion_length": 21.296875,
"epoch": 0.23430962343096234,
"grad_norm": 2.911484956741333,
"kl": 0.19775390625,
"learning_rate": 9.424686192468618e-07,
"loss": 0.0079,
"reward": 0.4774651974439621,
"reward_std": 0.1423085704445839,
"reward_std_all": 0.14563608914613724,
"rewards-max/forensics_tempsamp_iou_reward": 0.9101599752902985,
"rewards/forensics_tempsamp_iou_reward": 0.2899651825428009,
"step": 56
},
{
"completion_length": 19.3125,
"epoch": 0.2384937238493724,
"grad_norm": 3.1688504219055176,
"kl": 0.10498046875,
"learning_rate": 9.414225941422593e-07,
"loss": 0.0042,
"reward": 0.6184263825416565,
"reward_std": 0.09830507263541222,
"reward_std_all": 0.09904542937874794,
"rewards-max/forensics_tempsamp_iou_reward": 0.908462256193161,
"rewards/forensics_tempsamp_iou_reward": 0.4309263527393341,
"step": 57
},
{
"completion_length": 21.515625,
"epoch": 0.24267782426778242,
"grad_norm": 2.391568899154663,
"kl": 0.09765625,
"learning_rate": 9.403765690376568e-07,
"loss": 0.0039,
"reward": 0.6287861466407776,
"reward_std": 0.10878083482384682,
"reward_std_all": 0.106081772595644,
"rewards-max/forensics_tempsamp_iou_reward": 0.8520551025867462,
"rewards/forensics_tempsamp_iou_reward": 0.39128613471984863,
"step": 58
},
{
"completion_length": 20.421875,
"epoch": 0.24686192468619247,
"grad_norm": 2.8486862182617188,
"kl": 0.113525390625,
"learning_rate": 9.393305439330543e-07,
"loss": 0.0045,
"reward": 0.6490055322647095,
"reward_std": 0.16409865766763687,
"reward_std_all": 0.1534586027264595,
"rewards-max/forensics_tempsamp_iou_reward": 0.8877239525318146,
"rewards/forensics_tempsamp_iou_reward": 0.4365055114030838,
"step": 59
},
{
"completion_length": 20.578125,
"epoch": 0.2510460251046025,
"grad_norm": 3.5631942749023438,
"kl": 0.119384765625,
"learning_rate": 9.382845188284518e-07,
"loss": 0.0048,
"reward": 0.5227818489074707,
"reward_std": 0.17131606489419937,
"reward_std_all": 0.1592504382133484,
"rewards-max/forensics_tempsamp_iou_reward": 0.8584839701652527,
"rewards/forensics_tempsamp_iou_reward": 0.31653186678886414,
"step": 60
},
{
"completion_length": 20.953125,
"epoch": 0.25523012552301255,
"grad_norm": 3.4410464763641357,
"kl": 0.099365234375,
"learning_rate": 9.372384937238493e-07,
"loss": 0.004,
"reward": 0.6969785094261169,
"reward_std": 0.0951855406165123,
"reward_std_all": 0.11064835637807846,
"rewards-max/forensics_tempsamp_iou_reward": 0.8606584668159485,
"rewards/forensics_tempsamp_iou_reward": 0.4219784736633301,
"step": 61
},
{
"completion_length": 20.234375,
"epoch": 0.2594142259414226,
"grad_norm": 3.0359599590301514,
"kl": 0.1357421875,
"learning_rate": 9.361924686192469e-07,
"loss": 0.0054,
"reward": 0.4599929451942444,
"reward_std": 0.13580086082220078,
"reward_std_all": 0.1506674513220787,
"rewards-max/forensics_tempsamp_iou_reward": 0.8222869336605072,
"rewards/forensics_tempsamp_iou_reward": 0.2724929004907608,
"step": 62
},
{
"completion_length": 19.78125,
"epoch": 0.26359832635983266,
"grad_norm": 3.102254629135132,
"kl": 0.1298828125,
"learning_rate": 9.351464435146444e-07,
"loss": 0.0052,
"reward": 0.45794084668159485,
"reward_std": 0.1059187613427639,
"reward_std_all": 0.11500073224306107,
"rewards-max/forensics_tempsamp_iou_reward": 0.7993985414505005,
"rewards/forensics_tempsamp_iou_reward": 0.3141908049583435,
"step": 63
},
{
"completion_length": 21.390625,
"epoch": 0.26778242677824265,
"grad_norm": 2.9873480796813965,
"kl": 0.1650390625,
"learning_rate": 9.341004184100419e-07,
"loss": 0.0066,
"reward": 0.548797607421875,
"reward_std": 0.14933108538389206,
"reward_std_all": 0.17247333377599716,
"rewards-max/forensics_tempsamp_iou_reward": 0.8796218633651733,
"rewards/forensics_tempsamp_iou_reward": 0.3550475686788559,
"step": 64
},
{
"completion_length": 19.625,
"epoch": 0.2719665271966527,
"grad_norm": 2.914397716522217,
"kl": 0.162841796875,
"learning_rate": 9.330543933054393e-07,
"loss": 0.0065,
"reward": 0.4271164685487747,
"reward_std": 0.176457017660141,
"reward_std_all": 0.16859494149684906,
"rewards-max/forensics_tempsamp_iou_reward": 0.6577979624271393,
"rewards/forensics_tempsamp_iou_reward": 0.2646164447069168,
"step": 65
},
{
"completion_length": 21.015625,
"epoch": 0.27615062761506276,
"grad_norm": 2.940323829650879,
"kl": 0.12744140625,
"learning_rate": 9.320083682008367e-07,
"loss": 0.0051,
"reward": 0.4613303691148758,
"reward_std": 0.20807650685310364,
"reward_std_all": 0.24901169538497925,
"rewards-max/forensics_tempsamp_iou_reward": 0.8082454204559326,
"rewards/forensics_tempsamp_iou_reward": 0.28633034229278564,
"step": 66
},
{
"completion_length": 20.171875,
"epoch": 0.2803347280334728,
"grad_norm": 3.4595773220062256,
"kl": 0.108154296875,
"learning_rate": 9.309623430962342e-07,
"loss": 0.0043,
"reward": 0.6237208545207977,
"reward_std": 0.11725759506225586,
"reward_std_all": 0.14079942926764488,
"rewards-max/forensics_tempsamp_iou_reward": 0.8169123530387878,
"rewards/forensics_tempsamp_iou_reward": 0.40497082471847534,
"step": 67
},
{
"completion_length": 19.796875,
"epoch": 0.28451882845188287,
"grad_norm": 3.079150676727295,
"kl": 0.1640625,
"learning_rate": 9.299163179916317e-07,
"loss": 0.0066,
"reward": 0.42203669250011444,
"reward_std": 0.1659204661846161,
"reward_std_all": 0.1470695659518242,
"rewards-max/forensics_tempsamp_iou_reward": 0.8457357883453369,
"rewards/forensics_tempsamp_iou_reward": 0.2845366895198822,
"step": 68
},
{
"completion_length": 20.578125,
"epoch": 0.28870292887029286,
"grad_norm": 2.881035566329956,
"kl": 0.130126953125,
"learning_rate": 9.288702928870292e-07,
"loss": 0.0052,
"reward": 0.5574732422828674,
"reward_std": 0.17233731597661972,
"reward_std_all": 0.16549167037010193,
"rewards-max/forensics_tempsamp_iou_reward": 0.9052507877349854,
"rewards/forensics_tempsamp_iou_reward": 0.357473224401474,
"step": 69
},
{
"completion_length": 21.0625,
"epoch": 0.2928870292887029,
"grad_norm": 2.4439470767974854,
"kl": 0.1103515625,
"learning_rate": 9.278242677824268e-07,
"loss": 0.0044,
"reward": 0.49735109508037567,
"reward_std": 0.17203575745224953,
"reward_std_all": 0.18366790935397148,
"rewards-max/forensics_tempsamp_iou_reward": 0.7783780097961426,
"rewards/forensics_tempsamp_iou_reward": 0.27860110253095627,
"step": 70
},
{
"completion_length": 20.109375,
"epoch": 0.29707112970711297,
"grad_norm": 2.723458766937256,
"kl": 0.110107421875,
"learning_rate": 9.267782426778243e-07,
"loss": 0.0044,
"reward": 0.4916437566280365,
"reward_std": 0.1247636191546917,
"reward_std_all": 0.13529766723513603,
"rewards-max/forensics_tempsamp_iou_reward": 0.7954279780387878,
"rewards/forensics_tempsamp_iou_reward": 0.3291437029838562,
"step": 71
},
{
"completion_length": 20.5,
"epoch": 0.301255230125523,
"grad_norm": 2.779613494873047,
"kl": 0.11669921875,
"learning_rate": 9.257322175732218e-07,
"loss": 0.0047,
"reward": 0.5586876571178436,
"reward_std": 0.09621255099773407,
"reward_std_all": 0.10631534457206726,
"rewards-max/forensics_tempsamp_iou_reward": 0.9270484745502472,
"rewards/forensics_tempsamp_iou_reward": 0.3899376392364502,
"step": 72
},
{
"completion_length": 19.78125,
"epoch": 0.3054393305439331,
"grad_norm": 2.5644805431365967,
"kl": 0.14501953125,
"learning_rate": 9.246861924686192e-07,
"loss": 0.0058,
"reward": 0.5118520259857178,
"reward_std": 0.09660181403160095,
"reward_std_all": 0.11941217631101608,
"rewards-max/forensics_tempsamp_iou_reward": 0.8100393712520599,
"rewards/forensics_tempsamp_iou_reward": 0.31810200214385986,
"step": 73
},
{
"completion_length": 21.25,
"epoch": 0.30962343096234307,
"grad_norm": 2.8691303730010986,
"kl": 0.19970703125,
"learning_rate": 9.236401673640167e-07,
"loss": 0.008,
"reward": 0.4545446187257767,
"reward_std": 0.20929507911205292,
"reward_std_all": 0.186872698366642,
"rewards-max/forensics_tempsamp_iou_reward": 0.9758556187152863,
"rewards/forensics_tempsamp_iou_reward": 0.2982945889234543,
"step": 74
},
{
"completion_length": 20.5625,
"epoch": 0.3138075313807531,
"grad_norm": 2.3553333282470703,
"kl": 0.158447265625,
"learning_rate": 9.225941422594142e-07,
"loss": 0.0063,
"reward": 0.5489102005958557,
"reward_std": 0.17545481026172638,
"reward_std_all": 0.17468653619289398,
"rewards-max/forensics_tempsamp_iou_reward": 0.9595651924610138,
"rewards/forensics_tempsamp_iou_reward": 0.3801601827144623,
"step": 75
},
{
"completion_length": 21.25,
"epoch": 0.3179916317991632,
"grad_norm": 2.7080085277557373,
"kl": 0.127197265625,
"learning_rate": 9.215481171548116e-07,
"loss": 0.0051,
"reward": 0.6435751020908356,
"reward_std": 0.15118231996893883,
"reward_std_all": 0.1565876603126526,
"rewards-max/forensics_tempsamp_iou_reward": 0.8845405876636505,
"rewards/forensics_tempsamp_iou_reward": 0.42482510209083557,
"step": 76
},
{
"completion_length": 19.78125,
"epoch": 0.32217573221757323,
"grad_norm": 2.679133176803589,
"kl": 0.104248046875,
"learning_rate": 9.205020920502091e-07,
"loss": 0.0042,
"reward": 0.6479278206825256,
"reward_std": 0.14752116426825523,
"reward_std_all": 0.13395550847053528,
"rewards-max/forensics_tempsamp_iou_reward": 0.8823254704475403,
"rewards/forensics_tempsamp_iou_reward": 0.41042785346508026,
"step": 77
},
{
"completion_length": 21.265625,
"epoch": 0.3263598326359833,
"grad_norm": 3.6068270206451416,
"kl": 0.21484375,
"learning_rate": 9.194560669456066e-07,
"loss": 0.0086,
"reward": 0.46172989904880524,
"reward_std": 0.15821558982133865,
"reward_std_all": 0.15209771692752838,
"rewards-max/forensics_tempsamp_iou_reward": 0.7356246709823608,
"rewards/forensics_tempsamp_iou_reward": 0.292979896068573,
"step": 78
},
{
"completion_length": 22.46875,
"epoch": 0.3305439330543933,
"grad_norm": 2.6990373134613037,
"kl": 0.112548828125,
"learning_rate": 9.184100418410042e-07,
"loss": 0.0045,
"reward": 0.5244334638118744,
"reward_std": 0.08303441479802132,
"reward_std_all": 0.09305348247289658,
"rewards-max/forensics_tempsamp_iou_reward": 0.9404352307319641,
"rewards/forensics_tempsamp_iou_reward": 0.29943346977233887,
"step": 79
},
{
"completion_length": 19.5,
"epoch": 0.33472803347280333,
"grad_norm": 2.8264448642730713,
"kl": 0.135498046875,
"learning_rate": 9.173640167364017e-07,
"loss": 0.0054,
"reward": 0.5639037787914276,
"reward_std": 0.08217206969857216,
"reward_std_all": 0.10821424797177315,
"rewards-max/forensics_tempsamp_iou_reward": 0.867822140455246,
"rewards/forensics_tempsamp_iou_reward": 0.36390379071235657,
"step": 80
},
{
"completion_length": 20.28125,
"epoch": 0.3389121338912134,
"grad_norm": 2.7570881843566895,
"kl": 0.114013671875,
"learning_rate": 9.163179916317991e-07,
"loss": 0.0046,
"reward": 0.6428556442260742,
"reward_std": 0.13854744657874107,
"reward_std_all": 0.16615717858076096,
"rewards-max/forensics_tempsamp_iou_reward": 0.8808064758777618,
"rewards/forensics_tempsamp_iou_reward": 0.40535569190979004,
"step": 81
},
{
"completion_length": 20.9375,
"epoch": 0.34309623430962344,
"grad_norm": 2.5826575756073,
"kl": 0.140380859375,
"learning_rate": 9.152719665271966e-07,
"loss": 0.0056,
"reward": 0.5414496064186096,
"reward_std": 0.09446835517883301,
"reward_std_all": 0.09971410781145096,
"rewards-max/forensics_tempsamp_iou_reward": 0.8173652589321136,
"rewards/forensics_tempsamp_iou_reward": 0.3414496034383774,
"step": 82
},
{
"completion_length": 21.890625,
"epoch": 0.3472803347280335,
"grad_norm": 3.6538844108581543,
"kl": 0.123046875,
"learning_rate": 9.142259414225941e-07,
"loss": 0.0049,
"reward": 0.5600086748600006,
"reward_std": 0.13395218178629875,
"reward_std_all": 0.1936071813106537,
"rewards-max/forensics_tempsamp_iou_reward": 0.9001511633396149,
"rewards/forensics_tempsamp_iou_reward": 0.38500866293907166,
"step": 83
},
{
"completion_length": 20.796875,
"epoch": 0.3514644351464435,
"grad_norm": 3.264040946960449,
"kl": 0.142333984375,
"learning_rate": 9.131799163179916e-07,
"loss": 0.0057,
"reward": 0.5018160343170166,
"reward_std": 0.19062145054340363,
"reward_std_all": 0.1707218810915947,
"rewards-max/forensics_tempsamp_iou_reward": 0.9334363341331482,
"rewards/forensics_tempsamp_iou_reward": 0.32056601345539093,
"step": 84
},
{
"completion_length": 21.203125,
"epoch": 0.35564853556485354,
"grad_norm": 2.7290194034576416,
"kl": 0.0986328125,
"learning_rate": 9.121338912133891e-07,
"loss": 0.0039,
"reward": 0.5837530493736267,
"reward_std": 0.13614744693040848,
"reward_std_all": 0.1283663474023342,
"rewards-max/forensics_tempsamp_iou_reward": 0.9076276421546936,
"rewards/forensics_tempsamp_iou_reward": 0.41500304639339447,
"step": 85
},
{
"completion_length": 19.921875,
"epoch": 0.3598326359832636,
"grad_norm": 2.5453505516052246,
"kl": 0.144775390625,
"learning_rate": 9.110878661087865e-07,
"loss": 0.0058,
"reward": 0.537450522184372,
"reward_std": 0.10638216882944107,
"reward_std_all": 0.10008173063397408,
"rewards-max/forensics_tempsamp_iou_reward": 0.8307786583900452,
"rewards/forensics_tempsamp_iou_reward": 0.34370049834251404,
"step": 86
},
{
"completion_length": 23.109375,
"epoch": 0.36401673640167365,
"grad_norm": 3.0083727836608887,
"kl": 0.115966796875,
"learning_rate": 9.10041841004184e-07,
"loss": 0.0046,
"reward": 0.5294220745563507,
"reward_std": 0.13034368306398392,
"reward_std_all": 0.11388064175844193,
"rewards-max/forensics_tempsamp_iou_reward": 0.8815121352672577,
"rewards/forensics_tempsamp_iou_reward": 0.32317204773426056,
"step": 87
},
{
"completion_length": 20.015625,
"epoch": 0.3682008368200837,
"grad_norm": 3.0256857872009277,
"kl": 0.1572265625,
"learning_rate": 9.089958158995816e-07,
"loss": 0.0063,
"reward": 0.5603882670402527,
"reward_std": 0.15884365141391754,
"reward_std_all": 0.16377758979797363,
"rewards-max/forensics_tempsamp_iou_reward": 0.8297566175460815,
"rewards/forensics_tempsamp_iou_reward": 0.33538827300071716,
"step": 88
},
{
"completion_length": 19.34375,
"epoch": 0.3723849372384937,
"grad_norm": 2.6224913597106934,
"kl": 0.095458984375,
"learning_rate": 9.07949790794979e-07,
"loss": 0.0038,
"reward": 0.524655893445015,
"reward_std": 0.12259066477417946,
"reward_std_all": 0.1401650384068489,
"rewards-max/forensics_tempsamp_iou_reward": 0.9600845873355865,
"rewards/forensics_tempsamp_iou_reward": 0.33715587854385376,
"step": 89
},
{
"completion_length": 20.03125,
"epoch": 0.37656903765690375,
"grad_norm": 3.388490676879883,
"kl": 0.15283203125,
"learning_rate": 9.069037656903765e-07,
"loss": 0.0061,
"reward": 0.6477646827697754,
"reward_std": 0.07609965279698372,
"reward_std_all": 0.08818608522415161,
"rewards-max/forensics_tempsamp_iou_reward": 0.882521778345108,
"rewards/forensics_tempsamp_iou_reward": 0.41026467084884644,
"step": 90
},
{
"completion_length": 20.109375,
"epoch": 0.3807531380753138,
"grad_norm": 3.508925199508667,
"kl": 0.1884765625,
"learning_rate": 9.05857740585774e-07,
"loss": 0.0075,
"reward": 0.5745222419500351,
"reward_std": 0.08311565406620502,
"reward_std_all": 0.0873735137283802,
"rewards-max/forensics_tempsamp_iou_reward": 0.816737949848175,
"rewards/forensics_tempsamp_iou_reward": 0.3495222181081772,
"step": 91
},
{
"completion_length": 20.0625,
"epoch": 0.38493723849372385,
"grad_norm": 3.1605613231658936,
"kl": 0.130859375,
"learning_rate": 9.048117154811715e-07,
"loss": 0.0052,
"reward": 0.5416333079338074,
"reward_std": 0.10614687949419022,
"reward_std_all": 0.11964894458651543,
"rewards-max/forensics_tempsamp_iou_reward": 0.7582160234451294,
"rewards/forensics_tempsamp_iou_reward": 0.3291332721710205,
"step": 92
},
{
"completion_length": 19.390625,
"epoch": 0.3891213389121339,
"grad_norm": 3.5907034873962402,
"kl": 0.122802734375,
"learning_rate": 9.03765690376569e-07,
"loss": 0.0049,
"reward": 0.7001588642597198,
"reward_std": 0.1073397696018219,
"reward_std_all": 0.11316484585404396,
"rewards-max/forensics_tempsamp_iou_reward": 0.9511682689189911,
"rewards/forensics_tempsamp_iou_reward": 0.47515885531902313,
"step": 93
},
{
"completion_length": 19.1875,
"epoch": 0.39330543933054396,
"grad_norm": 4.301107883453369,
"kl": 0.11279296875,
"learning_rate": 9.027196652719665e-07,
"loss": 0.0045,
"reward": 0.4007134586572647,
"reward_std": 0.08340590074658394,
"reward_std_all": 0.10192243754863739,
"rewards-max/forensics_tempsamp_iou_reward": 0.780954897403717,
"rewards/forensics_tempsamp_iou_reward": 0.2444634586572647,
"step": 94
},
{
"completion_length": 19.546875,
"epoch": 0.39748953974895396,
"grad_norm": 2.6216611862182617,
"kl": 0.17578125,
"learning_rate": 9.01673640167364e-07,
"loss": 0.007,
"reward": 0.5914106070995331,
"reward_std": 0.0699322260916233,
"reward_std_all": 0.07492172718048096,
"rewards-max/forensics_tempsamp_iou_reward": 0.7070062905550003,
"rewards/forensics_tempsamp_iou_reward": 0.366410568356514,
"step": 95
},
{
"completion_length": 19.4375,
"epoch": 0.401673640167364,
"grad_norm": 3.522223472595215,
"kl": 0.1435546875,
"learning_rate": 9.006276150627615e-07,
"loss": 0.0058,
"reward": 0.5373824536800385,
"reward_std": 0.13062452524900436,
"reward_std_all": 0.13325966894626617,
"rewards-max/forensics_tempsamp_iou_reward": 0.8548015356063843,
"rewards/forensics_tempsamp_iou_reward": 0.3373824506998062,
"step": 96
},
{
"completion_length": 19.125,
"epoch": 0.40585774058577406,
"grad_norm": 3.333587884902954,
"kl": 0.17138671875,
"learning_rate": 8.99581589958159e-07,
"loss": 0.0068,
"reward": 0.5803160220384598,
"reward_std": 0.13402831554412842,
"reward_std_all": 0.1399470791220665,
"rewards-max/forensics_tempsamp_iou_reward": 0.7050029337406158,
"rewards/forensics_tempsamp_iou_reward": 0.3303160071372986,
"step": 97
},
{
"completion_length": 19.1875,
"epoch": 0.4100418410041841,
"grad_norm": 3.6864044666290283,
"kl": 0.109130859375,
"learning_rate": 8.985355648535564e-07,
"loss": 0.0044,
"reward": 0.6821727454662323,
"reward_std": 0.08076934516429901,
"reward_std_all": 0.07701401039958,
"rewards-max/forensics_tempsamp_iou_reward": 0.8251543343067169,
"rewards/forensics_tempsamp_iou_reward": 0.41967275738716125,
"step": 98
},
{
"completion_length": 19.375,
"epoch": 0.41422594142259417,
"grad_norm": 3.1023166179656982,
"kl": 0.13525390625,
"learning_rate": 8.974895397489539e-07,
"loss": 0.0054,
"reward": 0.8070760071277618,
"reward_std": 0.09482868388295174,
"reward_std_all": 0.1042621061205864,
"rewards-max/forensics_tempsamp_iou_reward": 0.9468148052692413,
"rewards/forensics_tempsamp_iou_reward": 0.5383259505033493,
"step": 99
},
{
"completion_length": 19.90625,
"epoch": 0.41841004184100417,
"grad_norm": 2.743863105773926,
"kl": 0.123779296875,
"learning_rate": 8.964435146443514e-07,
"loss": 0.005,
"reward": 0.4657103270292282,
"reward_std": 0.05713762901723385,
"reward_std_all": 0.06655811332166195,
"rewards-max/forensics_tempsamp_iou_reward": 0.8011219203472137,
"rewards/forensics_tempsamp_iou_reward": 0.278210312128067,
"step": 100
},
{
"completion_length": 19.046875,
"epoch": 0.4225941422594142,
"grad_norm": 2.8819234371185303,
"kl": 0.1767578125,
"learning_rate": 8.953974895397489e-07,
"loss": 0.0071,
"reward": 0.46632735431194305,
"reward_std": 0.12375126406550407,
"reward_std_all": 0.11176029406487942,
"rewards-max/forensics_tempsamp_iou_reward": 0.9048571288585663,
"rewards/forensics_tempsamp_iou_reward": 0.2975773513317108,
"step": 101
},
{
"completion_length": 19.203125,
"epoch": 0.42677824267782427,
"grad_norm": 2.6387970447540283,
"kl": 0.154296875,
"learning_rate": 8.943514644351464e-07,
"loss": 0.0062,
"reward": 0.6157117784023285,
"reward_std": 0.09256022796034813,
"reward_std_all": 0.09056741744279861,
"rewards-max/forensics_tempsamp_iou_reward": 0.8932575881481171,
"rewards/forensics_tempsamp_iou_reward": 0.37821173667907715,
"step": 102
},
{
"completion_length": 21.234375,
"epoch": 0.4309623430962343,
"grad_norm": 2.3753914833068848,
"kl": 0.137939453125,
"learning_rate": 8.933054393305439e-07,
"loss": 0.0055,
"reward": 0.4807351529598236,
"reward_std": 0.08148669078946114,
"reward_std_all": 0.08245671167969704,
"rewards-max/forensics_tempsamp_iou_reward": 0.8257184624671936,
"rewards/forensics_tempsamp_iou_reward": 0.311985120177269,
"step": 103
},
{
"completion_length": 19.375,
"epoch": 0.4351464435146444,
"grad_norm": 3.61794376373291,
"kl": 0.203125,
"learning_rate": 8.922594142259414e-07,
"loss": 0.0081,
"reward": 0.573622077703476,
"reward_std": 0.14514803886413574,
"reward_std_all": 0.13427859544754028,
"rewards-max/forensics_tempsamp_iou_reward": 0.8813057243824005,
"rewards/forensics_tempsamp_iou_reward": 0.3673720359802246,
"step": 104
},
{
"completion_length": 19.59375,
"epoch": 0.4393305439330544,
"grad_norm": 2.83689546585083,
"kl": 0.13623046875,
"learning_rate": 8.91213389121339e-07,
"loss": 0.0054,
"reward": 0.6099750101566315,
"reward_std": 0.11428412795066833,
"reward_std_all": 0.11192092671990395,
"rewards-max/forensics_tempsamp_iou_reward": 0.9034736454486847,
"rewards/forensics_tempsamp_iou_reward": 0.40997497737407684,
"step": 105
},
{
"completion_length": 19.46875,
"epoch": 0.4435146443514644,
"grad_norm": 3.1489293575286865,
"kl": 0.20947265625,
"learning_rate": 8.901673640167363e-07,
"loss": 0.0084,
"reward": 0.7348154485225677,
"reward_std": 0.08802950754761696,
"reward_std_all": 0.10468461364507675,
"rewards-max/forensics_tempsamp_iou_reward": 0.9567954242229462,
"rewards/forensics_tempsamp_iou_reward": 0.4785654991865158,
"step": 106
},
{
"completion_length": 19.625,
"epoch": 0.4476987447698745,
"grad_norm": 3.4177987575531006,
"kl": 0.17626953125,
"learning_rate": 8.891213389121338e-07,
"loss": 0.0071,
"reward": 0.7763835787773132,
"reward_std": 0.09483499825000763,
"reward_std_all": 0.12147068232297897,
"rewards-max/forensics_tempsamp_iou_reward": 0.9416802227497101,
"rewards/forensics_tempsamp_iou_reward": 0.5201335549354553,
"step": 107
},
{
"completion_length": 19.546875,
"epoch": 0.45188284518828453,
"grad_norm": 3.4469003677368164,
"kl": 0.1259765625,
"learning_rate": 8.880753138075313e-07,
"loss": 0.005,
"reward": 0.6357665956020355,
"reward_std": 0.08507804572582245,
"reward_std_all": 0.12414195761084557,
"rewards-max/forensics_tempsamp_iou_reward": 0.8827847242355347,
"rewards/forensics_tempsamp_iou_reward": 0.41701658070087433,
"step": 108
},
{
"completion_length": 19.3125,
"epoch": 0.4560669456066946,
"grad_norm": 3.4268412590026855,
"kl": 0.16259765625,
"learning_rate": 8.870292887029288e-07,
"loss": 0.0065,
"reward": 0.5584841668605804,
"reward_std": 0.13704559206962585,
"reward_std_all": 0.14479601383209229,
"rewards-max/forensics_tempsamp_iou_reward": 0.912041574716568,
"rewards/forensics_tempsamp_iou_reward": 0.358484148979187,
"step": 109
},
{
"completion_length": 19.546875,
"epoch": 0.4602510460251046,
"grad_norm": 2.518594264984131,
"kl": 0.1318359375,
"learning_rate": 8.859832635983263e-07,
"loss": 0.0053,
"reward": 0.6805315911769867,
"reward_std": 0.12360748648643494,
"reward_std_all": 0.1182909607887268,
"rewards-max/forensics_tempsamp_iou_reward": 0.9068763256072998,
"rewards/forensics_tempsamp_iou_reward": 0.4742816388607025,
"step": 110
},
{
"completion_length": 19.6875,
"epoch": 0.46443514644351463,
"grad_norm": 5.168785095214844,
"kl": 0.37353515625,
"learning_rate": 8.849372384937238e-07,
"loss": 0.0149,
"reward": 0.5722489058971405,
"reward_std": 0.11356965452432632,
"reward_std_all": 0.1257285252213478,
"rewards-max/forensics_tempsamp_iou_reward": 0.9028934240341187,
"rewards/forensics_tempsamp_iou_reward": 0.35974887013435364,
"step": 111
},
{
"completion_length": 19.453125,
"epoch": 0.4686192468619247,
"grad_norm": 2.9813051223754883,
"kl": 0.139404296875,
"learning_rate": 8.838912133891213e-07,
"loss": 0.0056,
"reward": 0.5408274531364441,
"reward_std": 0.12304104119539261,
"reward_std_all": 0.12511645257472992,
"rewards-max/forensics_tempsamp_iou_reward": 0.8407265543937683,
"rewards/forensics_tempsamp_iou_reward": 0.3595774471759796,
"step": 112
},
{
"completion_length": 19.46875,
"epoch": 0.47280334728033474,
"grad_norm": 2.716073989868164,
"kl": 0.16357421875,
"learning_rate": 8.828451882845189e-07,
"loss": 0.0065,
"reward": 0.6493111252784729,
"reward_std": 0.05750467628240585,
"reward_std_all": 0.10015316307544708,
"rewards-max/forensics_tempsamp_iou_reward": 0.8836049437522888,
"rewards/forensics_tempsamp_iou_reward": 0.38681110739707947,
"step": 113
},
{
"completion_length": 19.34375,
"epoch": 0.4769874476987448,
"grad_norm": 3.3483874797821045,
"kl": 0.188232421875,
"learning_rate": 8.817991631799164e-07,
"loss": 0.0076,
"reward": 0.5238321721553802,
"reward_std": 0.11165852099657059,
"reward_std_all": 0.12265713885426521,
"rewards-max/forensics_tempsamp_iou_reward": 0.8397451937198639,
"rewards/forensics_tempsamp_iou_reward": 0.34258218109607697,
"step": 114
},
{
"completion_length": 19.890625,
"epoch": 0.4811715481171548,
"grad_norm": 2.9880664348602295,
"kl": 0.095947265625,
"learning_rate": 8.807531380753138e-07,
"loss": 0.0038,
"reward": 0.5255553424358368,
"reward_std": 0.16401567310094833,
"reward_std_all": 0.1501457691192627,
"rewards-max/forensics_tempsamp_iou_reward": 0.9050915837287903,
"rewards/forensics_tempsamp_iou_reward": 0.3443053066730499,
"step": 115
},
{
"completion_length": 20.75,
"epoch": 0.48535564853556484,
"grad_norm": 2.703202724456787,
"kl": 0.097900390625,
"learning_rate": 8.797071129707112e-07,
"loss": 0.0039,
"reward": 0.563753604888916,
"reward_std": 0.10999362915754318,
"reward_std_all": 0.11372164636850357,
"rewards-max/forensics_tempsamp_iou_reward": 0.9164182245731354,
"rewards/forensics_tempsamp_iou_reward": 0.34500357508659363,
"step": 116
},
{
"completion_length": 19.953125,
"epoch": 0.4895397489539749,
"grad_norm": 3.0983669757843018,
"kl": 0.108642578125,
"learning_rate": 8.786610878661087e-07,
"loss": 0.0044,
"reward": 0.6306578516960144,
"reward_std": 0.1378808729350567,
"reward_std_all": 0.13978073745965958,
"rewards-max/forensics_tempsamp_iou_reward": 0.8773194849491119,
"rewards/forensics_tempsamp_iou_reward": 0.41815780103206635,
"step": 117
},
{
"completion_length": 19.484375,
"epoch": 0.49372384937238495,
"grad_norm": 4.01816463470459,
"kl": 0.11572265625,
"learning_rate": 8.776150627615062e-07,
"loss": 0.0046,
"reward": 0.4994307607412338,
"reward_std": 0.09322045370936394,
"reward_std_all": 0.0965443067252636,
"rewards-max/forensics_tempsamp_iou_reward": 0.8348630368709564,
"rewards/forensics_tempsamp_iou_reward": 0.34943071007728577,
"step": 118
},
{
"completion_length": 19.40625,
"epoch": 0.497907949790795,
"grad_norm": 3.2857773303985596,
"kl": 0.10986328125,
"learning_rate": 8.765690376569037e-07,
"loss": 0.0044,
"reward": 0.663908839225769,
"reward_std": 0.09742334857583046,
"reward_std_all": 0.1003817766904831,
"rewards-max/forensics_tempsamp_iou_reward": 0.915103405714035,
"rewards/forensics_tempsamp_iou_reward": 0.420158788561821,
"step": 119
},
{
"completion_length": 19.1875,
"epoch": 0.502092050209205,
"grad_norm": 3.7879159450531006,
"kl": 0.08837890625,
"learning_rate": 8.755230125523012e-07,
"loss": 0.0035,
"reward": 0.6525634527206421,
"reward_std": 0.08133722841739655,
"reward_std_all": 0.0806664228439331,
"rewards-max/forensics_tempsamp_iou_reward": 0.9502071142196655,
"rewards/forensics_tempsamp_iou_reward": 0.4650634378194809,
"step": 120
},
{
"completion_length": 19.890625,
"epoch": 0.5062761506276151,
"grad_norm": 3.491384983062744,
"kl": 0.096923828125,
"learning_rate": 8.744769874476987e-07,
"loss": 0.0039,
"reward": 0.7614250481128693,
"reward_std": 0.06946882233023643,
"reward_std_all": 0.10384571179747581,
"rewards-max/forensics_tempsamp_iou_reward": 0.9355865716934204,
"rewards/forensics_tempsamp_iou_reward": 0.5176750421524048,
"step": 121
},
{
"completion_length": 19.71875,
"epoch": 0.5104602510460251,
"grad_norm": 3.5175740718841553,
"kl": 0.216064453125,
"learning_rate": 8.734309623430963e-07,
"loss": 0.0086,
"reward": 0.42754748463630676,
"reward_std": 0.10989250987768173,
"reward_std_all": 0.11991768702864647,
"rewards-max/forensics_tempsamp_iou_reward": 0.8580279350280762,
"rewards/forensics_tempsamp_iou_reward": 0.25879746675491333,
"step": 122
},
{
"completion_length": 20.03125,
"epoch": 0.5146443514644351,
"grad_norm": 2.6294078826904297,
"kl": 0.1083984375,
"learning_rate": 8.723849372384937e-07,
"loss": 0.0043,
"reward": 0.5376076400279999,
"reward_std": 0.09668216854333878,
"reward_std_all": 0.11213506758213043,
"rewards-max/forensics_tempsamp_iou_reward": 0.7276684641838074,
"rewards/forensics_tempsamp_iou_reward": 0.33760765194892883,
"step": 123
},
{
"completion_length": 20.171875,
"epoch": 0.5188284518828452,
"grad_norm": 2.9117889404296875,
"kl": 0.095703125,
"learning_rate": 8.713389121338912e-07,
"loss": 0.0038,
"reward": 0.6464064419269562,
"reward_std": 0.09913130849599838,
"reward_std_all": 0.11221035942435265,
"rewards-max/forensics_tempsamp_iou_reward": 0.9282352328300476,
"rewards/forensics_tempsamp_iou_reward": 0.41515645384788513,
"step": 124
},
{
"completion_length": 19.796875,
"epoch": 0.5230125523012552,
"grad_norm": 3.049792766571045,
"kl": 0.099609375,
"learning_rate": 8.702928870292887e-07,
"loss": 0.004,
"reward": 0.6394393742084503,
"reward_std": 0.10049973800778389,
"reward_std_all": 0.10296470299363136,
"rewards-max/forensics_tempsamp_iou_reward": 0.8860098719596863,
"rewards/forensics_tempsamp_iou_reward": 0.4206893593072891,
"step": 125
},
{
"completion_length": 20.28125,
"epoch": 0.5271966527196653,
"grad_norm": 2.7640981674194336,
"kl": 0.093994140625,
"learning_rate": 8.692468619246861e-07,
"loss": 0.0038,
"reward": 0.7162806391716003,
"reward_std": 0.07534628920257092,
"reward_std_all": 0.08709894865751266,
"rewards-max/forensics_tempsamp_iou_reward": 0.7907997071743011,
"rewards/forensics_tempsamp_iou_reward": 0.43503066897392273,
"step": 126
},
{
"completion_length": 19.703125,
"epoch": 0.5313807531380753,
"grad_norm": 2.470123052597046,
"kl": 0.08203125,
"learning_rate": 8.682008368200836e-07,
"loss": 0.0033,
"reward": 0.7533744871616364,
"reward_std": 0.10253429412841797,
"reward_std_all": 0.11671300232410431,
"rewards-max/forensics_tempsamp_iou_reward": 0.9602864682674408,
"rewards/forensics_tempsamp_iou_reward": 0.5096244812011719,
"step": 127
},
{
"completion_length": 20.1875,
"epoch": 0.5355648535564853,
"grad_norm": 3.3922882080078125,
"kl": 0.08203125,
"learning_rate": 8.671548117154811e-07,
"loss": 0.0033,
"reward": 0.6069016754627228,
"reward_std": 0.10482721030712128,
"reward_std_all": 0.11041027307510376,
"rewards-max/forensics_tempsamp_iou_reward": 0.9339583218097687,
"rewards/forensics_tempsamp_iou_reward": 0.3944016546010971,
"step": 128
},
{
"completion_length": 20.03125,
"epoch": 0.5397489539748954,
"grad_norm": 2.7597382068634033,
"kl": 0.08447265625,
"learning_rate": 8.661087866108786e-07,
"loss": 0.0034,
"reward": 0.6423919498920441,
"reward_std": 0.08990656584501266,
"reward_std_all": 0.08616301417350769,
"rewards-max/forensics_tempsamp_iou_reward": 0.9240418374538422,
"rewards/forensics_tempsamp_iou_reward": 0.4611419588327408,
"step": 129
},
{
"completion_length": 19.75,
"epoch": 0.5439330543933054,
"grad_norm": 4.291018009185791,
"kl": 0.139892578125,
"learning_rate": 8.650627615062761e-07,
"loss": 0.0056,
"reward": 0.6174978613853455,
"reward_std": 0.1197182647883892,
"reward_std_all": 0.10610115900635719,
"rewards-max/forensics_tempsamp_iou_reward": 0.7394010722637177,
"rewards/forensics_tempsamp_iou_reward": 0.35499782860279083,
"step": 130
},
{
"completion_length": 20.234375,
"epoch": 0.5481171548117155,
"grad_norm": 3.0171236991882324,
"kl": 0.11376953125,
"learning_rate": 8.640167364016737e-07,
"loss": 0.0045,
"reward": 0.4117601215839386,
"reward_std": 0.14897873252630234,
"reward_std_all": 0.1318921186029911,
"rewards-max/forensics_tempsamp_iou_reward": 0.8174600899219513,
"rewards/forensics_tempsamp_iou_reward": 0.26801010966300964,
"step": 131
},
{
"completion_length": 19.90625,
"epoch": 0.5523012552301255,
"grad_norm": 3.703280210494995,
"kl": 0.102783203125,
"learning_rate": 8.629707112970711e-07,
"loss": 0.0041,
"reward": 0.7159865200519562,
"reward_std": 0.08132747560739517,
"reward_std_all": 0.10887794941663742,
"rewards-max/forensics_tempsamp_iou_reward": 0.9532874226570129,
"rewards/forensics_tempsamp_iou_reward": 0.4722365140914917,
"step": 132
},
{
"completion_length": 20.3125,
"epoch": 0.5564853556485355,
"grad_norm": 3.684596538543701,
"kl": 0.115478515625,
"learning_rate": 8.619246861924686e-07,
"loss": 0.0046,
"reward": 0.5029393136501312,
"reward_std": 0.10861274600028992,
"reward_std_all": 0.12947794795036316,
"rewards-max/forensics_tempsamp_iou_reward": 0.8091475069522858,
"rewards/forensics_tempsamp_iou_reward": 0.31543926894664764,
"step": 133
},
{
"completion_length": 19.78125,
"epoch": 0.5606694560669456,
"grad_norm": 3.312018394470215,
"kl": 0.12548828125,
"learning_rate": 8.608786610878661e-07,
"loss": 0.005,
"reward": 0.7007938623428345,
"reward_std": 0.07947684824466705,
"reward_std_all": 0.07821428030729294,
"rewards-max/forensics_tempsamp_iou_reward": 0.9106453061103821,
"rewards/forensics_tempsamp_iou_reward": 0.4195438027381897,
"step": 134
},
{
"completion_length": 20.59375,
"epoch": 0.5648535564853556,
"grad_norm": 3.725069522857666,
"kl": 0.138427734375,
"learning_rate": 8.598326359832636e-07,
"loss": 0.0055,
"reward": 0.5663236081600189,
"reward_std": 0.1311035305261612,
"reward_std_all": 0.12641489505767822,
"rewards-max/forensics_tempsamp_iou_reward": 0.8615466058254242,
"rewards/forensics_tempsamp_iou_reward": 0.35382357239723206,
"step": 135
},
{
"completion_length": 19.0,
"epoch": 0.5690376569037657,
"grad_norm": 2.757084369659424,
"kl": 0.1826171875,
"learning_rate": 8.58786610878661e-07,
"loss": 0.0073,
"reward": 0.5639887750148773,
"reward_std": 0.09152401611208916,
"reward_std_all": 0.10838593542575836,
"rewards-max/forensics_tempsamp_iou_reward": 0.9545789957046509,
"rewards/forensics_tempsamp_iou_reward": 0.35773877799510956,
"step": 136
},
{
"completion_length": 19.09375,
"epoch": 0.5732217573221757,
"grad_norm": 4.684041500091553,
"kl": 0.089599609375,
"learning_rate": 8.577405857740585e-07,
"loss": 0.0036,
"reward": 0.6443935036659241,
"reward_std": 0.12464046850800514,
"reward_std_all": 0.13631170615553856,
"rewards-max/forensics_tempsamp_iou_reward": 0.994156539440155,
"rewards/forensics_tempsamp_iou_reward": 0.4381434917449951,
"step": 137
},
{
"completion_length": 19.859375,
"epoch": 0.5774058577405857,
"grad_norm": 4.605729103088379,
"kl": 0.14794921875,
"learning_rate": 8.56694560669456e-07,
"loss": 0.0059,
"reward": 0.6449767351150513,
"reward_std": 0.134316548705101,
"reward_std_all": 0.13964197039604187,
"rewards-max/forensics_tempsamp_iou_reward": 0.9189265370368958,
"rewards/forensics_tempsamp_iou_reward": 0.42622675001621246,
"step": 138
},
{
"completion_length": 19.703125,
"epoch": 0.5815899581589958,
"grad_norm": 2.977785110473633,
"kl": 0.118896484375,
"learning_rate": 8.556485355648536e-07,
"loss": 0.0048,
"reward": 0.7730225026607513,
"reward_std": 0.08469157293438911,
"reward_std_all": 0.10889023169875145,
"rewards-max/forensics_tempsamp_iou_reward": 0.9013156890869141,
"rewards/forensics_tempsamp_iou_reward": 0.5292725116014481,
"step": 139
},
{
"completion_length": 19.484375,
"epoch": 0.5857740585774058,
"grad_norm": 3.1831421852111816,
"kl": 0.14794921875,
"learning_rate": 8.54602510460251e-07,
"loss": 0.0059,
"reward": 0.7621346414089203,
"reward_std": 0.12075449153780937,
"reward_std_all": 0.1277744248509407,
"rewards-max/forensics_tempsamp_iou_reward": 0.8965995013713837,
"rewards/forensics_tempsamp_iou_reward": 0.4808846712112427,
"step": 140
},
{
"completion_length": 19.5625,
"epoch": 0.5899581589958159,
"grad_norm": 3.96526837348938,
"kl": 0.09228515625,
"learning_rate": 8.535564853556485e-07,
"loss": 0.0037,
"reward": 0.7992464303970337,
"reward_std": 0.09363288804888725,
"reward_std_all": 0.0859257634729147,
"rewards-max/forensics_tempsamp_iou_reward": 0.943135142326355,
"rewards/forensics_tempsamp_iou_reward": 0.5554964244365692,
"step": 141
},
{
"completion_length": 20.1875,
"epoch": 0.5941422594142259,
"grad_norm": 2.7952589988708496,
"kl": 0.102294921875,
"learning_rate": 8.52510460251046e-07,
"loss": 0.0041,
"reward": 0.5597171783447266,
"reward_std": 0.15080993622541428,
"reward_std_all": 0.13793885335326195,
"rewards-max/forensics_tempsamp_iou_reward": 0.8772997856140137,
"rewards/forensics_tempsamp_iou_reward": 0.36596715450286865,
"step": 142
},
{
"completion_length": 19.984375,
"epoch": 0.5983263598326359,
"grad_norm": 3.114821434020996,
"kl": 0.1220703125,
"learning_rate": 8.514644351464435e-07,
"loss": 0.0049,
"reward": 0.6239852607250214,
"reward_std": 0.08609616011381149,
"reward_std_all": 0.08595015853643417,
"rewards-max/forensics_tempsamp_iou_reward": 0.9668470621109009,
"rewards/forensics_tempsamp_iou_reward": 0.4177352786064148,
"step": 143
},
{
"completion_length": 19.203125,
"epoch": 0.602510460251046,
"grad_norm": 3.173027753829956,
"kl": 0.126708984375,
"learning_rate": 8.50418410041841e-07,
"loss": 0.0051,
"reward": 0.7091273069381714,
"reward_std": 0.0885816179215908,
"reward_std_all": 0.08516302704811096,
"rewards-max/forensics_tempsamp_iou_reward": 0.9411433935165405,
"rewards/forensics_tempsamp_iou_reward": 0.4653773009777069,
"step": 144
},
{
"completion_length": 20.015625,
"epoch": 0.606694560669456,
"grad_norm": 4.5959978103637695,
"kl": 0.135986328125,
"learning_rate": 8.493723849372385e-07,
"loss": 0.0055,
"reward": 0.4787372052669525,
"reward_std": 0.10888916254043579,
"reward_std_all": 0.12533752247691154,
"rewards-max/forensics_tempsamp_iou_reward": 0.7306699305772781,
"rewards/forensics_tempsamp_iou_reward": 0.3099871873855591,
"step": 145
},
{
"completion_length": 19.671875,
"epoch": 0.6108786610878661,
"grad_norm": 3.0862998962402344,
"kl": 0.115966796875,
"learning_rate": 8.483263598326359e-07,
"loss": 0.0046,
"reward": 0.6556963920593262,
"reward_std": 0.15058965981006622,
"reward_std_all": 0.159757100045681,
"rewards-max/forensics_tempsamp_iou_reward": 0.8669654726982117,
"rewards/forensics_tempsamp_iou_reward": 0.43694639205932617,
"step": 146
},
{
"completion_length": 19.15625,
"epoch": 0.6150627615062761,
"grad_norm": 2.893033027648926,
"kl": 0.1201171875,
"learning_rate": 8.472803347280334e-07,
"loss": 0.0048,
"reward": 0.7848754525184631,
"reward_std": 0.10618319734930992,
"reward_std_all": 0.11609451472759247,
"rewards-max/forensics_tempsamp_iou_reward": 0.9871051013469696,
"rewards/forensics_tempsamp_iou_reward": 0.5286253988742828,
"step": 147
},
{
"completion_length": 19.046875,
"epoch": 0.6192468619246861,
"grad_norm": 3.475217580795288,
"kl": 0.1552734375,
"learning_rate": 8.462343096234308e-07,
"loss": 0.0062,
"reward": 0.5500079095363617,
"reward_std": 0.10092710703611374,
"reward_std_all": 0.10135620087385178,
"rewards-max/forensics_tempsamp_iou_reward": 0.8728050887584686,
"rewards/forensics_tempsamp_iou_reward": 0.38125792145729065,
"step": 148
},
{
"completion_length": 19.21875,
"epoch": 0.6234309623430963,
"grad_norm": 2.6733691692352295,
"kl": 0.105224609375,
"learning_rate": 8.451882845188284e-07,
"loss": 0.0042,
"reward": 0.5993291884660721,
"reward_std": 0.06402122788131237,
"reward_std_all": 0.06105411611497402,
"rewards-max/forensics_tempsamp_iou_reward": 0.7858968079090118,
"rewards/forensics_tempsamp_iou_reward": 0.3930791765451431,
"step": 149
},
{
"completion_length": 20.3125,
"epoch": 0.6276150627615062,
"grad_norm": 3.8565430641174316,
"kl": 0.3115234375,
"learning_rate": 8.441422594142259e-07,
"loss": 0.0124,
"reward": 0.493300199508667,
"reward_std": 0.10041061788797379,
"reward_std_all": 0.16417020559310913,
"rewards-max/forensics_tempsamp_iou_reward": 0.684904158115387,
"rewards/forensics_tempsamp_iou_reward": 0.28080014884471893,
"step": 150
},
{
"completion_length": 19.609375,
"epoch": 0.6317991631799164,
"grad_norm": 4.840059280395508,
"kl": 0.38671875,
"learning_rate": 8.430962343096234e-07,
"loss": 0.0154,
"reward": 0.623344212770462,
"reward_std": 0.10479855537414551,
"reward_std_all": 0.13236500322818756,
"rewards-max/forensics_tempsamp_iou_reward": 0.8849445283412933,
"rewards/forensics_tempsamp_iou_reward": 0.37334421277046204,
"step": 151
},
{
"completion_length": 20.125,
"epoch": 0.6359832635983264,
"grad_norm": 4.237565517425537,
"kl": 0.21533203125,
"learning_rate": 8.420502092050209e-07,
"loss": 0.0086,
"reward": 0.6248132884502411,
"reward_std": 0.09466112405061722,
"reward_std_all": 0.10595813393592834,
"rewards-max/forensics_tempsamp_iou_reward": 0.9069314002990723,
"rewards/forensics_tempsamp_iou_reward": 0.41231323778629303,
"step": 152
},
{
"completion_length": 19.671875,
"epoch": 0.6401673640167364,
"grad_norm": 4.066334247589111,
"kl": 0.17822265625,
"learning_rate": 8.410041841004184e-07,
"loss": 0.0071,
"reward": 0.6793281137943268,
"reward_std": 0.10661736875772476,
"reward_std_all": 0.09847169741988182,
"rewards-max/forensics_tempsamp_iou_reward": 0.9248282611370087,
"rewards/forensics_tempsamp_iou_reward": 0.4543280601501465,
"step": 153
},
{
"completion_length": 19.0,
"epoch": 0.6443514644351465,
"grad_norm": 5.887523174285889,
"kl": 0.14501953125,
"learning_rate": 8.399581589958159e-07,
"loss": 0.0058,
"reward": 0.5762135088443756,
"reward_std": 0.102585818618536,
"reward_std_all": 0.11943919211626053,
"rewards-max/forensics_tempsamp_iou_reward": 0.9726110100746155,
"rewards/forensics_tempsamp_iou_reward": 0.38871346414089203,
"step": 154
},
{
"completion_length": 20.484375,
"epoch": 0.6485355648535565,
"grad_norm": 3.2659943103790283,
"kl": 0.1240234375,
"learning_rate": 8.389121338912134e-07,
"loss": 0.005,
"reward": 0.6631394922733307,
"reward_std": 0.13347340747714043,
"reward_std_all": 0.14471854269504547,
"rewards-max/forensics_tempsamp_iou_reward": 0.9437447786331177,
"rewards/forensics_tempsamp_iou_reward": 0.4568895101547241,
"step": 155
},
{
"completion_length": 19.234375,
"epoch": 0.6527196652719666,
"grad_norm": 3.34234356880188,
"kl": 0.19970703125,
"learning_rate": 8.378661087866107e-07,
"loss": 0.008,
"reward": 0.5906393229961395,
"reward_std": 0.1377449333667755,
"reward_std_all": 0.14244873821735382,
"rewards-max/forensics_tempsamp_iou_reward": 0.8063830733299255,
"rewards/forensics_tempsamp_iou_reward": 0.3906393051147461,
"step": 156
},
{
"completion_length": 20.65625,
"epoch": 0.6569037656903766,
"grad_norm": 2.7090909481048584,
"kl": 0.100830078125,
"learning_rate": 8.368200836820083e-07,
"loss": 0.004,
"reward": 0.528645396232605,
"reward_std": 0.056314751505851746,
"reward_std_all": 0.0785459615290165,
"rewards-max/forensics_tempsamp_iou_reward": 0.7931099236011505,
"rewards/forensics_tempsamp_iou_reward": 0.3411453515291214,
"step": 157
},
{
"completion_length": 19.375,
"epoch": 0.6610878661087866,
"grad_norm": 3.113210678100586,
"kl": 0.106689453125,
"learning_rate": 8.357740585774058e-07,
"loss": 0.0043,
"reward": 0.6412312090396881,
"reward_std": 0.06803805567324162,
"reward_std_all": 0.08268315345048904,
"rewards-max/forensics_tempsamp_iou_reward": 0.8803583085536957,
"rewards/forensics_tempsamp_iou_reward": 0.43498116731643677,
"step": 158
},
{
"completion_length": 19.6875,
"epoch": 0.6652719665271967,
"grad_norm": 4.548762798309326,
"kl": 0.139404296875,
"learning_rate": 8.347280334728033e-07,
"loss": 0.0056,
"reward": 0.6290504932403564,
"reward_std": 0.08222807198762894,
"reward_std_all": 0.09418345242738724,
"rewards-max/forensics_tempsamp_iou_reward": 0.9216017127037048,
"rewards/forensics_tempsamp_iou_reward": 0.4228005111217499,
"step": 159
},
{
"completion_length": 19.5625,
"epoch": 0.6694560669456067,
"grad_norm": 4.2907233238220215,
"kl": 0.15625,
"learning_rate": 8.336820083682008e-07,
"loss": 0.0062,
"reward": 0.5531624853610992,
"reward_std": 0.12579137459397316,
"reward_std_all": 0.14344793558120728,
"rewards-max/forensics_tempsamp_iou_reward": 0.8949334919452667,
"rewards/forensics_tempsamp_iou_reward": 0.3844124674797058,
"step": 160
},
{
"completion_length": 21.0625,
"epoch": 0.6736401673640168,
"grad_norm": 3.9313104152679443,
"kl": 0.1552734375,
"learning_rate": 8.326359832635983e-07,
"loss": 0.0062,
"reward": 0.5714309811592102,
"reward_std": 0.0675281286239624,
"reward_std_all": 0.08629157394170761,
"rewards-max/forensics_tempsamp_iou_reward": 0.8716887533664703,
"rewards/forensics_tempsamp_iou_reward": 0.352680966258049,
"step": 161
},
{
"completion_length": 21.359375,
"epoch": 0.6778242677824268,
"grad_norm": 3.3187482357025146,
"kl": 0.1689453125,
"learning_rate": 8.315899581589958e-07,
"loss": 0.0067,
"reward": 0.7525539696216583,
"reward_std": 0.08226880803704262,
"reward_std_all": 0.1416272446513176,
"rewards-max/forensics_tempsamp_iou_reward": 0.8967974781990051,
"rewards/forensics_tempsamp_iou_reward": 0.4900539517402649,
"step": 162
},
{
"completion_length": 20.953125,
"epoch": 0.6820083682008368,
"grad_norm": 4.148557186126709,
"kl": 0.120361328125,
"learning_rate": 8.305439330543933e-07,
"loss": 0.0048,
"reward": 0.7093664109706879,
"reward_std": 0.13203652203083038,
"reward_std_all": 0.12856262549757957,
"rewards-max/forensics_tempsamp_iou_reward": 0.8901089131832123,
"rewards/forensics_tempsamp_iou_reward": 0.47811639308929443,
"step": 163
},
{
"completion_length": 19.609375,
"epoch": 0.6861924686192469,
"grad_norm": 3.648867607116699,
"kl": 0.15283203125,
"learning_rate": 8.294979079497908e-07,
"loss": 0.0061,
"reward": 0.5274137854576111,
"reward_std": 0.06567574106156826,
"reward_std_all": 0.08531443774700165,
"rewards-max/forensics_tempsamp_iou_reward": 0.7994304895401001,
"rewards/forensics_tempsamp_iou_reward": 0.3399137854576111,
"step": 164
},
{
"completion_length": 20.578125,
"epoch": 0.6903765690376569,
"grad_norm": 3.7906179428100586,
"kl": 0.1318359375,
"learning_rate": 8.284518828451882e-07,
"loss": 0.0053,
"reward": 0.7236236929893494,
"reward_std": 0.09863743931055069,
"reward_std_all": 0.10556072741746902,
"rewards-max/forensics_tempsamp_iou_reward": 0.9181312024593353,
"rewards/forensics_tempsamp_iou_reward": 0.47362369298934937,
"step": 165
},
{
"completion_length": 21.3125,
"epoch": 0.694560669456067,
"grad_norm": 4.071810245513916,
"kl": 0.15185546875,
"learning_rate": 8.274058577405857e-07,
"loss": 0.0061,
"reward": 0.5480881184339523,
"reward_std": 0.09892648085951805,
"reward_std_all": 0.09314671903848648,
"rewards-max/forensics_tempsamp_iou_reward": 0.9033875167369843,
"rewards/forensics_tempsamp_iou_reward": 0.3793381005525589,
"step": 166
},
{
"completion_length": 19.984375,
"epoch": 0.698744769874477,
"grad_norm": 4.564801216125488,
"kl": 0.19189453125,
"learning_rate": 8.263598326359832e-07,
"loss": 0.0077,
"reward": 0.7085209488868713,
"reward_std": 0.1416328325867653,
"reward_std_all": 0.15204282850027084,
"rewards-max/forensics_tempsamp_iou_reward": 0.8963311910629272,
"rewards/forensics_tempsamp_iou_reward": 0.45852090418338776,
"step": 167
},
{
"completion_length": 20.0,
"epoch": 0.702928870292887,
"grad_norm": 3.7447965145111084,
"kl": 0.18359375,
"learning_rate": 8.253138075313807e-07,
"loss": 0.0073,
"reward": 0.7695732116699219,
"reward_std": 0.07782651484012604,
"reward_std_all": 0.09474381059408188,
"rewards-max/forensics_tempsamp_iou_reward": 0.92705237865448,
"rewards/forensics_tempsamp_iou_reward": 0.5070731937885284,
"step": 168
},
{
"completion_length": 20.484375,
"epoch": 0.7071129707112971,
"grad_norm": 2.954232692718506,
"kl": 0.186767578125,
"learning_rate": 8.242677824267782e-07,
"loss": 0.0075,
"reward": 0.5750656723976135,
"reward_std": 0.091611098498106,
"reward_std_all": 0.13426023721694946,
"rewards-max/forensics_tempsamp_iou_reward": 0.9451993107795715,
"rewards/forensics_tempsamp_iou_reward": 0.36256568133831024,
"step": 169
},
{
"completion_length": 21.40625,
"epoch": 0.7112970711297071,
"grad_norm": 3.4808061122894287,
"kl": 0.1025390625,
"learning_rate": 8.232217573221757e-07,
"loss": 0.0041,
"reward": 0.7457993030548096,
"reward_std": 0.10333950631320477,
"reward_std_all": 0.12351281940937042,
"rewards-max/forensics_tempsamp_iou_reward": 0.9144946038722992,
"rewards/forensics_tempsamp_iou_reward": 0.4707992970943451,
"step": 170
},
{
"completion_length": 21.34375,
"epoch": 0.7154811715481172,
"grad_norm": 3.7498137950897217,
"kl": 0.20458984375,
"learning_rate": 8.221757322175732e-07,
"loss": 0.0082,
"reward": 0.5861375629901886,
"reward_std": 0.11713692545890808,
"reward_std_all": 0.11201807484030724,
"rewards-max/forensics_tempsamp_iou_reward": 0.7611555755138397,
"rewards/forensics_tempsamp_iou_reward": 0.35488754510879517,
"step": 171
},
{
"completion_length": 19.859375,
"epoch": 0.7196652719665272,
"grad_norm": 3.1784815788269043,
"kl": 0.17626953125,
"learning_rate": 8.211297071129707e-07,
"loss": 0.0071,
"reward": 0.5476935803890228,
"reward_std": 0.11565112322568893,
"reward_std_all": 0.10529893264174461,
"rewards-max/forensics_tempsamp_iou_reward": 0.8245013654232025,
"rewards/forensics_tempsamp_iou_reward": 0.34144356846809387,
"step": 172
},
{
"completion_length": 20.265625,
"epoch": 0.7238493723849372,
"grad_norm": 3.1133148670196533,
"kl": 0.188720703125,
"learning_rate": 8.200836820083682e-07,
"loss": 0.0075,
"reward": 0.7108959555625916,
"reward_std": 0.0980963408946991,
"reward_std_all": 0.1040746234357357,
"rewards-max/forensics_tempsamp_iou_reward": 0.91154545545578,
"rewards/forensics_tempsamp_iou_reward": 0.4483959376811981,
"step": 173
},
{
"completion_length": 20.9375,
"epoch": 0.7280334728033473,
"grad_norm": 3.376473903656006,
"kl": 0.14208984375,
"learning_rate": 8.190376569037657e-07,
"loss": 0.0057,
"reward": 0.675089955329895,
"reward_std": 0.10379363968968391,
"reward_std_all": 0.10794961452484131,
"rewards-max/forensics_tempsamp_iou_reward": 0.8427824079990387,
"rewards/forensics_tempsamp_iou_reward": 0.4063398987054825,
"step": 174
},
{
"completion_length": 21.453125,
"epoch": 0.7322175732217573,
"grad_norm": 4.696430683135986,
"kl": 0.116943359375,
"learning_rate": 8.179916317991632e-07,
"loss": 0.0047,
"reward": 0.5889217555522919,
"reward_std": 0.11626537889242172,
"reward_std_all": 0.12262865155935287,
"rewards-max/forensics_tempsamp_iou_reward": 0.8391507863998413,
"rewards/forensics_tempsamp_iou_reward": 0.3764217644929886,
"step": 175
},
{
"completion_length": 21.46875,
"epoch": 0.7364016736401674,
"grad_norm": 3.50065279006958,
"kl": 0.15185546875,
"learning_rate": 8.169456066945606e-07,
"loss": 0.0061,
"reward": 0.6088903546333313,
"reward_std": 0.12188740074634552,
"reward_std_all": 0.17425956577062607,
"rewards-max/forensics_tempsamp_iou_reward": 0.8975813090801239,
"rewards/forensics_tempsamp_iou_reward": 0.3776403218507767,
"step": 176
},
{
"completion_length": 20.03125,
"epoch": 0.7405857740585774,
"grad_norm": 3.500638961791992,
"kl": 0.14013671875,
"learning_rate": 8.158995815899581e-07,
"loss": 0.0056,
"reward": 0.6307119131088257,
"reward_std": 0.12364876642823219,
"reward_std_all": 0.1466245800256729,
"rewards-max/forensics_tempsamp_iou_reward": 0.9525042772293091,
"rewards/forensics_tempsamp_iou_reward": 0.41821183264255524,
"step": 177
},
{
"completion_length": 21.671875,
"epoch": 0.7447698744769874,
"grad_norm": 4.977306365966797,
"kl": 0.16064453125,
"learning_rate": 8.148535564853556e-07,
"loss": 0.0064,
"reward": 0.803900808095932,
"reward_std": 0.13980262726545334,
"reward_std_all": 0.14784205704927444,
"rewards-max/forensics_tempsamp_iou_reward": 0.9419832229614258,
"rewards/forensics_tempsamp_iou_reward": 0.522650808095932,
"step": 178
},
{
"completion_length": 20.53125,
"epoch": 0.7489539748953975,
"grad_norm": 5.4983930587768555,
"kl": 0.1884765625,
"learning_rate": 8.138075313807531e-07,
"loss": 0.0075,
"reward": 0.7270975708961487,
"reward_std": 0.07203448191285133,
"reward_std_all": 0.0757376030087471,
"rewards-max/forensics_tempsamp_iou_reward": 0.9036745727062225,
"rewards/forensics_tempsamp_iou_reward": 0.5020975470542908,
"step": 179
},
{
"completion_length": 23.53125,
"epoch": 0.7531380753138075,
"grad_norm": 3.955732583999634,
"kl": 0.16552734375,
"learning_rate": 8.127615062761506e-07,
"loss": 0.0066,
"reward": 0.7134921848773956,
"reward_std": 0.140335351228714,
"reward_std_all": 0.1529119722545147,
"rewards-max/forensics_tempsamp_iou_reward": 0.9676766693592072,
"rewards/forensics_tempsamp_iou_reward": 0.4759921729564667,
"step": 180
},
{
"completion_length": 22.03125,
"epoch": 0.7573221757322176,
"grad_norm": 4.426943302154541,
"kl": 0.21044921875,
"learning_rate": 8.117154811715481e-07,
"loss": 0.0084,
"reward": 0.7432798743247986,
"reward_std": 0.15820610523223877,
"reward_std_all": 0.19281521439552307,
"rewards-max/forensics_tempsamp_iou_reward": 0.8993062078952789,
"rewards/forensics_tempsamp_iou_reward": 0.48702993988990784,
"step": 181
},
{
"completion_length": 21.15625,
"epoch": 0.7615062761506276,
"grad_norm": 3.2728111743927,
"kl": 0.2568359375,
"learning_rate": 8.106694560669455e-07,
"loss": 0.0103,
"reward": 0.7888374924659729,
"reward_std": 0.1021641381084919,
"reward_std_all": 0.10562888160347939,
"rewards-max/forensics_tempsamp_iou_reward": 0.8884918689727783,
"rewards/forensics_tempsamp_iou_reward": 0.5138374269008636,
"step": 182
},
{
"completion_length": 20.3125,
"epoch": 0.7656903765690377,
"grad_norm": 3.4235565662384033,
"kl": 0.146484375,
"learning_rate": 8.096234309623431e-07,
"loss": 0.0058,
"reward": 0.702350914478302,
"reward_std": 0.08149327710270882,
"reward_std_all": 0.10334543511271477,
"rewards-max/forensics_tempsamp_iou_reward": 0.9267154037952423,
"rewards/forensics_tempsamp_iou_reward": 0.46485088765621185,
"step": 183
},
{
"completion_length": 19.78125,
"epoch": 0.7698744769874477,
"grad_norm": 3.3342247009277344,
"kl": 0.15478515625,
"learning_rate": 8.085774058577406e-07,
"loss": 0.0062,
"reward": 0.5964082181453705,
"reward_std": 0.09782892093062401,
"reward_std_all": 0.10299720615148544,
"rewards-max/forensics_tempsamp_iou_reward": 0.9563372135162354,
"rewards/forensics_tempsamp_iou_reward": 0.3901582211256027,
"step": 184
},
{
"completion_length": 21.21875,
"epoch": 0.7740585774058577,
"grad_norm": 4.920086860656738,
"kl": 0.17529296875,
"learning_rate": 8.075313807531381e-07,
"loss": 0.007,
"reward": 0.5646756589412689,
"reward_std": 0.14160168915987015,
"reward_std_all": 0.15787725895643234,
"rewards-max/forensics_tempsamp_iou_reward": 0.8811449706554413,
"rewards/forensics_tempsamp_iou_reward": 0.3771756589412689,
"step": 185
},
{
"completion_length": 20.3125,
"epoch": 0.7782426778242678,
"grad_norm": 4.2797980308532715,
"kl": 0.1630859375,
"learning_rate": 8.064853556485355e-07,
"loss": 0.0065,
"reward": 0.754219651222229,
"reward_std": 0.09982122480869293,
"reward_std_all": 0.1099339909851551,
"rewards-max/forensics_tempsamp_iou_reward": 0.9234360456466675,
"rewards/forensics_tempsamp_iou_reward": 0.5104696303606033,
"step": 186
},
{
"completion_length": 19.828125,
"epoch": 0.7824267782426778,
"grad_norm": 3.352888822555542,
"kl": 0.16064453125,
"learning_rate": 8.05439330543933e-07,
"loss": 0.0064,
"reward": 0.6668307483196259,
"reward_std": 0.08947615325450897,
"reward_std_all": 0.08575053885579109,
"rewards-max/forensics_tempsamp_iou_reward": 0.7402969598770142,
"rewards/forensics_tempsamp_iou_reward": 0.4293307363986969,
"step": 187
},
{
"completion_length": 20.21875,
"epoch": 0.7866108786610879,
"grad_norm": 4.427997589111328,
"kl": 0.17333984375,
"learning_rate": 8.043933054393305e-07,
"loss": 0.0069,
"reward": 0.7005333304405212,
"reward_std": 0.11792734265327454,
"reward_std_all": 0.1388806253671646,
"rewards-max/forensics_tempsamp_iou_reward": 0.9032189249992371,
"rewards/forensics_tempsamp_iou_reward": 0.4567832797765732,
"step": 188
},
{
"completion_length": 21.46875,
"epoch": 0.7907949790794979,
"grad_norm": 3.770003080368042,
"kl": 0.12890625,
"learning_rate": 8.03347280334728e-07,
"loss": 0.0052,
"reward": 0.5472402125597,
"reward_std": 0.08739959448575974,
"reward_std_all": 0.08724266290664673,
"rewards-max/forensics_tempsamp_iou_reward": 0.8065541684627533,
"rewards/forensics_tempsamp_iou_reward": 0.3847402036190033,
"step": 189
},
{
"completion_length": 20.421875,
"epoch": 0.7949790794979079,
"grad_norm": 3.9401254653930664,
"kl": 0.141357421875,
"learning_rate": 8.023012552301254e-07,
"loss": 0.0057,
"reward": 0.7131828665733337,
"reward_std": 0.11332910507917404,
"reward_std_all": 0.12954271957278252,
"rewards-max/forensics_tempsamp_iou_reward": 0.8893219828605652,
"rewards/forensics_tempsamp_iou_reward": 0.4881828725337982,
"step": 190
},
{
"completion_length": 20.21875,
"epoch": 0.799163179916318,
"grad_norm": 3.6429383754730225,
"kl": 0.205078125,
"learning_rate": 8.01255230125523e-07,
"loss": 0.0082,
"reward": 0.6849219501018524,
"reward_std": 0.13561861962080002,
"reward_std_all": 0.13866001740098,
"rewards-max/forensics_tempsamp_iou_reward": 0.966516762971878,
"rewards/forensics_tempsamp_iou_reward": 0.4536719471216202,
"step": 191
},
{
"completion_length": 19.5625,
"epoch": 0.803347280334728,
"grad_norm": 3.901139736175537,
"kl": 0.1201171875,
"learning_rate": 8.002092050209205e-07,
"loss": 0.0048,
"reward": 0.6551456451416016,
"reward_std": 0.1037692055106163,
"reward_std_all": 0.113624706864357,
"rewards-max/forensics_tempsamp_iou_reward": 0.9364243745803833,
"rewards/forensics_tempsamp_iou_reward": 0.46139562129974365,
"step": 192
},
{
"completion_length": 19.5625,
"epoch": 0.8075313807531381,
"grad_norm": 3.4095418453216553,
"kl": 0.164794921875,
"learning_rate": 7.99163179916318e-07,
"loss": 0.0066,
"reward": 0.5803510248661041,
"reward_std": 0.06214088201522827,
"reward_std_all": 0.07704145461320877,
"rewards-max/forensics_tempsamp_iou_reward": 0.827402800321579,
"rewards/forensics_tempsamp_iou_reward": 0.3741009682416916,
"step": 193
},
{
"completion_length": 19.28125,
"epoch": 0.8117154811715481,
"grad_norm": 3.160827875137329,
"kl": 0.15576171875,
"learning_rate": 7.981171548117155e-07,
"loss": 0.0062,
"reward": 0.7622280418872833,
"reward_std": 0.11397382244467735,
"reward_std_all": 0.12475274130702019,
"rewards-max/forensics_tempsamp_iou_reward": 0.8586688041687012,
"rewards/forensics_tempsamp_iou_reward": 0.505977988243103,
"step": 194
},
{
"completion_length": 20.671875,
"epoch": 0.8158995815899581,
"grad_norm": 4.40226936340332,
"kl": 0.15771484375,
"learning_rate": 7.97071129707113e-07,
"loss": 0.0063,
"reward": 0.5836629718542099,
"reward_std": 0.07431427389383316,
"reward_std_all": 0.08958205953240395,
"rewards-max/forensics_tempsamp_iou_reward": 0.8576148152351379,
"rewards/forensics_tempsamp_iou_reward": 0.3711629807949066,
"step": 195
},
{
"completion_length": 18.984375,
"epoch": 0.8200836820083682,
"grad_norm": 3.2660207748413086,
"kl": 0.2021484375,
"learning_rate": 7.960251046025104e-07,
"loss": 0.0081,
"reward": 0.698920875787735,
"reward_std": 0.06629294157028198,
"reward_std_all": 0.06239023804664612,
"rewards-max/forensics_tempsamp_iou_reward": 0.7531600892543793,
"rewards/forensics_tempsamp_iou_reward": 0.4551708698272705,
"step": 196
},
{
"completion_length": 19.421875,
"epoch": 0.8242677824267782,
"grad_norm": 3.5282511711120605,
"kl": 0.173583984375,
"learning_rate": 7.949790794979079e-07,
"loss": 0.0069,
"reward": 0.7174321711063385,
"reward_std": 0.08786897547543049,
"reward_std_all": 0.09843263402581215,
"rewards-max/forensics_tempsamp_iou_reward": 0.7791793942451477,
"rewards/forensics_tempsamp_iou_reward": 0.4611821472644806,
"step": 197
},
{
"completion_length": 20.234375,
"epoch": 0.8284518828451883,
"grad_norm": 3.4154021739959717,
"kl": 0.15380859375,
"learning_rate": 7.939330543933053e-07,
"loss": 0.0061,
"reward": 0.5844378769397736,
"reward_std": 0.06075112894177437,
"reward_std_all": 0.05604073032736778,
"rewards-max/forensics_tempsamp_iou_reward": 0.9718809127807617,
"rewards/forensics_tempsamp_iou_reward": 0.3781878650188446,
"step": 198
},
{
"completion_length": 20.5,
"epoch": 0.8326359832635983,
"grad_norm": 2.5856120586395264,
"kl": 0.173828125,
"learning_rate": 7.928870292887028e-07,
"loss": 0.0069,
"reward": 0.5871795117855072,
"reward_std": 0.0729019045829773,
"reward_std_all": 0.0793045274913311,
"rewards-max/forensics_tempsamp_iou_reward": 0.8402941823005676,
"rewards/forensics_tempsamp_iou_reward": 0.38717950880527496,
"step": 199
},
{
"completion_length": 19.765625,
"epoch": 0.8368200836820083,
"grad_norm": 4.4460883140563965,
"kl": 0.14794921875,
"learning_rate": 7.918410041841004e-07,
"loss": 0.0059,
"reward": 0.5985688269138336,
"reward_std": 0.08084096759557724,
"reward_std_all": 0.11753160879015923,
"rewards-max/forensics_tempsamp_iou_reward": 0.8805758953094482,
"rewards/forensics_tempsamp_iou_reward": 0.38606879115104675,
"step": 200
},
{
"completion_length": 19.40625,
"epoch": 0.8410041841004184,
"grad_norm": 4.5806427001953125,
"kl": 0.134033203125,
"learning_rate": 7.907949790794979e-07,
"loss": 0.0054,
"reward": 0.6195500493049622,
"reward_std": 0.15169352293014526,
"reward_std_all": 0.13997479528188705,
"rewards-max/forensics_tempsamp_iou_reward": 0.9506289660930634,
"rewards/forensics_tempsamp_iou_reward": 0.4383000433444977,
"step": 201
},
{
"completion_length": 19.71875,
"epoch": 0.8451882845188284,
"grad_norm": 4.264894962310791,
"kl": 0.134765625,
"learning_rate": 7.897489539748954e-07,
"loss": 0.0054,
"reward": 0.576414480805397,
"reward_std": 0.14312824606895447,
"reward_std_all": 0.14257797226309776,
"rewards-max/forensics_tempsamp_iou_reward": 0.9468026757240295,
"rewards/forensics_tempsamp_iou_reward": 0.3764144331216812,
"step": 202
},
{
"completion_length": 20.109375,
"epoch": 0.8493723849372385,
"grad_norm": 3.587698221206665,
"kl": 0.1357421875,
"learning_rate": 7.887029288702929e-07,
"loss": 0.0054,
"reward": 0.598547488451004,
"reward_std": 0.09413411468267441,
"reward_std_all": 0.09310927614569664,
"rewards-max/forensics_tempsamp_iou_reward": 0.8318417370319366,
"rewards/forensics_tempsamp_iou_reward": 0.38604749739170074,
"step": 203
},
{
"completion_length": 19.5,
"epoch": 0.8535564853556485,
"grad_norm": 4.011575698852539,
"kl": 0.2958984375,
"learning_rate": 7.876569037656904e-07,
"loss": 0.0118,
"reward": 0.7982805371284485,
"reward_std": 0.15229914337396622,
"reward_std_all": 0.15662623941898346,
"rewards-max/forensics_tempsamp_iou_reward": 0.9222330749034882,
"rewards/forensics_tempsamp_iou_reward": 0.5607805252075195,
"step": 204
},
{
"completion_length": 19.4375,
"epoch": 0.8577405857740585,
"grad_norm": 3.569589614868164,
"kl": 0.1552734375,
"learning_rate": 7.866108786610879e-07,
"loss": 0.0062,
"reward": 0.7749476432800293,
"reward_std": 0.11703173443675041,
"reward_std_all": 0.11760030314326286,
"rewards-max/forensics_tempsamp_iou_reward": 0.9126115143299103,
"rewards/forensics_tempsamp_iou_reward": 0.4936976134777069,
"step": 205
},
{
"completion_length": 19.5625,
"epoch": 0.8619246861924686,
"grad_norm": 4.900283336639404,
"kl": 0.13525390625,
"learning_rate": 7.855648535564853e-07,
"loss": 0.0054,
"reward": 0.558889850974083,
"reward_std": 0.0997723788022995,
"reward_std_all": 0.09135255962610245,
"rewards-max/forensics_tempsamp_iou_reward": 0.8452762961387634,
"rewards/forensics_tempsamp_iou_reward": 0.3526398688554764,
"step": 206
},
{
"completion_length": 20.375,
"epoch": 0.8661087866108786,
"grad_norm": 3.7589802742004395,
"kl": 0.3310546875,
"learning_rate": 7.845188284518827e-07,
"loss": 0.0132,
"reward": 0.5896714925765991,
"reward_std": 0.09271131828427315,
"reward_std_all": 0.09907573089003563,
"rewards-max/forensics_tempsamp_iou_reward": 0.9230414032936096,
"rewards/forensics_tempsamp_iou_reward": 0.4021714776754379,
"step": 207
},
{
"completion_length": 19.265625,
"epoch": 0.8702928870292888,
"grad_norm": 3.6380887031555176,
"kl": 0.153076171875,
"learning_rate": 7.834728033472802e-07,
"loss": 0.0061,
"reward": 0.6101679801940918,
"reward_std": 0.09456677734851837,
"reward_std_all": 0.0883694514632225,
"rewards-max/forensics_tempsamp_iou_reward": 0.830268532037735,
"rewards/forensics_tempsamp_iou_reward": 0.40391795337200165,
"step": 208
},
{
"completion_length": 20.40625,
"epoch": 0.8744769874476988,
"grad_norm": 3.845865249633789,
"kl": 0.138671875,
"learning_rate": 7.824267782426778e-07,
"loss": 0.0055,
"reward": 0.5966683328151703,
"reward_std": 0.10695075616240501,
"reward_std_all": 0.12642833217978477,
"rewards-max/forensics_tempsamp_iou_reward": 0.8535995185375214,
"rewards/forensics_tempsamp_iou_reward": 0.4154183268547058,
"step": 209
},
{
"completion_length": 19.375,
"epoch": 0.8786610878661087,
"grad_norm": 3.5281190872192383,
"kl": 0.14404296875,
"learning_rate": 7.813807531380753e-07,
"loss": 0.0058,
"reward": 0.7226667106151581,
"reward_std": 0.08900719322264194,
"reward_std_all": 0.09529306553304195,
"rewards-max/forensics_tempsamp_iou_reward": 0.9524787068367004,
"rewards/forensics_tempsamp_iou_reward": 0.49141670763492584,
"step": 210
},
{
"completion_length": 19.03125,
"epoch": 0.8828451882845189,
"grad_norm": 2.0846750736236572,
"kl": 0.134033203125,
"learning_rate": 7.803347280334728e-07,
"loss": 0.0054,
"reward": 0.7082947194576263,
"reward_std": 0.06829556450247765,
"reward_std_all": 0.07871583476662636,
"rewards-max/forensics_tempsamp_iou_reward": 0.9251136183738708,
"rewards/forensics_tempsamp_iou_reward": 0.47079475224018097,
"step": 211
},
{
"completion_length": 20.640625,
"epoch": 0.8870292887029289,
"grad_norm": 5.543247699737549,
"kl": 0.15673828125,
"learning_rate": 7.792887029288703e-07,
"loss": 0.0063,
"reward": 0.7889683246612549,
"reward_std": 0.11299203708767891,
"reward_std_all": 0.1339603289961815,
"rewards-max/forensics_tempsamp_iou_reward": 0.8886179029941559,
"rewards/forensics_tempsamp_iou_reward": 0.5452183485031128,
"step": 212
},
{
"completion_length": 20.40625,
"epoch": 0.891213389121339,
"grad_norm": 4.3788557052612305,
"kl": 0.17333984375,
"learning_rate": 7.782426778242678e-07,
"loss": 0.0069,
"reward": 0.6563570201396942,
"reward_std": 0.11173320934176445,
"reward_std_all": 0.11471512168645859,
"rewards-max/forensics_tempsamp_iou_reward": 0.8998058140277863,
"rewards/forensics_tempsamp_iou_reward": 0.45010703802108765,
"step": 213
},
{
"completion_length": 19.96875,
"epoch": 0.895397489539749,
"grad_norm": 3.976327419281006,
"kl": 0.16357421875,
"learning_rate": 7.771966527196653e-07,
"loss": 0.0065,
"reward": 0.7960340082645416,
"reward_std": 0.08072687312960625,
"reward_std_all": 0.07200118154287338,
"rewards-max/forensics_tempsamp_iou_reward": 0.9286450147628784,
"rewards/forensics_tempsamp_iou_reward": 0.5335339903831482,
"step": 214
},
{
"completion_length": 20.234375,
"epoch": 0.899581589958159,
"grad_norm": 7.997798442840576,
"kl": 0.38623046875,
"learning_rate": 7.761506276150628e-07,
"loss": 0.0155,
"reward": 0.5269762873649597,
"reward_std": 0.1438157930970192,
"reward_std_all": 0.1332736238837242,
"rewards-max/forensics_tempsamp_iou_reward": 0.7449223101139069,
"rewards/forensics_tempsamp_iou_reward": 0.3269762694835663,
"step": 215
},
{
"completion_length": 20.890625,
"epoch": 0.9037656903765691,
"grad_norm": 3.779489517211914,
"kl": 0.140625,
"learning_rate": 7.751046025104601e-07,
"loss": 0.0056,
"reward": 0.6292147040367126,
"reward_std": 0.13804341852664948,
"reward_std_all": 0.13636070489883423,
"rewards-max/forensics_tempsamp_iou_reward": 0.9223089814186096,
"rewards/forensics_tempsamp_iou_reward": 0.4229647219181061,
"step": 216
},
{
"completion_length": 20.703125,
"epoch": 0.9079497907949791,
"grad_norm": 3.3512744903564453,
"kl": 0.1669921875,
"learning_rate": 7.740585774058576e-07,
"loss": 0.0067,
"reward": 0.6755582094192505,
"reward_std": 0.12087374180555344,
"reward_std_all": 0.11531540006399155,
"rewards-max/forensics_tempsamp_iou_reward": 0.8294035494327545,
"rewards/forensics_tempsamp_iou_reward": 0.463058203458786,
"step": 217
},
{
"completion_length": 19.015625,
"epoch": 0.9121338912133892,
"grad_norm": 4.656987190246582,
"kl": 0.148193359375,
"learning_rate": 7.730125523012552e-07,
"loss": 0.0059,
"reward": 0.6993460059165955,
"reward_std": 0.0915173701941967,
"reward_std_all": 0.10322220996022224,
"rewards-max/forensics_tempsamp_iou_reward": 0.9744487106800079,
"rewards/forensics_tempsamp_iou_reward": 0.468095988035202,
"step": 218
},
{
"completion_length": 19.375,
"epoch": 0.9163179916317992,
"grad_norm": 2.879117488861084,
"kl": 0.142578125,
"learning_rate": 7.719665271966527e-07,
"loss": 0.0057,
"reward": 0.6920916438102722,
"reward_std": 0.08463651686906815,
"reward_std_all": 0.096174456179142,
"rewards-max/forensics_tempsamp_iou_reward": 0.9508965313434601,
"rewards/forensics_tempsamp_iou_reward": 0.44834162294864655,
"step": 219
},
{
"completion_length": 19.65625,
"epoch": 0.9205020920502092,
"grad_norm": 2.9190778732299805,
"kl": 0.18701171875,
"learning_rate": 7.709205020920502e-07,
"loss": 0.0075,
"reward": 0.769116222858429,
"reward_std": 0.1003452017903328,
"reward_std_all": 0.10294641181826591,
"rewards-max/forensics_tempsamp_iou_reward": 0.882039338350296,
"rewards/forensics_tempsamp_iou_reward": 0.5191161632537842,
"step": 220
},
{
"completion_length": 20.421875,
"epoch": 0.9246861924686193,
"grad_norm": 2.7252843379974365,
"kl": 0.130859375,
"learning_rate": 7.698744769874477e-07,
"loss": 0.0052,
"reward": 0.7196080684661865,
"reward_std": 0.14298423379659653,
"reward_std_all": 0.133404191583395,
"rewards-max/forensics_tempsamp_iou_reward": 0.8671830296516418,
"rewards/forensics_tempsamp_iou_reward": 0.4633580446243286,
"step": 221
},
{
"completion_length": 19.171875,
"epoch": 0.9288702928870293,
"grad_norm": 3.36360764503479,
"kl": 0.131591796875,
"learning_rate": 7.688284518828452e-07,
"loss": 0.0053,
"reward": 0.5286433100700378,
"reward_std": 0.07335886359214783,
"reward_std_all": 0.07470282725989819,
"rewards-max/forensics_tempsamp_iou_reward": 0.8020534217357635,
"rewards/forensics_tempsamp_iou_reward": 0.3223932534456253,
"step": 222
},
{
"completion_length": 20.3125,
"epoch": 0.9330543933054394,
"grad_norm": 3.171438455581665,
"kl": 0.17529296875,
"learning_rate": 7.677824267782427e-07,
"loss": 0.007,
"reward": 0.6529997885227203,
"reward_std": 0.15431802719831467,
"reward_std_all": 0.16333500295877457,
"rewards-max/forensics_tempsamp_iou_reward": 0.8863241374492645,
"rewards/forensics_tempsamp_iou_reward": 0.42799973487854004,
"step": 223
},
{
"completion_length": 19.375,
"epoch": 0.9372384937238494,
"grad_norm": 3.920062780380249,
"kl": 0.1328125,
"learning_rate": 7.667364016736401e-07,
"loss": 0.0053,
"reward": 0.7695104479789734,
"reward_std": 0.0596079770475626,
"reward_std_all": 0.060124386101961136,
"rewards-max/forensics_tempsamp_iou_reward": 0.8895895779132843,
"rewards/forensics_tempsamp_iou_reward": 0.5070104002952576,
"step": 224
},
{
"completion_length": 20.28125,
"epoch": 0.9414225941422594,
"grad_norm": 5.1845269203186035,
"kl": 0.5107421875,
"learning_rate": 7.656903765690376e-07,
"loss": 0.0205,
"reward": 0.708317756652832,
"reward_std": 0.23377583175897598,
"reward_std_all": 0.22285260260105133,
"rewards-max/forensics_tempsamp_iou_reward": 0.9203220903873444,
"rewards/forensics_tempsamp_iou_reward": 0.4708177149295807,
"step": 225
},
{
"completion_length": 19.984375,
"epoch": 0.9456066945606695,
"grad_norm": 4.046517848968506,
"kl": 0.119384765625,
"learning_rate": 7.64644351464435e-07,
"loss": 0.0048,
"reward": 0.6381121575832367,
"reward_std": 0.07567423582077026,
"reward_std_all": 0.11983679607510567,
"rewards-max/forensics_tempsamp_iou_reward": 0.8893300294876099,
"rewards/forensics_tempsamp_iou_reward": 0.40061214566230774,
"step": 226
},
{
"completion_length": 19.28125,
"epoch": 0.9497907949790795,
"grad_norm": 3.2434003353118896,
"kl": 0.138916015625,
"learning_rate": 7.635983263598326e-07,
"loss": 0.0056,
"reward": 0.6685264706611633,
"reward_std": 0.10820972174406052,
"reward_std_all": 0.10475587472319603,
"rewards-max/forensics_tempsamp_iou_reward": 0.9644721746444702,
"rewards/forensics_tempsamp_iou_reward": 0.45602649450302124,
"step": 227
},
{
"completion_length": 19.9375,
"epoch": 0.9539748953974896,
"grad_norm": 4.680345058441162,
"kl": 0.31103515625,
"learning_rate": 7.625523012552301e-07,
"loss": 0.0124,
"reward": 0.553364634513855,
"reward_std": 0.13703011348843575,
"reward_std_all": 0.15132080018520355,
"rewards-max/forensics_tempsamp_iou_reward": 0.8587295114994049,
"rewards/forensics_tempsamp_iou_reward": 0.3408646434545517,
"step": 228
},
{
"completion_length": 20.390625,
"epoch": 0.9581589958158996,
"grad_norm": 3.478628396987915,
"kl": 0.146484375,
"learning_rate": 7.615062761506276e-07,
"loss": 0.0059,
"reward": 0.6485503017902374,
"reward_std": 0.0816543772816658,
"reward_std_all": 0.1038474440574646,
"rewards-max/forensics_tempsamp_iou_reward": 0.852632462978363,
"rewards/forensics_tempsamp_iou_reward": 0.40480026602745056,
"step": 229
},
{
"completion_length": 19.734375,
"epoch": 0.9623430962343096,
"grad_norm": 3.6983039379119873,
"kl": 0.24169921875,
"learning_rate": 7.604602510460251e-07,
"loss": 0.0097,
"reward": 0.7362335324287415,
"reward_std": 0.1267005279660225,
"reward_std_all": 0.1380893848836422,
"rewards-max/forensics_tempsamp_iou_reward": 0.8785521984100342,
"rewards/forensics_tempsamp_iou_reward": 0.4737335294485092,
"step": 230
},
{
"completion_length": 21.015625,
"epoch": 0.9665271966527197,
"grad_norm": 3.7569174766540527,
"kl": 0.1484375,
"learning_rate": 7.594142259414226e-07,
"loss": 0.0059,
"reward": 0.6200248003005981,
"reward_std": 0.1994505673646927,
"reward_std_all": 0.18976269662380219,
"rewards-max/forensics_tempsamp_iou_reward": 0.9066827595233917,
"rewards/forensics_tempsamp_iou_reward": 0.4075247794389725,
"step": 231
},
{
"completion_length": 19.328125,
"epoch": 0.9707112970711297,
"grad_norm": 4.459153652191162,
"kl": 0.166015625,
"learning_rate": 7.5836820083682e-07,
"loss": 0.0066,
"reward": 0.6576421856880188,
"reward_std": 0.14278147369623184,
"reward_std_all": 0.1570827066898346,
"rewards-max/forensics_tempsamp_iou_reward": 0.8811975717544556,
"rewards/forensics_tempsamp_iou_reward": 0.4326421618461609,
"step": 232
},
{
"completion_length": 20.375,
"epoch": 0.9748953974895398,
"grad_norm": 5.548072338104248,
"kl": 0.32470703125,
"learning_rate": 7.573221757322175e-07,
"loss": 0.013,
"reward": 0.5401813387870789,
"reward_std": 0.08271211758255959,
"reward_std_all": 0.07565135508775711,
"rewards-max/forensics_tempsamp_iou_reward": 0.9183681011199951,
"rewards/forensics_tempsamp_iou_reward": 0.35268133878707886,
"step": 233
},
{
"completion_length": 21.59375,
"epoch": 0.9790794979079498,
"grad_norm": 4.524199962615967,
"kl": 0.112060546875,
"learning_rate": 7.56276150627615e-07,
"loss": 0.0045,
"reward": 0.5827466249465942,
"reward_std": 0.11440194398164749,
"reward_std_all": 0.11190897785127163,
"rewards-max/forensics_tempsamp_iou_reward": 0.9365072846412659,
"rewards/forensics_tempsamp_iou_reward": 0.4014965891838074,
"step": 234
},
{
"completion_length": 19.5625,
"epoch": 0.9832635983263598,
"grad_norm": 3.5351545810699463,
"kl": 0.11328125,
"learning_rate": 7.552301255230126e-07,
"loss": 0.0045,
"reward": 0.5880230665206909,
"reward_std": 0.08096093311905861,
"reward_std_all": 0.08350123092532158,
"rewards-max/forensics_tempsamp_iou_reward": 0.8926212191581726,
"rewards/forensics_tempsamp_iou_reward": 0.40052300691604614,
"step": 235
},
{
"completion_length": 20.34375,
"epoch": 0.9874476987447699,
"grad_norm": 4.107608795166016,
"kl": 0.1083984375,
"learning_rate": 7.5418410041841e-07,
"loss": 0.0043,
"reward": 0.6340355277061462,
"reward_std": 0.10700971633195877,
"reward_std_all": 0.10822470486164093,
"rewards-max/forensics_tempsamp_iou_reward": 0.9022594094276428,
"rewards/forensics_tempsamp_iou_reward": 0.3965354859828949,
"step": 236
},
{
"completion_length": 19.703125,
"epoch": 0.9916317991631799,
"grad_norm": 4.573081016540527,
"kl": 0.29296875,
"learning_rate": 7.531380753138075e-07,
"loss": 0.0117,
"reward": 0.7835548520088196,
"reward_std": 0.10273891314864159,
"reward_std_all": 0.0907827615737915,
"rewards-max/forensics_tempsamp_iou_reward": 0.9580665230751038,
"rewards/forensics_tempsamp_iou_reward": 0.5398048758506775,
"step": 237
},
{
"completion_length": 19.859375,
"epoch": 0.99581589958159,
"grad_norm": 4.4154839515686035,
"kl": 0.1923828125,
"learning_rate": 7.52092050209205e-07,
"loss": 0.0077,
"reward": 0.5382711440324783,
"reward_std": 0.1794649362564087,
"reward_std_all": 0.16889217495918274,
"rewards-max/forensics_tempsamp_iou_reward": 0.8965228796005249,
"rewards/forensics_tempsamp_iou_reward": 0.35702113807201385,
"step": 238
},
{
"completion_length": 20.142857551574707,
"epoch": 1.0,
"grad_norm": 3.2369298934936523,
"kl": 0.123779296875,
"learning_rate": 7.510460251046025e-07,
"loss": 0.0055,
"reward": 0.7365807741880417,
"reward_std": 0.1285949870944023,
"reward_std_all": 0.10958047956228256,
"rewards-max/forensics_tempsamp_iou_reward": 0.8574134409427643,
"rewards/forensics_tempsamp_iou_reward": 0.5222949981689453,
"step": 239
},
{
"completion_length": 19.0,
"epoch": 1.00418410041841,
"grad_norm": 3.2070844173431396,
"kl": 0.125732421875,
"learning_rate": 7.5e-07,
"loss": 0.005,
"reward": 0.7402958571910858,
"reward_std": 0.07445434294641018,
"reward_std_all": 0.07363547571003437,
"rewards-max/forensics_tempsamp_iou_reward": 0.9195303618907928,
"rewards/forensics_tempsamp_iou_reward": 0.5152958929538727,
"step": 240
},
{
"completion_length": 19.21875,
"epoch": 1.00836820083682,
"grad_norm": 3.4245619773864746,
"kl": 0.15869140625,
"learning_rate": 7.489539748953974e-07,
"loss": 0.0063,
"reward": 0.8178026974201202,
"reward_std": 0.10247482359409332,
"reward_std_all": 0.09839946031570435,
"rewards-max/forensics_tempsamp_iou_reward": 0.9710021018981934,
"rewards/forensics_tempsamp_iou_reward": 0.5678026974201202,
"step": 241
},
{
"completion_length": 19.84375,
"epoch": 1.0125523012552302,
"grad_norm": 5.821074485778809,
"kl": 0.122314453125,
"learning_rate": 7.479079497907949e-07,
"loss": 0.0049,
"reward": 0.7857174873352051,
"reward_std": 0.08718603104352951,
"reward_std_all": 0.09708574414253235,
"rewards-max/forensics_tempsamp_iou_reward": 0.8563569188117981,
"rewards/forensics_tempsamp_iou_reward": 0.5044675320386887,
"step": 242
},
{
"completion_length": 19.671875,
"epoch": 1.0167364016736402,
"grad_norm": 3.40218448638916,
"kl": 0.14794921875,
"learning_rate": 7.468619246861925e-07,
"loss": 0.0059,
"reward": 0.6997632682323456,
"reward_std": 0.12255222350358963,
"reward_std_all": 0.12720779329538345,
"rewards-max/forensics_tempsamp_iou_reward": 0.9110048115253448,
"rewards/forensics_tempsamp_iou_reward": 0.4872632622718811,
"step": 243
},
{
"completion_length": 19.953125,
"epoch": 1.0209205020920502,
"grad_norm": 3.603423595428467,
"kl": 0.107421875,
"learning_rate": 7.4581589958159e-07,
"loss": 0.0043,
"reward": 0.7888345420360565,
"reward_std": 0.09963475912809372,
"reward_std_all": 0.10853961855173111,
"rewards-max/forensics_tempsamp_iou_reward": 0.8581246435642242,
"rewards/forensics_tempsamp_iou_reward": 0.5263345390558243,
"step": 244
},
{
"completion_length": 19.109375,
"epoch": 1.0251046025104602,
"grad_norm": 4.744433403015137,
"kl": 0.1416015625,
"learning_rate": 7.447698744769874e-07,
"loss": 0.0057,
"reward": 0.7006538510322571,
"reward_std": 0.08508478105068207,
"reward_std_all": 0.08367323502898216,
"rewards-max/forensics_tempsamp_iou_reward": 0.9225366711616516,
"rewards/forensics_tempsamp_iou_reward": 0.4819037914276123,
"step": 245
},
{
"completion_length": 19.28125,
"epoch": 1.0292887029288702,
"grad_norm": 2.88657808303833,
"kl": 0.15087890625,
"learning_rate": 7.437238493723849e-07,
"loss": 0.006,
"reward": 0.7298277616500854,
"reward_std": 0.08168625459074974,
"reward_std_all": 0.08683764934539795,
"rewards-max/forensics_tempsamp_iou_reward": 0.9348137676715851,
"rewards/forensics_tempsamp_iou_reward": 0.467327743768692,
"step": 246
},
{
"completion_length": 19.890625,
"epoch": 1.0334728033472804,
"grad_norm": 5.582184314727783,
"kl": 0.154296875,
"learning_rate": 7.426778242677824e-07,
"loss": 0.0062,
"reward": 0.7443016767501831,
"reward_std": 0.07070762291550636,
"reward_std_all": 0.07571989670395851,
"rewards-max/forensics_tempsamp_iou_reward": 0.8957376778125763,
"rewards/forensics_tempsamp_iou_reward": 0.4630516916513443,
"step": 247
},
{
"completion_length": 19.234375,
"epoch": 1.0376569037656904,
"grad_norm": 3.3252713680267334,
"kl": 0.14404296875,
"learning_rate": 7.416317991631799e-07,
"loss": 0.0058,
"reward": 0.696300208568573,
"reward_std": 0.0668882466852665,
"reward_std_all": 0.06456394121050835,
"rewards-max/forensics_tempsamp_iou_reward": 0.9636271595954895,
"rewards/forensics_tempsamp_iou_reward": 0.4713001251220703,
"step": 248
},
{
"completion_length": 19.6875,
"epoch": 1.0418410041841004,
"grad_norm": 3.2247414588928223,
"kl": 0.12890625,
"learning_rate": 7.405857740585773e-07,
"loss": 0.0052,
"reward": 0.6140792071819305,
"reward_std": 0.08666180074214935,
"reward_std_all": 0.08961177989840508,
"rewards-max/forensics_tempsamp_iou_reward": 0.8776941001415253,
"rewards/forensics_tempsamp_iou_reward": 0.38907918334007263,
"step": 249
},
{
"completion_length": 19.484375,
"epoch": 1.0460251046025104,
"grad_norm": 4.755808353424072,
"kl": 0.1416015625,
"learning_rate": 7.395397489539748e-07,
"loss": 0.0057,
"reward": 0.8658774495124817,
"reward_std": 0.07848777249455452,
"reward_std_all": 0.08584912866353989,
"rewards-max/forensics_tempsamp_iou_reward": 0.9612065553665161,
"rewards/forensics_tempsamp_iou_reward": 0.5846273899078369,
"step": 250
},
{
"completion_length": 19.265625,
"epoch": 1.0502092050209204,
"grad_norm": 3.377537965774536,
"kl": 0.1298828125,
"learning_rate": 7.384937238493723e-07,
"loss": 0.0052,
"reward": 0.7169099450111389,
"reward_std": 0.10495249927043915,
"reward_std_all": 0.10227680206298828,
"rewards-max/forensics_tempsamp_iou_reward": 0.9632412493228912,
"rewards/forensics_tempsamp_iou_reward": 0.47940996289253235,
"step": 251
},
{
"completion_length": 19.390625,
"epoch": 1.0543933054393306,
"grad_norm": 3.3763623237609863,
"kl": 0.208984375,
"learning_rate": 7.374476987447699e-07,
"loss": 0.0083,
"reward": 0.6496513187885284,
"reward_std": 0.13427338749170303,
"reward_std_all": 0.13779878243803978,
"rewards-max/forensics_tempsamp_iou_reward": 0.9158765375614166,
"rewards/forensics_tempsamp_iou_reward": 0.4121513068675995,
"step": 252
},
{
"completion_length": 22.390625,
"epoch": 1.0585774058577406,
"grad_norm": 3.287010669708252,
"kl": 0.146484375,
"learning_rate": 7.364016736401674e-07,
"loss": 0.0059,
"reward": 0.7030723690986633,
"reward_std": 0.12338831648230553,
"reward_std_all": 0.12123875319957733,
"rewards-max/forensics_tempsamp_iou_reward": 0.8850791156291962,
"rewards/forensics_tempsamp_iou_reward": 0.4718223661184311,
"step": 253
},
{
"completion_length": 19.0625,
"epoch": 1.0627615062761506,
"grad_norm": 3.9326415061950684,
"kl": 0.15478515625,
"learning_rate": 7.353556485355649e-07,
"loss": 0.0062,
"reward": 0.7218747138977051,
"reward_std": 0.11054966598749161,
"reward_std_all": 0.10754947364330292,
"rewards-max/forensics_tempsamp_iou_reward": 0.9297492504119873,
"rewards/forensics_tempsamp_iou_reward": 0.4781247079372406,
"step": 254
},
{
"completion_length": 20.453125,
"epoch": 1.0669456066945606,
"grad_norm": 3.4414095878601074,
"kl": 0.15576171875,
"learning_rate": 7.343096234309623e-07,
"loss": 0.0062,
"reward": 0.5539441108703613,
"reward_std": 0.13167854398488998,
"reward_std_all": 0.13583699986338615,
"rewards-max/forensics_tempsamp_iou_reward": 0.9086376130580902,
"rewards/forensics_tempsamp_iou_reward": 0.3601941168308258,
"step": 255
},
{
"completion_length": 19.3125,
"epoch": 1.0711297071129706,
"grad_norm": 3.5028128623962402,
"kl": 0.19384765625,
"learning_rate": 7.332635983263598e-07,
"loss": 0.0078,
"reward": 0.6455120742321014,
"reward_std": 0.1317693367600441,
"reward_std_all": 0.1194295585155487,
"rewards-max/forensics_tempsamp_iou_reward": 0.8472772240638733,
"rewards/forensics_tempsamp_iou_reward": 0.4392620325088501,
"step": 256
},
{
"completion_length": 19.515625,
"epoch": 1.0753138075313808,
"grad_norm": 3.6268503665924072,
"kl": 0.121337890625,
"learning_rate": 7.322175732217572e-07,
"loss": 0.0049,
"reward": 0.8025315999984741,
"reward_std": 0.1315588280558586,
"reward_std_all": 0.13428010791540146,
"rewards-max/forensics_tempsamp_iou_reward": 0.9585660994052887,
"rewards/forensics_tempsamp_iou_reward": 0.5462815761566162,
"step": 257
},
{
"completion_length": 19.921875,
"epoch": 1.0794979079497908,
"grad_norm": 3.335343360900879,
"kl": 0.27001953125,
"learning_rate": 7.311715481171547e-07,
"loss": 0.0108,
"reward": 0.6105412840843201,
"reward_std": 0.09624198451638222,
"reward_std_all": 0.09496890753507614,
"rewards-max/forensics_tempsamp_iou_reward": 0.7547692656517029,
"rewards/forensics_tempsamp_iou_reward": 0.3667912632226944,
"step": 258
},
{
"completion_length": 20.140625,
"epoch": 1.0836820083682008,
"grad_norm": 3.509091854095459,
"kl": 0.16064453125,
"learning_rate": 7.301255230125522e-07,
"loss": 0.0064,
"reward": 0.5913798213005066,
"reward_std": 0.06756634637713432,
"reward_std_all": 0.0678316131234169,
"rewards-max/forensics_tempsamp_iou_reward": 0.9347826242446899,
"rewards/forensics_tempsamp_iou_reward": 0.38512980937957764,
"step": 259
},
{
"completion_length": 19.421875,
"epoch": 1.0878661087866108,
"grad_norm": 5.805680274963379,
"kl": 0.19970703125,
"learning_rate": 7.290794979079497e-07,
"loss": 0.008,
"reward": 0.7209292650222778,
"reward_std": 0.1154615730047226,
"reward_std_all": 0.11316338181495667,
"rewards-max/forensics_tempsamp_iou_reward": 0.8836654424667358,
"rewards/forensics_tempsamp_iou_reward": 0.4646792560815811,
"step": 260
},
{
"completion_length": 20.421875,
"epoch": 1.0920502092050208,
"grad_norm": 4.661555290222168,
"kl": 0.20361328125,
"learning_rate": 7.280334728033473e-07,
"loss": 0.0081,
"reward": 0.636713519692421,
"reward_std": 0.11495475843548775,
"reward_std_all": 0.10758049786090851,
"rewards-max/forensics_tempsamp_iou_reward": 0.7900475561618805,
"rewards/forensics_tempsamp_iou_reward": 0.4179634749889374,
"step": 261
},
{
"completion_length": 19.875,
"epoch": 1.096234309623431,
"grad_norm": 3.4593286514282227,
"kl": 0.119873046875,
"learning_rate": 7.269874476987448e-07,
"loss": 0.0048,
"reward": 0.6370747089385986,
"reward_std": 0.11314995586872101,
"reward_std_all": 0.10164114460349083,
"rewards-max/forensics_tempsamp_iou_reward": 0.9092923104763031,
"rewards/forensics_tempsamp_iou_reward": 0.42457470297813416,
"step": 262
},
{
"completion_length": 19.40625,
"epoch": 1.100418410041841,
"grad_norm": 2.94830322265625,
"kl": 0.1220703125,
"learning_rate": 7.259414225941423e-07,
"loss": 0.0049,
"reward": 0.6149381697177887,
"reward_std": 0.06825099885463715,
"reward_std_all": 0.07904601283371449,
"rewards-max/forensics_tempsamp_iou_reward": 0.977925032377243,
"rewards/forensics_tempsamp_iou_reward": 0.4274381399154663,
"step": 263
},
{
"completion_length": 20.015625,
"epoch": 1.104602510460251,
"grad_norm": 4.411858081817627,
"kl": 0.12158203125,
"learning_rate": 7.248953974895398e-07,
"loss": 0.0049,
"reward": 0.7880650162696838,
"reward_std": 0.1032198891043663,
"reward_std_all": 0.14656197279691696,
"rewards-max/forensics_tempsamp_iou_reward": 0.9084629416465759,
"rewards/forensics_tempsamp_iou_reward": 0.5068149864673615,
"step": 264
},
{
"completion_length": 20.8125,
"epoch": 1.108786610878661,
"grad_norm": 4.749746799468994,
"kl": 0.14453125,
"learning_rate": 7.238493723849371e-07,
"loss": 0.0058,
"reward": 0.6899360716342926,
"reward_std": 0.12369690835475922,
"reward_std_all": 0.11563741788268089,
"rewards-max/forensics_tempsamp_iou_reward": 0.922626256942749,
"rewards/forensics_tempsamp_iou_reward": 0.4711860716342926,
"step": 265
},
{
"completion_length": 20.0625,
"epoch": 1.112970711297071,
"grad_norm": 2.987792491912842,
"kl": 0.129638671875,
"learning_rate": 7.228033472803346e-07,
"loss": 0.0052,
"reward": 0.8313600718975067,
"reward_std": 0.1476704329252243,
"reward_std_all": 0.13838262110948563,
"rewards-max/forensics_tempsamp_iou_reward": 0.9640261232852936,
"rewards/forensics_tempsamp_iou_reward": 0.5876100659370422,
"step": 266
},
{
"completion_length": 19.28125,
"epoch": 1.1171548117154813,
"grad_norm": 4.441258907318115,
"kl": 0.21484375,
"learning_rate": 7.217573221757321e-07,
"loss": 0.0086,
"reward": 0.7332529127597809,
"reward_std": 0.07857915759086609,
"reward_std_all": 0.07971587404608727,
"rewards-max/forensics_tempsamp_iou_reward": 0.9090140759944916,
"rewards/forensics_tempsamp_iou_reward": 0.4832529276609421,
"step": 267
},
{
"completion_length": 19.640625,
"epoch": 1.1213389121338913,
"grad_norm": 8.39754867553711,
"kl": 0.21142578125,
"learning_rate": 7.207112970711296e-07,
"loss": 0.0085,
"reward": 0.6259461045265198,
"reward_std": 0.08134353160858154,
"reward_std_all": 0.08597713336348534,
"rewards-max/forensics_tempsamp_iou_reward": 0.9454138576984406,
"rewards/forensics_tempsamp_iou_reward": 0.40094611048698425,
"step": 268
},
{
"completion_length": 19.484375,
"epoch": 1.1255230125523012,
"grad_norm": 4.282088756561279,
"kl": 0.24755859375,
"learning_rate": 7.196652719665272e-07,
"loss": 0.0099,
"reward": 0.7280746698379517,
"reward_std": 0.1626838743686676,
"reward_std_all": 0.1483580470085144,
"rewards-max/forensics_tempsamp_iou_reward": 0.888947069644928,
"rewards/forensics_tempsamp_iou_reward": 0.49682462215423584,
"step": 269
},
{
"completion_length": 19.359375,
"epoch": 1.1297071129707112,
"grad_norm": 2.9209485054016113,
"kl": 0.17626953125,
"learning_rate": 7.186192468619247e-07,
"loss": 0.0071,
"reward": 0.6655852198600769,
"reward_std": 0.06852049939334393,
"reward_std_all": 0.10184568166732788,
"rewards-max/forensics_tempsamp_iou_reward": 0.9325712621212006,
"rewards/forensics_tempsamp_iou_reward": 0.4030851572751999,
"step": 270
},
{
"completion_length": 19.21875,
"epoch": 1.1338912133891212,
"grad_norm": 4.453962326049805,
"kl": 0.208984375,
"learning_rate": 7.175732217573222e-07,
"loss": 0.0084,
"reward": 0.7045314908027649,
"reward_std": 0.11179085448384285,
"reward_std_all": 0.10787885263562202,
"rewards-max/forensics_tempsamp_iou_reward": 0.9506552517414093,
"rewards/forensics_tempsamp_iou_reward": 0.4607814699411392,
"step": 271
},
{
"completion_length": 19.390625,
"epoch": 1.1380753138075315,
"grad_norm": 3.100008010864258,
"kl": 0.158203125,
"learning_rate": 7.165271966527197e-07,
"loss": 0.0063,
"reward": 0.7824777960777283,
"reward_std": 0.10305990651249886,
"reward_std_all": 0.11032971739768982,
"rewards-max/forensics_tempsamp_iou_reward": 0.9835217297077179,
"rewards/forensics_tempsamp_iou_reward": 0.5512278079986572,
"step": 272
},
{
"completion_length": 20.125,
"epoch": 1.1422594142259415,
"grad_norm": 3.753262758255005,
"kl": 0.16748046875,
"learning_rate": 7.154811715481172e-07,
"loss": 0.0067,
"reward": 0.7727021872997284,
"reward_std": 0.12818147614598274,
"reward_std_all": 0.11294513568282127,
"rewards-max/forensics_tempsamp_iou_reward": 0.9752940237522125,
"rewards/forensics_tempsamp_iou_reward": 0.553952157497406,
"step": 273
},
{
"completion_length": 19.625,
"epoch": 1.1464435146443515,
"grad_norm": 4.3785719871521,
"kl": 0.13525390625,
"learning_rate": 7.144351464435146e-07,
"loss": 0.0054,
"reward": 0.7639661133289337,
"reward_std": 0.13438645377755165,
"reward_std_all": 0.1317894160747528,
"rewards-max/forensics_tempsamp_iou_reward": 0.9339457452297211,
"rewards/forensics_tempsamp_iou_reward": 0.5202160775661469,
"step": 274
},
{
"completion_length": 20.3125,
"epoch": 1.1506276150627615,
"grad_norm": 4.000943183898926,
"kl": 0.15478515625,
"learning_rate": 7.13389121338912e-07,
"loss": 0.0062,
"reward": 0.5842623710632324,
"reward_std": 0.0411395076662302,
"reward_std_all": 0.05628310516476631,
"rewards-max/forensics_tempsamp_iou_reward": 0.8427961468696594,
"rewards/forensics_tempsamp_iou_reward": 0.3780123293399811,
"step": 275
},
{
"completion_length": 19.40625,
"epoch": 1.1548117154811715,
"grad_norm": 4.078647613525391,
"kl": 0.14111328125,
"learning_rate": 7.123430962343095e-07,
"loss": 0.0056,
"reward": 0.6317146122455597,
"reward_std": 0.09913697093725204,
"reward_std_all": 0.1110260933637619,
"rewards-max/forensics_tempsamp_iou_reward": 0.8654181063175201,
"rewards/forensics_tempsamp_iou_reward": 0.450464591383934,
"step": 276
},
{
"completion_length": 19.609375,
"epoch": 1.1589958158995817,
"grad_norm": 3.9737563133239746,
"kl": 0.15478515625,
"learning_rate": 7.11297071129707e-07,
"loss": 0.0062,
"reward": 0.8137139976024628,
"reward_std": 0.10499542579054832,
"reward_std_all": 0.10935528948903084,
"rewards-max/forensics_tempsamp_iou_reward": 0.8963817954063416,
"rewards/forensics_tempsamp_iou_reward": 0.5324639976024628,
"step": 277
},
{
"completion_length": 19.84375,
"epoch": 1.1631799163179917,
"grad_norm": 4.024538516998291,
"kl": 0.19580078125,
"learning_rate": 7.102510460251046e-07,
"loss": 0.0078,
"reward": 0.6521580517292023,
"reward_std": 0.06935657933354378,
"reward_std_all": 0.10336433723568916,
"rewards-max/forensics_tempsamp_iou_reward": 0.9031019806861877,
"rewards/forensics_tempsamp_iou_reward": 0.42715805768966675,
"step": 278
},
{
"completion_length": 19.640625,
"epoch": 1.1673640167364017,
"grad_norm": 3.9806020259857178,
"kl": 0.32421875,
"learning_rate": 7.092050209205021e-07,
"loss": 0.0129,
"reward": 0.8054725527763367,
"reward_std": 0.1319577470421791,
"reward_std_all": 0.13158239424228668,
"rewards-max/forensics_tempsamp_iou_reward": 0.9444242417812347,
"rewards/forensics_tempsamp_iou_reward": 0.5304725766181946,
"step": 279
},
{
"completion_length": 21.953125,
"epoch": 1.1715481171548117,
"grad_norm": 3.649444818496704,
"kl": 0.1455078125,
"learning_rate": 7.081589958158996e-07,
"loss": 0.0058,
"reward": 0.710620105266571,
"reward_std": 0.12927617877721786,
"reward_std_all": 0.15325172245502472,
"rewards-max/forensics_tempsamp_iou_reward": 0.8763737976551056,
"rewards/forensics_tempsamp_iou_reward": 0.4731200933456421,
"step": 280
},
{
"completion_length": 19.890625,
"epoch": 1.1757322175732217,
"grad_norm": 3.549935817718506,
"kl": 0.1630859375,
"learning_rate": 7.071129707112971e-07,
"loss": 0.0065,
"reward": 0.5777578949928284,
"reward_std": 0.09938506782054901,
"reward_std_all": 0.09068893641233444,
"rewards-max/forensics_tempsamp_iou_reward": 0.9009036123752594,
"rewards/forensics_tempsamp_iou_reward": 0.40900787711143494,
"step": 281
},
{
"completion_length": 20.71875,
"epoch": 1.1799163179916319,
"grad_norm": 6.097288608551025,
"kl": 0.118408203125,
"learning_rate": 7.060669456066946e-07,
"loss": 0.0047,
"reward": 0.7558645606040955,
"reward_std": 0.05522201210260391,
"reward_std_all": 0.07038208097219467,
"rewards-max/forensics_tempsamp_iou_reward": 0.9308333396911621,
"rewards/forensics_tempsamp_iou_reward": 0.5121145397424698,
"step": 282
},
{
"completion_length": 19.828125,
"epoch": 1.1841004184100419,
"grad_norm": 5.764952182769775,
"kl": 0.203125,
"learning_rate": 7.05020920502092e-07,
"loss": 0.0081,
"reward": 0.7194656133651733,
"reward_std": 0.12163973227143288,
"reward_std_all": 0.12499072402715683,
"rewards-max/forensics_tempsamp_iou_reward": 0.9525426626205444,
"rewards/forensics_tempsamp_iou_reward": 0.5007155984640121,
"step": 283
},
{
"completion_length": 19.375,
"epoch": 1.1882845188284519,
"grad_norm": 3.0479636192321777,
"kl": 0.15087890625,
"learning_rate": 7.039748953974895e-07,
"loss": 0.006,
"reward": 0.7241898477077484,
"reward_std": 0.11363746970891953,
"reward_std_all": 0.11250801384449005,
"rewards-max/forensics_tempsamp_iou_reward": 0.8149999976158142,
"rewards/forensics_tempsamp_iou_reward": 0.4679398387670517,
"step": 284
},
{
"completion_length": 19.59375,
"epoch": 1.1924686192468619,
"grad_norm": 4.745907783508301,
"kl": 0.35009765625,
"learning_rate": 7.029288702928869e-07,
"loss": 0.014,
"reward": 0.5928189158439636,
"reward_std": 0.08539650589227676,
"reward_std_all": 0.09370918571949005,
"rewards-max/forensics_tempsamp_iou_reward": 0.9238400459289551,
"rewards/forensics_tempsamp_iou_reward": 0.3990688771009445,
"step": 285
},
{
"completion_length": 19.203125,
"epoch": 1.196652719665272,
"grad_norm": 4.125890731811523,
"kl": 0.2158203125,
"learning_rate": 7.018828451882844e-07,
"loss": 0.0086,
"reward": 0.656903475522995,
"reward_std": 0.11475157737731934,
"reward_std_all": 0.12106338515877724,
"rewards-max/forensics_tempsamp_iou_reward": 0.9726780652999878,
"rewards/forensics_tempsamp_iou_reward": 0.4319034814834595,
"step": 286
},
{
"completion_length": 21.421875,
"epoch": 1.200836820083682,
"grad_norm": 4.531501770019531,
"kl": 0.14892578125,
"learning_rate": 7.00836820083682e-07,
"loss": 0.006,
"reward": 0.43299029767513275,
"reward_std": 0.06262175738811493,
"reward_std_all": 0.08430252596735954,
"rewards-max/forensics_tempsamp_iou_reward": 0.7603079676628113,
"rewards/forensics_tempsamp_iou_reward": 0.27674029767513275,
"step": 287
},
{
"completion_length": 19.4375,
"epoch": 1.205020920502092,
"grad_norm": 4.28378963470459,
"kl": 0.185546875,
"learning_rate": 6.997907949790795e-07,
"loss": 0.0074,
"reward": 0.7546142935752869,
"reward_std": 0.13969606906175613,
"reward_std_all": 0.12854139506816864,
"rewards-max/forensics_tempsamp_iou_reward": 0.893515020608902,
"rewards/forensics_tempsamp_iou_reward": 0.5046142935752869,
"step": 288
},
{
"completion_length": 19.609375,
"epoch": 1.209205020920502,
"grad_norm": 3.986689805984497,
"kl": 0.26220703125,
"learning_rate": 6.98744769874477e-07,
"loss": 0.0105,
"reward": 0.7603932023048401,
"reward_std": 0.1388854756951332,
"reward_std_all": 0.1506015807390213,
"rewards-max/forensics_tempsamp_iou_reward": 0.9368930459022522,
"rewards/forensics_tempsamp_iou_reward": 0.5166431665420532,
"step": 289
},
{
"completion_length": 22.140625,
"epoch": 1.213389121338912,
"grad_norm": 4.982595920562744,
"kl": 0.23583984375,
"learning_rate": 6.976987447698745e-07,
"loss": 0.0094,
"reward": 0.7508475482463837,
"reward_std": 0.12996838241815567,
"reward_std_all": 0.14654944464564323,
"rewards-max/forensics_tempsamp_iou_reward": 0.9251897633075714,
"rewards/forensics_tempsamp_iou_reward": 0.49459749460220337,
"step": 290
},
{
"completion_length": 19.34375,
"epoch": 1.217573221757322,
"grad_norm": 4.353562355041504,
"kl": 0.15576171875,
"learning_rate": 6.966527196652719e-07,
"loss": 0.0062,
"reward": 0.8882339298725128,
"reward_std": 0.0974634476006031,
"reward_std_all": 0.11018915101885796,
"rewards-max/forensics_tempsamp_iou_reward": 0.9436149895191193,
"rewards/forensics_tempsamp_iou_reward": 0.5882339477539062,
"step": 291
},
{
"completion_length": 19.40625,
"epoch": 1.2217573221757323,
"grad_norm": 3.851653814315796,
"kl": 0.13916015625,
"learning_rate": 6.956066945606694e-07,
"loss": 0.0056,
"reward": 0.7204906642436981,
"reward_std": 0.11298034340143204,
"reward_std_all": 0.10203060507774353,
"rewards-max/forensics_tempsamp_iou_reward": 0.8800690174102783,
"rewards/forensics_tempsamp_iou_reward": 0.4892406165599823,
"step": 292
},
{
"completion_length": 19.296875,
"epoch": 1.2259414225941423,
"grad_norm": 5.361990451812744,
"kl": 0.29736328125,
"learning_rate": 6.945606694560669e-07,
"loss": 0.0119,
"reward": 0.544624388217926,
"reward_std": 0.11196567490696907,
"reward_std_all": 0.1049218401312828,
"rewards-max/forensics_tempsamp_iou_reward": 0.902434229850769,
"rewards/forensics_tempsamp_iou_reward": 0.3821244090795517,
"step": 293
},
{
"completion_length": 19.96875,
"epoch": 1.2301255230125523,
"grad_norm": 3.7838830947875977,
"kl": 0.22802734375,
"learning_rate": 6.935146443514644e-07,
"loss": 0.0091,
"reward": 0.7454541921615601,
"reward_std": 0.12816663831472397,
"reward_std_all": 0.11791441589593887,
"rewards-max/forensics_tempsamp_iou_reward": 0.8113306164741516,
"rewards/forensics_tempsamp_iou_reward": 0.4704542011022568,
"step": 294
},
{
"completion_length": 20.828125,
"epoch": 1.2343096234309623,
"grad_norm": 4.114897727966309,
"kl": 0.20166015625,
"learning_rate": 6.924686192468618e-07,
"loss": 0.0081,
"reward": 0.6058124601840973,
"reward_std": 0.20323900133371353,
"reward_std_all": 0.20296727120876312,
"rewards-max/forensics_tempsamp_iou_reward": 0.9300575256347656,
"rewards/forensics_tempsamp_iou_reward": 0.3995624780654907,
"step": 295
},
{
"completion_length": 20.46875,
"epoch": 1.2384937238493725,
"grad_norm": 10.470465660095215,
"kl": 0.146484375,
"learning_rate": 6.914225941422594e-07,
"loss": 0.0059,
"reward": 0.7336924076080322,
"reward_std": 0.1411115899682045,
"reward_std_all": 0.14991428703069687,
"rewards-max/forensics_tempsamp_iou_reward": 0.9432699084281921,
"rewards/forensics_tempsamp_iou_reward": 0.5211923569440842,
"step": 296
},
{
"completion_length": 19.828125,
"epoch": 1.2426778242677825,
"grad_norm": 4.327788829803467,
"kl": 0.26220703125,
"learning_rate": 6.903765690376569e-07,
"loss": 0.0105,
"reward": 0.5445718765258789,
"reward_std": 0.05282280221581459,
"reward_std_all": 0.07847754284739494,
"rewards-max/forensics_tempsamp_iou_reward": 0.8865465223789215,
"rewards/forensics_tempsamp_iou_reward": 0.36332184076309204,
"step": 297
},
{
"completion_length": 19.34375,
"epoch": 1.2468619246861925,
"grad_norm": 3.5038492679595947,
"kl": 0.17333984375,
"learning_rate": 6.893305439330544e-07,
"loss": 0.0069,
"reward": 0.7130473554134369,
"reward_std": 0.06485592573881149,
"reward_std_all": 0.0636558085680008,
"rewards-max/forensics_tempsamp_iou_reward": 0.9510868489742279,
"rewards/forensics_tempsamp_iou_reward": 0.488047331571579,
"step": 298
},
{
"completion_length": 20.046875,
"epoch": 1.2510460251046025,
"grad_norm": 4.8839287757873535,
"kl": 0.1767578125,
"learning_rate": 6.882845188284518e-07,
"loss": 0.0071,
"reward": 0.6693384051322937,
"reward_std": 0.23990894854068756,
"reward_std_all": 0.2168598398566246,
"rewards-max/forensics_tempsamp_iou_reward": 0.9747258424758911,
"rewards/forensics_tempsamp_iou_reward": 0.4193383455276489,
"step": 299
},
{
"completion_length": 21.015625,
"epoch": 1.2552301255230125,
"grad_norm": 4.420351982116699,
"kl": 0.2109375,
"learning_rate": 6.872384937238493e-07,
"loss": 0.0084,
"reward": 0.5161411613225937,
"reward_std": 0.08750268071889877,
"reward_std_all": 0.10527441650629044,
"rewards-max/forensics_tempsamp_iou_reward": 0.7713386118412018,
"rewards/forensics_tempsamp_iou_reward": 0.31614120304584503,
"step": 300
},
{
"completion_length": 19.515625,
"epoch": 1.2594142259414225,
"grad_norm": 3.3127756118774414,
"kl": 0.13720703125,
"learning_rate": 6.861924686192468e-07,
"loss": 0.0055,
"reward": 0.5951897203922272,
"reward_std": 0.05882934108376503,
"reward_std_all": 0.06455271318554878,
"rewards-max/forensics_tempsamp_iou_reward": 0.9169295132160187,
"rewards/forensics_tempsamp_iou_reward": 0.42643968760967255,
"step": 301
},
{
"completion_length": 22.390625,
"epoch": 1.2635983263598327,
"grad_norm": 64.51338958740234,
"kl": 2.1416015625,
"learning_rate": 6.851464435146443e-07,
"loss": 0.0853,
"reward": 0.7670691609382629,
"reward_std": 0.07736121490597725,
"reward_std_all": 0.09182779118418694,
"rewards-max/forensics_tempsamp_iou_reward": 0.926673024892807,
"rewards/forensics_tempsamp_iou_reward": 0.5233191251754761,
"step": 302
},
{
"completion_length": 19.640625,
"epoch": 1.2677824267782427,
"grad_norm": 3.9268367290496826,
"kl": 0.2255859375,
"learning_rate": 6.841004184100418e-07,
"loss": 0.009,
"reward": 0.602776825428009,
"reward_std": 0.10657112300395966,
"reward_std_all": 0.12860267981886864,
"rewards-max/forensics_tempsamp_iou_reward": 0.8611536920070648,
"rewards/forensics_tempsamp_iou_reward": 0.39027678966522217,
"step": 303
},
{
"completion_length": 20.859375,
"epoch": 1.2719665271966527,
"grad_norm": 3.744616746902466,
"kl": 0.14013671875,
"learning_rate": 6.830543933054394e-07,
"loss": 0.0056,
"reward": 0.8892806172370911,
"reward_std": 0.10317644849419594,
"reward_std_all": 0.11018054559826851,
"rewards-max/forensics_tempsamp_iou_reward": 0.920332282781601,
"rewards/forensics_tempsamp_iou_reward": 0.6080305874347687,
"step": 304
},
{
"completion_length": 20.671875,
"epoch": 1.2761506276150627,
"grad_norm": 3.184330701828003,
"kl": 0.208740234375,
"learning_rate": 6.820083682008368e-07,
"loss": 0.0084,
"reward": 0.7188332974910736,
"reward_std": 0.06890767067670822,
"reward_std_all": 0.0988575927913189,
"rewards-max/forensics_tempsamp_iou_reward": 0.9574073851108551,
"rewards/forensics_tempsamp_iou_reward": 0.4625832885503769,
"step": 305
},
{
"completion_length": 22.796875,
"epoch": 1.280334728033473,
"grad_norm": 4.4518351554870605,
"kl": 0.1748046875,
"learning_rate": 6.809623430962343e-07,
"loss": 0.007,
"reward": 0.7945725321769714,
"reward_std": 0.09565131366252899,
"reward_std_all": 0.12561800330877304,
"rewards-max/forensics_tempsamp_iou_reward": 0.9434128701686859,
"rewards/forensics_tempsamp_iou_reward": 0.5383225381374359,
"step": 306
},
{
"completion_length": 22.375,
"epoch": 1.284518828451883,
"grad_norm": 4.991385459899902,
"kl": 0.18212890625,
"learning_rate": 6.799163179916317e-07,
"loss": 0.0073,
"reward": 0.6700207591056824,
"reward_std": 0.08363308385014534,
"reward_std_all": 0.10450924932956696,
"rewards-max/forensics_tempsamp_iou_reward": 0.8694372773170471,
"rewards/forensics_tempsamp_iou_reward": 0.40752077102661133,
"step": 307
},
{
"completion_length": 22.375,
"epoch": 1.288702928870293,
"grad_norm": 4.359987735748291,
"kl": 0.15234375,
"learning_rate": 6.788702928870292e-07,
"loss": 0.0061,
"reward": 0.7145775556564331,
"reward_std": 0.10803907364606857,
"reward_std_all": 0.11696535721421242,
"rewards-max/forensics_tempsamp_iou_reward": 0.9795078933238983,
"rewards/forensics_tempsamp_iou_reward": 0.4958275258541107,
"step": 308
},
{
"completion_length": 24.46875,
"epoch": 1.292887029288703,
"grad_norm": 4.255894660949707,
"kl": 0.203125,
"learning_rate": 6.778242677824267e-07,
"loss": 0.0081,
"reward": 0.8791455626487732,
"reward_std": 0.1460135281085968,
"reward_std_all": 0.1387924700975418,
"rewards-max/forensics_tempsamp_iou_reward": 0.9415033757686615,
"rewards/forensics_tempsamp_iou_reward": 0.6041455268859863,
"step": 309
},
{
"completion_length": 24.234375,
"epoch": 1.297071129707113,
"grad_norm": 5.635138511657715,
"kl": 0.35009765625,
"learning_rate": 6.767782426778242e-07,
"loss": 0.014,
"reward": 0.6815848648548126,
"reward_std": 0.15034694969654083,
"reward_std_all": 0.16681719571352005,
"rewards-max/forensics_tempsamp_iou_reward": 0.9179382920265198,
"rewards/forensics_tempsamp_iou_reward": 0.46908488869667053,
"step": 310
},
{
"completion_length": 23.734375,
"epoch": 1.301255230125523,
"grad_norm": 3.9815752506256104,
"kl": 0.19384765625,
"learning_rate": 6.757322175732217e-07,
"loss": 0.0078,
"reward": 0.711831733584404,
"reward_std": 0.11079496145248413,
"reward_std_all": 0.10711118206381798,
"rewards-max/forensics_tempsamp_iou_reward": 0.9034110903739929,
"rewards/forensics_tempsamp_iou_reward": 0.48058173060417175,
"step": 311
},
{
"completion_length": 25.046875,
"epoch": 1.3054393305439331,
"grad_norm": 3.7703802585601807,
"kl": 0.2060546875,
"learning_rate": 6.746861924686192e-07,
"loss": 0.0082,
"reward": 0.47628749907016754,
"reward_std": 0.11545359343290329,
"reward_std_all": 0.13748958706855774,
"rewards-max/forensics_tempsamp_iou_reward": 0.640492171049118,
"rewards/forensics_tempsamp_iou_reward": 0.2762874662876129,
"step": 312
},
{
"completion_length": 30.8125,
"epoch": 1.3096234309623431,
"grad_norm": 4.395818710327148,
"kl": 0.1923828125,
"learning_rate": 6.736401673640168e-07,
"loss": 0.0077,
"reward": 0.6138305366039276,
"reward_std": 0.09843810275197029,
"reward_std_all": 0.12396695464849472,
"rewards-max/forensics_tempsamp_iou_reward": 0.8907928466796875,
"rewards/forensics_tempsamp_iou_reward": 0.4013305604457855,
"step": 313
},
{
"completion_length": 31.09375,
"epoch": 1.3138075313807531,
"grad_norm": 4.391767501831055,
"kl": 0.24072265625,
"learning_rate": 6.725941422594143e-07,
"loss": 0.0096,
"reward": 0.9139377474784851,
"reward_std": 0.19696582853794098,
"reward_std_all": 0.19296948611736298,
"rewards-max/forensics_tempsamp_iou_reward": 0.9578733742237091,
"rewards/forensics_tempsamp_iou_reward": 0.6201877593994141,
"step": 314
},
{
"completion_length": 25.890625,
"epoch": 1.3179916317991631,
"grad_norm": 5.209534645080566,
"kl": 0.1865234375,
"learning_rate": 6.715481171548117e-07,
"loss": 0.0075,
"reward": 0.8345903754234314,
"reward_std": 0.10101427510380745,
"reward_std_all": 0.14287832379341125,
"rewards-max/forensics_tempsamp_iou_reward": 0.9116553962230682,
"rewards/forensics_tempsamp_iou_reward": 0.5595903992652893,
"step": 315
},
{
"completion_length": 28.296875,
"epoch": 1.3221757322175733,
"grad_norm": 4.31824254989624,
"kl": 0.142578125,
"learning_rate": 6.705020920502091e-07,
"loss": 0.0057,
"reward": 0.7061009705066681,
"reward_std": 0.06569057144224644,
"reward_std_all": 0.08382108062505722,
"rewards-max/forensics_tempsamp_iou_reward": 0.8945572078227997,
"rewards/forensics_tempsamp_iou_reward": 0.48110097646713257,
"step": 316
},
{
"completion_length": 28.984375,
"epoch": 1.3263598326359833,
"grad_norm": 4.1526875495910645,
"kl": 0.1728515625,
"learning_rate": 6.694560669456066e-07,
"loss": 0.0069,
"reward": 0.7623976171016693,
"reward_std": 0.12360857799649239,
"reward_std_all": 0.10899084061384201,
"rewards-max/forensics_tempsamp_iou_reward": 0.8795387148857117,
"rewards/forensics_tempsamp_iou_reward": 0.49364760518074036,
"step": 317
},
{
"completion_length": 27.890625,
"epoch": 1.3305439330543933,
"grad_norm": 3.183825731277466,
"kl": 0.14208984375,
"learning_rate": 6.684100418410041e-07,
"loss": 0.0057,
"reward": 0.69048210978508,
"reward_std": 0.08396032825112343,
"reward_std_all": 0.10438704863190651,
"rewards-max/forensics_tempsamp_iou_reward": 0.9490048885345459,
"rewards/forensics_tempsamp_iou_reward": 0.5029820501804352,
"step": 318
},
{
"completion_length": 25.734375,
"epoch": 1.3347280334728033,
"grad_norm": 4.414980888366699,
"kl": 0.2509765625,
"learning_rate": 6.673640167364016e-07,
"loss": 0.01,
"reward": 0.6625126004219055,
"reward_std": 0.10448610782623291,
"reward_std_all": 0.11216845363378525,
"rewards-max/forensics_tempsamp_iou_reward": 0.9166888892650604,
"rewards/forensics_tempsamp_iou_reward": 0.4312625825405121,
"step": 319
},
{
"completion_length": 24.21875,
"epoch": 1.3389121338912133,
"grad_norm": 5.139053821563721,
"kl": 0.2060546875,
"learning_rate": 6.663179916317991e-07,
"loss": 0.0082,
"reward": 0.6253743469715118,
"reward_std": 0.11461586505174637,
"reward_std_all": 0.1214773878455162,
"rewards-max/forensics_tempsamp_iou_reward": 0.8780259490013123,
"rewards/forensics_tempsamp_iou_reward": 0.40037432312965393,
"step": 320
},
{
"completion_length": 21.5625,
"epoch": 1.3430962343096233,
"grad_norm": 6.983429908752441,
"kl": 0.3408203125,
"learning_rate": 6.652719665271967e-07,
"loss": 0.0136,
"reward": 0.7228768169879913,
"reward_std": 0.10217709839344025,
"reward_std_all": 0.11542605236172676,
"rewards-max/forensics_tempsamp_iou_reward": 0.926908552646637,
"rewards/forensics_tempsamp_iou_reward": 0.47912682592868805,
"step": 321
},
{
"completion_length": 22.953125,
"epoch": 1.3472803347280335,
"grad_norm": 4.167449951171875,
"kl": 0.1865234375,
"learning_rate": 6.642259414225942e-07,
"loss": 0.0075,
"reward": 0.7321462631225586,
"reward_std": 0.07345201820135117,
"reward_std_all": 0.09989822283387184,
"rewards-max/forensics_tempsamp_iou_reward": 0.9465283453464508,
"rewards/forensics_tempsamp_iou_reward": 0.5071462392807007,
"step": 322
},
{
"completion_length": 24.453125,
"epoch": 1.3514644351464435,
"grad_norm": 4.583470821380615,
"kl": 0.2099609375,
"learning_rate": 6.631799163179917e-07,
"loss": 0.0084,
"reward": 0.6327186822891235,
"reward_std": 0.09131934493780136,
"reward_std_all": 0.12098771333694458,
"rewards-max/forensics_tempsamp_iou_reward": 0.8574323356151581,
"rewards/forensics_tempsamp_iou_reward": 0.4327186644077301,
"step": 323
},
{
"completion_length": 20.828125,
"epoch": 1.3556485355648535,
"grad_norm": 4.095149040222168,
"kl": 0.17236328125,
"learning_rate": 6.621338912133892e-07,
"loss": 0.0069,
"reward": 0.8503212928771973,
"reward_std": 0.09703220427036285,
"reward_std_all": 0.09956958517432213,
"rewards-max/forensics_tempsamp_iou_reward": 0.9068489968776703,
"rewards/forensics_tempsamp_iou_reward": 0.5815712660551071,
"step": 324
},
{
"completion_length": 19.765625,
"epoch": 1.3598326359832635,
"grad_norm": 4.892597675323486,
"kl": 0.16943359375,
"learning_rate": 6.610878661087865e-07,
"loss": 0.0068,
"reward": 0.6752539277076721,
"reward_std": 0.08493788167834282,
"reward_std_all": 0.09230585396289825,
"rewards-max/forensics_tempsamp_iou_reward": 0.9050355851650238,
"rewards/forensics_tempsamp_iou_reward": 0.4752538800239563,
"step": 325
},
{
"completion_length": 21.296875,
"epoch": 1.3640167364016738,
"grad_norm": 3.5764424800872803,
"kl": 0.13818359375,
"learning_rate": 6.60041841004184e-07,
"loss": 0.0055,
"reward": 0.727758377790451,
"reward_std": 0.07567137479782104,
"reward_std_all": 0.08000505343079567,
"rewards-max/forensics_tempsamp_iou_reward": 0.9355675578117371,
"rewards/forensics_tempsamp_iou_reward": 0.5027583837509155,
"step": 326
},
{
"completion_length": 21.703125,
"epoch": 1.3682008368200838,
"grad_norm": 4.163809299468994,
"kl": 0.2412109375,
"learning_rate": 6.589958158995815e-07,
"loss": 0.0096,
"reward": 0.5220455825328827,
"reward_std": 0.09583726897835732,
"reward_std_all": 0.09698926657438278,
"rewards-max/forensics_tempsamp_iou_reward": 0.9336329698562622,
"rewards/forensics_tempsamp_iou_reward": 0.31579558551311493,
"step": 327
},
{
"completion_length": 20.25,
"epoch": 1.3723849372384938,
"grad_norm": 3.116504192352295,
"kl": 0.16455078125,
"learning_rate": 6.57949790794979e-07,
"loss": 0.0066,
"reward": 0.7794105410575867,
"reward_std": 0.05285605788230896,
"reward_std_all": 0.07386208698153496,
"rewards-max/forensics_tempsamp_iou_reward": 0.8948909044265747,
"rewards/forensics_tempsamp_iou_reward": 0.49816054105758667,
"step": 328
},
{
"completion_length": 22.875,
"epoch": 1.3765690376569037,
"grad_norm": 3.9405908584594727,
"kl": 0.18603515625,
"learning_rate": 6.569037656903765e-07,
"loss": 0.0074,
"reward": 0.7433749735355377,
"reward_std": 0.1326543390750885,
"reward_std_all": 0.12087294086813927,
"rewards-max/forensics_tempsamp_iou_reward": 0.9267794191837311,
"rewards/forensics_tempsamp_iou_reward": 0.4871249496936798,
"step": 329
},
{
"completion_length": 20.90625,
"epoch": 1.3807531380753137,
"grad_norm": 3.641704559326172,
"kl": 0.23388671875,
"learning_rate": 6.558577405857741e-07,
"loss": 0.0094,
"reward": 0.8207233846187592,
"reward_std": 0.09848406538367271,
"reward_std_all": 0.10303851217031479,
"rewards-max/forensics_tempsamp_iou_reward": 0.9869769811630249,
"rewards/forensics_tempsamp_iou_reward": 0.5582233667373657,
"step": 330
},
{
"completion_length": 20.484375,
"epoch": 1.3849372384937237,
"grad_norm": 3.9989140033721924,
"kl": 0.19580078125,
"learning_rate": 6.548117154811716e-07,
"loss": 0.0078,
"reward": 0.7530893981456757,
"reward_std": 0.07827362790703773,
"reward_std_all": 0.07362083345651627,
"rewards-max/forensics_tempsamp_iou_reward": 0.9151604771614075,
"rewards/forensics_tempsamp_iou_reward": 0.5030893981456757,
"step": 331
},
{
"completion_length": 19.890625,
"epoch": 1.389121338912134,
"grad_norm": 3.991621971130371,
"kl": 0.131591796875,
"learning_rate": 6.537656903765691e-07,
"loss": 0.0053,
"reward": 0.6878316700458527,
"reward_std": 0.10280732810497284,
"reward_std_all": 0.10824109241366386,
"rewards-max/forensics_tempsamp_iou_reward": 0.9193782210350037,
"rewards/forensics_tempsamp_iou_reward": 0.4690816253423691,
"step": 332
},
{
"completion_length": 21.21875,
"epoch": 1.393305439330544,
"grad_norm": 3.379209041595459,
"kl": 0.14208984375,
"learning_rate": 6.527196652719665e-07,
"loss": 0.0057,
"reward": 0.7441627383232117,
"reward_std": 0.10539986193180084,
"reward_std_all": 0.09726296737790108,
"rewards-max/forensics_tempsamp_iou_reward": 0.9142676293849945,
"rewards/forensics_tempsamp_iou_reward": 0.500412717461586,
"step": 333
},
{
"completion_length": 19.1875,
"epoch": 1.397489539748954,
"grad_norm": 3.920252561569214,
"kl": 0.21044921875,
"learning_rate": 6.51673640167364e-07,
"loss": 0.0084,
"reward": 0.6706031560897827,
"reward_std": 0.13978128135204315,
"reward_std_all": 0.16774099320173264,
"rewards-max/forensics_tempsamp_iou_reward": 0.9531669914722443,
"rewards/forensics_tempsamp_iou_reward": 0.433103084564209,
"step": 334
},
{
"completion_length": 20.09375,
"epoch": 1.401673640167364,
"grad_norm": 4.283573150634766,
"kl": 0.16796875,
"learning_rate": 6.506276150627614e-07,
"loss": 0.0067,
"reward": 0.6819342374801636,
"reward_std": 0.07598416320979595,
"reward_std_all": 0.07031166180968285,
"rewards-max/forensics_tempsamp_iou_reward": 0.9737009108066559,
"rewards/forensics_tempsamp_iou_reward": 0.45693422853946686,
"step": 335
},
{
"completion_length": 20.375,
"epoch": 1.4058577405857742,
"grad_norm": 4.301535129547119,
"kl": 0.166015625,
"learning_rate": 6.495815899581589e-07,
"loss": 0.0067,
"reward": 0.606118232011795,
"reward_std": 0.08488035947084427,
"reward_std_all": 0.08354498445987701,
"rewards-max/forensics_tempsamp_iou_reward": 0.8169165849685669,
"rewards/forensics_tempsamp_iou_reward": 0.4123682379722595,
"step": 336
},
{
"completion_length": 20.5625,
"epoch": 1.4100418410041842,
"grad_norm": 4.678075790405273,
"kl": 0.1865234375,
"learning_rate": 6.485355648535564e-07,
"loss": 0.0075,
"reward": 0.6583587825298309,
"reward_std": 0.08498156815767288,
"reward_std_all": 0.08553614281117916,
"rewards-max/forensics_tempsamp_iou_reward": 0.8164297044277191,
"rewards/forensics_tempsamp_iou_reward": 0.4271087795495987,
"step": 337
},
{
"completion_length": 19.078125,
"epoch": 1.4142259414225942,
"grad_norm": 5.301628589630127,
"kl": 0.1552734375,
"learning_rate": 6.47489539748954e-07,
"loss": 0.0062,
"reward": 0.5501285195350647,
"reward_std": 0.054558608680963516,
"reward_std_all": 0.07382160797715187,
"rewards-max/forensics_tempsamp_iou_reward": 0.9098122715950012,
"rewards/forensics_tempsamp_iou_reward": 0.3626285046339035,
"step": 338
},
{
"completion_length": 19.515625,
"epoch": 1.4184100418410042,
"grad_norm": 4.072910785675049,
"kl": 0.15380859375,
"learning_rate": 6.464435146443515e-07,
"loss": 0.0062,
"reward": 0.7734572887420654,
"reward_std": 0.07339539751410484,
"reward_std_all": 0.10351938009262085,
"rewards-max/forensics_tempsamp_iou_reward": 0.9348401725292206,
"rewards/forensics_tempsamp_iou_reward": 0.5109573304653168,
"step": 339
},
{
"completion_length": 19.8125,
"epoch": 1.4225941422594142,
"grad_norm": 5.296221733093262,
"kl": 0.17431640625,
"learning_rate": 6.45397489539749e-07,
"loss": 0.007,
"reward": 0.7334912121295929,
"reward_std": 0.05972724035382271,
"reward_std_all": 0.08514976315200329,
"rewards-max/forensics_tempsamp_iou_reward": 0.8932410180568695,
"rewards/forensics_tempsamp_iou_reward": 0.49599121510982513,
"step": 340
},
{
"completion_length": 19.3125,
"epoch": 1.4267782426778242,
"grad_norm": 4.615549087524414,
"kl": 0.1845703125,
"learning_rate": 6.443514644351464e-07,
"loss": 0.0074,
"reward": 0.7595143020153046,
"reward_std": 0.06654370203614235,
"reward_std_all": 0.07354473695158958,
"rewards-max/forensics_tempsamp_iou_reward": 0.9415877461433411,
"rewards/forensics_tempsamp_iou_reward": 0.5032642781734467,
"step": 341
},
{
"completion_length": 21.015625,
"epoch": 1.4309623430962344,
"grad_norm": 4.345317840576172,
"kl": 0.23388671875,
"learning_rate": 6.433054393305439e-07,
"loss": 0.0094,
"reward": 0.752809077501297,
"reward_std": 0.14859123528003693,
"reward_std_all": 0.14787033200263977,
"rewards-max/forensics_tempsamp_iou_reward": 0.8886229693889618,
"rewards/forensics_tempsamp_iou_reward": 0.4903090298175812,
"step": 342
},
{
"completion_length": 18.71875,
"epoch": 1.4351464435146444,
"grad_norm": 5.5943403244018555,
"kl": 0.2255859375,
"learning_rate": 6.422594142259414e-07,
"loss": 0.009,
"reward": 0.8426127135753632,
"reward_std": 0.07154106348752975,
"reward_std_all": 0.07814988121390343,
"rewards-max/forensics_tempsamp_iou_reward": 0.8504079878330231,
"rewards/forensics_tempsamp_iou_reward": 0.5426127016544342,
"step": 343
},
{
"completion_length": 19.265625,
"epoch": 1.4393305439330544,
"grad_norm": 6.4163641929626465,
"kl": 0.18603515625,
"learning_rate": 6.412133891213389e-07,
"loss": 0.0074,
"reward": 0.6658518314361572,
"reward_std": 0.09378122538328171,
"reward_std_all": 0.0948740690946579,
"rewards-max/forensics_tempsamp_iou_reward": 0.9052205979824066,
"rewards/forensics_tempsamp_iou_reward": 0.42835181951522827,
"step": 344
},
{
"completion_length": 19.140625,
"epoch": 1.4435146443514644,
"grad_norm": 6.023801326751709,
"kl": 0.16455078125,
"learning_rate": 6.401673640167363e-07,
"loss": 0.0066,
"reward": 0.7680287659168243,
"reward_std": 0.07642362639307976,
"reward_std_all": 0.1125335544347763,
"rewards-max/forensics_tempsamp_iou_reward": 0.9586975872516632,
"rewards/forensics_tempsamp_iou_reward": 0.5242787599563599,
"step": 345
},
{
"completion_length": 19.171875,
"epoch": 1.4476987447698746,
"grad_norm": 3.776806592941284,
"kl": 0.2685546875,
"learning_rate": 6.391213389121338e-07,
"loss": 0.0107,
"reward": 0.668417364358902,
"reward_std": 0.05609360430389643,
"reward_std_all": 0.06052548810839653,
"rewards-max/forensics_tempsamp_iou_reward": 0.9059432744979858,
"rewards/forensics_tempsamp_iou_reward": 0.4246673583984375,
"step": 346
},
{
"completion_length": 19.359375,
"epoch": 1.4518828451882846,
"grad_norm": 4.42294979095459,
"kl": 0.14599609375,
"learning_rate": 6.380753138075314e-07,
"loss": 0.0058,
"reward": 0.6186015009880066,
"reward_std": 0.06285391189157963,
"reward_std_all": 0.060977621003985405,
"rewards-max/forensics_tempsamp_iou_reward": 0.8870429396629333,
"rewards/forensics_tempsamp_iou_reward": 0.41235144436359406,
"step": 347
},
{
"completion_length": 19.9375,
"epoch": 1.4560669456066946,
"grad_norm": 6.964395046234131,
"kl": 0.23974609375,
"learning_rate": 6.370292887029289e-07,
"loss": 0.0096,
"reward": 0.786225438117981,
"reward_std": 0.11651625484228134,
"reward_std_all": 0.11519875377416611,
"rewards-max/forensics_tempsamp_iou_reward": 0.941349595785141,
"rewards/forensics_tempsamp_iou_reward": 0.5237254500389099,
"step": 348
},
{
"completion_length": 19.96875,
"epoch": 1.4602510460251046,
"grad_norm": 5.059910297393799,
"kl": 0.17041015625,
"learning_rate": 6.359832635983263e-07,
"loss": 0.0068,
"reward": 0.7098332047462463,
"reward_std": 0.03770196624100208,
"reward_std_all": 0.07342774793505669,
"rewards-max/forensics_tempsamp_iou_reward": 0.9466278254985809,
"rewards/forensics_tempsamp_iou_reward": 0.503583163022995,
"step": 349
},
{
"completion_length": 19.421875,
"epoch": 1.4644351464435146,
"grad_norm": 4.280508518218994,
"kl": 0.28515625,
"learning_rate": 6.349372384937238e-07,
"loss": 0.0114,
"reward": 0.8710715174674988,
"reward_std": 0.12632274627685547,
"reward_std_all": 0.12493560463190079,
"rewards-max/forensics_tempsamp_iou_reward": 0.9256309866905212,
"rewards/forensics_tempsamp_iou_reward": 0.6148215234279633,
"step": 350
},
{
"completion_length": 20.4375,
"epoch": 1.4686192468619246,
"grad_norm": 6.721954345703125,
"kl": 0.1982421875,
"learning_rate": 6.338912133891213e-07,
"loss": 0.0079,
"reward": 0.8160971403121948,
"reward_std": 0.11630572006106377,
"reward_std_all": 0.11770603060722351,
"rewards-max/forensics_tempsamp_iou_reward": 0.9495206475257874,
"rewards/forensics_tempsamp_iou_reward": 0.5598471462726593,
"step": 351
},
{
"completion_length": 19.390625,
"epoch": 1.4728033472803348,
"grad_norm": 3.9730606079101562,
"kl": 0.21435546875,
"learning_rate": 6.328451882845188e-07,
"loss": 0.0086,
"reward": 0.5548087656497955,
"reward_std": 0.19363372027873993,
"reward_std_all": 0.17897489666938782,
"rewards-max/forensics_tempsamp_iou_reward": 0.8675292730331421,
"rewards/forensics_tempsamp_iou_reward": 0.3798087537288666,
"step": 352
},
{
"completion_length": 19.484375,
"epoch": 1.4769874476987448,
"grad_norm": 4.822771072387695,
"kl": 0.18701171875,
"learning_rate": 6.317991631799163e-07,
"loss": 0.0075,
"reward": 0.6104840338230133,
"reward_std": 0.1203470528125763,
"reward_std_all": 0.1089002788066864,
"rewards-max/forensics_tempsamp_iou_reward": 0.9748005270957947,
"rewards/forensics_tempsamp_iou_reward": 0.39798402786254883,
"step": 353
},
{
"completion_length": 19.171875,
"epoch": 1.4811715481171548,
"grad_norm": 4.341469764709473,
"kl": 0.2080078125,
"learning_rate": 6.307531380753138e-07,
"loss": 0.0083,
"reward": 0.689783364534378,
"reward_std": 0.07247291132807732,
"reward_std_all": 0.09163950383663177,
"rewards-max/forensics_tempsamp_iou_reward": 0.897395670413971,
"rewards/forensics_tempsamp_iou_reward": 0.46478332579135895,
"step": 354
},
{
"completion_length": 19.875,
"epoch": 1.4853556485355648,
"grad_norm": 3.9398820400238037,
"kl": 0.15966796875,
"learning_rate": 6.297071129707112e-07,
"loss": 0.0064,
"reward": 0.8064073622226715,
"reward_std": 0.05604461207985878,
"reward_std_all": 0.0642822366207838,
"rewards-max/forensics_tempsamp_iou_reward": 0.9546734094619751,
"rewards/forensics_tempsamp_iou_reward": 0.5314073860645294,
"step": 355
},
{
"completion_length": 19.4375,
"epoch": 1.489539748953975,
"grad_norm": 5.106343746185303,
"kl": 0.20751953125,
"learning_rate": 6.286610878661088e-07,
"loss": 0.0083,
"reward": 0.7412002384662628,
"reward_std": 0.12852585315704346,
"reward_std_all": 0.12553340196609497,
"rewards-max/forensics_tempsamp_iou_reward": 0.934188038110733,
"rewards/forensics_tempsamp_iou_reward": 0.49745023250579834,
"step": 356
},
{
"completion_length": 19.828125,
"epoch": 1.493723849372385,
"grad_norm": 3.1529664993286133,
"kl": 0.15869140625,
"learning_rate": 6.276150627615063e-07,
"loss": 0.0063,
"reward": 0.687052458524704,
"reward_std": 0.05527842789888382,
"reward_std_all": 0.05929508060216904,
"rewards-max/forensics_tempsamp_iou_reward": 0.8835247755050659,
"rewards/forensics_tempsamp_iou_reward": 0.45580244064331055,
"step": 357
},
{
"completion_length": 20.015625,
"epoch": 1.497907949790795,
"grad_norm": 3.901362180709839,
"kl": 0.140625,
"learning_rate": 6.265690376569037e-07,
"loss": 0.0056,
"reward": 0.8502953052520752,
"reward_std": 0.07031768560409546,
"reward_std_all": 0.07727281376719475,
"rewards-max/forensics_tempsamp_iou_reward": 0.9457861185073853,
"rewards/forensics_tempsamp_iou_reward": 0.6127952635288239,
"step": 358
},
{
"completion_length": 20.109375,
"epoch": 1.502092050209205,
"grad_norm": 12.702938079833984,
"kl": 0.1591796875,
"learning_rate": 6.255230125523012e-07,
"loss": 0.0064,
"reward": 0.7612243592739105,
"reward_std": 0.0801387894898653,
"reward_std_all": 0.07612470537424088,
"rewards-max/forensics_tempsamp_iou_reward": 0.942497193813324,
"rewards/forensics_tempsamp_iou_reward": 0.4987243562936783,
"step": 359
},
{
"completion_length": 19.171875,
"epoch": 1.506276150627615,
"grad_norm": 4.231260299682617,
"kl": 0.3447265625,
"learning_rate": 6.244769874476987e-07,
"loss": 0.0138,
"reward": 0.7766593992710114,
"reward_std": 0.08531934022903442,
"reward_std_all": 0.08036176487803459,
"rewards-max/forensics_tempsamp_iou_reward": 0.8920369148254395,
"rewards/forensics_tempsamp_iou_reward": 0.5141594111919403,
"step": 360
},
{
"completion_length": 19.859375,
"epoch": 1.510460251046025,
"grad_norm": 14.111891746520996,
"kl": 0.2373046875,
"learning_rate": 6.234309623430962e-07,
"loss": 0.0095,
"reward": 0.6942225992679596,
"reward_std": 0.0476329792290926,
"reward_std_all": 0.0674137994647026,
"rewards-max/forensics_tempsamp_iou_reward": 0.9291628301143646,
"rewards/forensics_tempsamp_iou_reward": 0.487972617149353,
"step": 361
},
{
"completion_length": 19.109375,
"epoch": 1.514644351464435,
"grad_norm": 4.603570938110352,
"kl": 0.19921875,
"learning_rate": 6.223849372384937e-07,
"loss": 0.008,
"reward": 0.7590403854846954,
"reward_std": 0.11631745100021362,
"reward_std_all": 0.1163734681904316,
"rewards-max/forensics_tempsamp_iou_reward": 0.8415163159370422,
"rewards/forensics_tempsamp_iou_reward": 0.5340403467416763,
"step": 362
},
{
"completion_length": 20.25,
"epoch": 1.5188284518828452,
"grad_norm": 4.827236175537109,
"kl": 0.1142578125,
"learning_rate": 6.213389121338912e-07,
"loss": 0.0046,
"reward": 0.6580938398838043,
"reward_std": 0.09933791682124138,
"reward_std_all": 0.11805080249905586,
"rewards-max/forensics_tempsamp_iou_reward": 0.9096912443637848,
"rewards/forensics_tempsamp_iou_reward": 0.4580938220024109,
"step": 363
},
{
"completion_length": 20.1875,
"epoch": 1.5230125523012552,
"grad_norm": 7.909019470214844,
"kl": 0.166259765625,
"learning_rate": 6.202928870292888e-07,
"loss": 0.0066,
"reward": 0.6920763850212097,
"reward_std": 0.05603438429534435,
"reward_std_all": 0.06567902863025665,
"rewards-max/forensics_tempsamp_iou_reward": 0.9176453649997711,
"rewards/forensics_tempsamp_iou_reward": 0.4858263432979584,
"step": 364
},
{
"completion_length": 19.53125,
"epoch": 1.5271966527196654,
"grad_norm": 4.437346935272217,
"kl": 0.2802734375,
"learning_rate": 6.192468619246862e-07,
"loss": 0.0112,
"reward": 0.7083640098571777,
"reward_std": 0.10862875357270241,
"reward_std_all": 0.110087089240551,
"rewards-max/forensics_tempsamp_iou_reward": 0.8934192955493927,
"rewards/forensics_tempsamp_iou_reward": 0.45836398005485535,
"step": 365
},
{
"completion_length": 19.71875,
"epoch": 1.5313807531380754,
"grad_norm": 4.136433124542236,
"kl": 0.16845703125,
"learning_rate": 6.182008368200836e-07,
"loss": 0.0067,
"reward": 0.4000803828239441,
"reward_std": 0.07327224686741829,
"reward_std_all": 0.08795398846268654,
"rewards-max/forensics_tempsamp_iou_reward": 0.722610741853714,
"rewards/forensics_tempsamp_iou_reward": 0.281330406665802,
"step": 366
},
{
"completion_length": 20.015625,
"epoch": 1.5355648535564854,
"grad_norm": 4.324184894561768,
"kl": 0.18896484375,
"learning_rate": 6.171548117154811e-07,
"loss": 0.0076,
"reward": 0.7813405394554138,
"reward_std": 0.09480735287070274,
"reward_std_all": 0.0989840105175972,
"rewards-max/forensics_tempsamp_iou_reward": 0.9382022619247437,
"rewards/forensics_tempsamp_iou_reward": 0.5563405603170395,
"step": 367
},
{
"completion_length": 19.203125,
"epoch": 1.5397489539748954,
"grad_norm": 4.675383567810059,
"kl": 0.138427734375,
"learning_rate": 6.161087866108786e-07,
"loss": 0.0055,
"reward": 0.9581067562103271,
"reward_std": 0.09578951820731163,
"reward_std_all": 0.09966465458273888,
"rewards-max/forensics_tempsamp_iou_reward": 0.9872419834136963,
"rewards/forensics_tempsamp_iou_reward": 0.6831067204475403,
"step": 368
},
{
"completion_length": 19.359375,
"epoch": 1.5439330543933054,
"grad_norm": 5.557816505432129,
"kl": 0.1748046875,
"learning_rate": 6.150627615062761e-07,
"loss": 0.007,
"reward": 0.6533156335353851,
"reward_std": 0.09746003150939941,
"reward_std_all": 0.09253546968102455,
"rewards-max/forensics_tempsamp_iou_reward": 0.9695588648319244,
"rewards/forensics_tempsamp_iou_reward": 0.44081561267375946,
"step": 369
},
{
"completion_length": 19.828125,
"epoch": 1.5481171548117154,
"grad_norm": 4.149466037750244,
"kl": 0.1494140625,
"learning_rate": 6.140167364016736e-07,
"loss": 0.006,
"reward": 0.8242698907852173,
"reward_std": 0.10261883586645126,
"reward_std_all": 0.10632168501615524,
"rewards-max/forensics_tempsamp_iou_reward": 0.9604272246360779,
"rewards/forensics_tempsamp_iou_reward": 0.5867698788642883,
"step": 370
},
{
"completion_length": 20.09375,
"epoch": 1.5523012552301254,
"grad_norm": 5.221569061279297,
"kl": 0.23193359375,
"learning_rate": 6.129707112970711e-07,
"loss": 0.0093,
"reward": 0.5343354642391205,
"reward_std": 0.05670253373682499,
"reward_std_all": 0.06742007285356522,
"rewards-max/forensics_tempsamp_iou_reward": 0.7072303891181946,
"rewards/forensics_tempsamp_iou_reward": 0.290585458278656,
"step": 371
},
{
"completion_length": 20.65625,
"epoch": 1.5564853556485354,
"grad_norm": 3.3939411640167236,
"kl": 0.22900390625,
"learning_rate": 6.119246861924686e-07,
"loss": 0.0092,
"reward": 0.6064125299453735,
"reward_std": 0.07433366402983665,
"reward_std_all": 0.07769619673490524,
"rewards-max/forensics_tempsamp_iou_reward": 0.919095367193222,
"rewards/forensics_tempsamp_iou_reward": 0.4376625269651413,
"step": 372
},
{
"completion_length": 20.234375,
"epoch": 1.5606694560669456,
"grad_norm": 4.620502471923828,
"kl": 0.181640625,
"learning_rate": 6.108786610878662e-07,
"loss": 0.0073,
"reward": 0.7547047734260559,
"reward_std": 0.1096886545419693,
"reward_std_all": 0.11672518774867058,
"rewards-max/forensics_tempsamp_iou_reward": 0.8871729969978333,
"rewards/forensics_tempsamp_iou_reward": 0.4984547346830368,
"step": 373
},
{
"completion_length": 19.390625,
"epoch": 1.5648535564853556,
"grad_norm": 5.551793575286865,
"kl": 0.24951171875,
"learning_rate": 6.098326359832637e-07,
"loss": 0.01,
"reward": 0.8385690152645111,
"reward_std": 0.062306204810738564,
"reward_std_all": 0.08394905924797058,
"rewards-max/forensics_tempsamp_iou_reward": 0.9032203555107117,
"rewards/forensics_tempsamp_iou_reward": 0.5573190152645111,
"step": 374
},
{
"completion_length": 19.65625,
"epoch": 1.5690376569037658,
"grad_norm": 6.643610000610352,
"kl": 0.1455078125,
"learning_rate": 6.08786610878661e-07,
"loss": 0.0058,
"reward": 0.8116883635520935,
"reward_std": 0.09610997512936592,
"reward_std_all": 0.10216786339879036,
"rewards-max/forensics_tempsamp_iou_reward": 0.9471110701560974,
"rewards/forensics_tempsamp_iou_reward": 0.5491883605718613,
"step": 375
},
{
"completion_length": 19.125,
"epoch": 1.5732217573221758,
"grad_norm": 4.574708461761475,
"kl": 0.117919921875,
"learning_rate": 6.077405857740585e-07,
"loss": 0.0047,
"reward": 0.8414853513240814,
"reward_std": 0.08248526230454445,
"reward_std_all": 0.08644799143075943,
"rewards-max/forensics_tempsamp_iou_reward": 0.9479278326034546,
"rewards/forensics_tempsamp_iou_reward": 0.5789854228496552,
"step": 376
},
{
"completion_length": 19.671875,
"epoch": 1.5774058577405858,
"grad_norm": 4.045705318450928,
"kl": 0.15283203125,
"learning_rate": 6.06694560669456e-07,
"loss": 0.0061,
"reward": 0.7221877872943878,
"reward_std": 0.0808490477502346,
"reward_std_all": 0.08108093962073326,
"rewards-max/forensics_tempsamp_iou_reward": 0.951321542263031,
"rewards/forensics_tempsamp_iou_reward": 0.459687739610672,
"step": 377
},
{
"completion_length": 20.0625,
"epoch": 1.5815899581589958,
"grad_norm": 5.769662380218506,
"kl": 0.34228515625,
"learning_rate": 6.056485355648535e-07,
"loss": 0.0136,
"reward": 0.6445153653621674,
"reward_std": 0.11345713958144188,
"reward_std_all": 0.140708576887846,
"rewards-max/forensics_tempsamp_iou_reward": 0.9839609861373901,
"rewards/forensics_tempsamp_iou_reward": 0.4257653057575226,
"step": 378
},
{
"completion_length": 20.390625,
"epoch": 1.5857740585774058,
"grad_norm": 3.7109971046447754,
"kl": 0.1376953125,
"learning_rate": 6.04602510460251e-07,
"loss": 0.0055,
"reward": 0.7714243233203888,
"reward_std": 0.08222905360162258,
"reward_std_all": 0.09612865746021271,
"rewards-max/forensics_tempsamp_iou_reward": 0.9052736461162567,
"rewards/forensics_tempsamp_iou_reward": 0.5276742875576019,
"step": 379
},
{
"completion_length": 19.953125,
"epoch": 1.5899581589958158,
"grad_norm": 3.0301547050476074,
"kl": 0.21142578125,
"learning_rate": 6.035564853556485e-07,
"loss": 0.0085,
"reward": 0.8037003576755524,
"reward_std": 0.0990007147192955,
"reward_std_all": 0.1116030365228653,
"rewards-max/forensics_tempsamp_iou_reward": 0.83189657330513,
"rewards/forensics_tempsamp_iou_reward": 0.5474503636360168,
"step": 380
},
{
"completion_length": 19.5625,
"epoch": 1.5941422594142258,
"grad_norm": 7.319838047027588,
"kl": 0.4169921875,
"learning_rate": 6.02510460251046e-07,
"loss": 0.0166,
"reward": 0.613265872001648,
"reward_std": 0.10758588835597038,
"reward_std_all": 0.10475553572177887,
"rewards-max/forensics_tempsamp_iou_reward": 0.8893766403198242,
"rewards/forensics_tempsamp_iou_reward": 0.4132658541202545,
"step": 381
},
{
"completion_length": 21.171875,
"epoch": 1.5983263598326358,
"grad_norm": 5.047714710235596,
"kl": 0.14794921875,
"learning_rate": 6.014644351464436e-07,
"loss": 0.0059,
"reward": 0.7172858715057373,
"reward_std": 0.06746296212077141,
"reward_std_all": 0.09526190906763077,
"rewards-max/forensics_tempsamp_iou_reward": 0.9236111044883728,
"rewards/forensics_tempsamp_iou_reward": 0.4922858327627182,
"step": 382
},
{
"completion_length": 19.671875,
"epoch": 1.602510460251046,
"grad_norm": 4.262624263763428,
"kl": 0.15771484375,
"learning_rate": 6.00418410041841e-07,
"loss": 0.0063,
"reward": 0.7051416039466858,
"reward_std": 0.09424840658903122,
"reward_std_all": 0.0969877764582634,
"rewards-max/forensics_tempsamp_iou_reward": 0.890315592288971,
"rewards/forensics_tempsamp_iou_reward": 0.4801415950059891,
"step": 383
},
{
"completion_length": 20.6875,
"epoch": 1.606694560669456,
"grad_norm": 3.243622303009033,
"kl": 0.14013671875,
"learning_rate": 5.993723849372385e-07,
"loss": 0.0056,
"reward": 0.8012344837188721,
"reward_std": 0.1097191609442234,
"reward_std_all": 0.10152425244450569,
"rewards-max/forensics_tempsamp_iou_reward": 0.9742083847522736,
"rewards/forensics_tempsamp_iou_reward": 0.5449844002723694,
"step": 384
},
{
"completion_length": 20.828125,
"epoch": 1.6108786610878663,
"grad_norm": 3.9262754917144775,
"kl": 0.24169921875,
"learning_rate": 5.983263598326359e-07,
"loss": 0.0097,
"reward": 0.5642031133174896,
"reward_std": 0.11672480404376984,
"reward_std_all": 0.1083630919456482,
"rewards-max/forensics_tempsamp_iou_reward": 0.9425374865531921,
"rewards/forensics_tempsamp_iou_reward": 0.3704531341791153,
"step": 385
},
{
"completion_length": 19.78125,
"epoch": 1.6150627615062763,
"grad_norm": 4.146145820617676,
"kl": 0.1591796875,
"learning_rate": 5.972803347280334e-07,
"loss": 0.0064,
"reward": 0.7142733633518219,
"reward_std": 0.09842085093259811,
"reward_std_all": 0.12222578376531601,
"rewards-max/forensics_tempsamp_iou_reward": 0.939790815114975,
"rewards/forensics_tempsamp_iou_reward": 0.4642733484506607,
"step": 386
},
{
"completion_length": 19.859375,
"epoch": 1.6192468619246863,
"grad_norm": 4.9521484375,
"kl": 0.22021484375,
"learning_rate": 5.962343096234309e-07,
"loss": 0.0088,
"reward": 0.7345221042633057,
"reward_std": 0.09285440295934677,
"reward_std_all": 0.11335794255137444,
"rewards-max/forensics_tempsamp_iou_reward": 0.8880292475223541,
"rewards/forensics_tempsamp_iou_reward": 0.48452208936214447,
"step": 387
},
{
"completion_length": 19.734375,
"epoch": 1.6234309623430963,
"grad_norm": 3.7840123176574707,
"kl": 0.27294921875,
"learning_rate": 5.951882845188284e-07,
"loss": 0.0109,
"reward": 0.831823080778122,
"reward_std": 0.121603824198246,
"reward_std_all": 0.11839570105075836,
"rewards-max/forensics_tempsamp_iou_reward": 0.9297360777854919,
"rewards/forensics_tempsamp_iou_reward": 0.5568230599164963,
"step": 388
},
{
"completion_length": 19.28125,
"epoch": 1.6276150627615062,
"grad_norm": 4.968148231506348,
"kl": 0.197265625,
"learning_rate": 5.941422594142259e-07,
"loss": 0.0079,
"reward": 0.8054818511009216,
"reward_std": 0.1523936204612255,
"reward_std_all": 0.16030821204185486,
"rewards-max/forensics_tempsamp_iou_reward": 0.9263400435447693,
"rewards/forensics_tempsamp_iou_reward": 0.5179818272590637,
"step": 389
},
{
"completion_length": 20.78125,
"epoch": 1.6317991631799162,
"grad_norm": 3.8251447677612305,
"kl": 0.122314453125,
"learning_rate": 5.930962343096235e-07,
"loss": 0.0049,
"reward": 0.7346691489219666,
"reward_std": 0.10089613497257233,
"reward_std_all": 0.0997934341430664,
"rewards-max/forensics_tempsamp_iou_reward": 0.89878049492836,
"rewards/forensics_tempsamp_iou_reward": 0.47841912508010864,
"step": 390
},
{
"completion_length": 22.046875,
"epoch": 1.6359832635983262,
"grad_norm": 3.1214230060577393,
"kl": 0.14599609375,
"learning_rate": 5.92050209205021e-07,
"loss": 0.0059,
"reward": 0.7487303614616394,
"reward_std": 0.09307082742452621,
"reward_std_all": 0.11043956503272057,
"rewards-max/forensics_tempsamp_iou_reward": 0.8986805975437164,
"rewards/forensics_tempsamp_iou_reward": 0.5049803555011749,
"step": 391
},
{
"completion_length": 22.0,
"epoch": 1.6401673640167362,
"grad_norm": 4.429520606994629,
"kl": 0.267578125,
"learning_rate": 5.910041841004184e-07,
"loss": 0.0107,
"reward": 0.88264000415802,
"reward_std": 0.11040260270237923,
"reward_std_all": 0.13103562965989113,
"rewards-max/forensics_tempsamp_iou_reward": 0.9423631727695465,
"rewards/forensics_tempsamp_iou_reward": 0.5826400518417358,
"step": 392
},
{
"completion_length": 20.625,
"epoch": 1.6443514644351465,
"grad_norm": 4.093129634857178,
"kl": 0.141357421875,
"learning_rate": 5.899581589958159e-07,
"loss": 0.0056,
"reward": 0.6111509799957275,
"reward_std": 0.0978346411138773,
"reward_std_all": 0.10533474013209343,
"rewards-max/forensics_tempsamp_iou_reward": 0.9212121367454529,
"rewards/forensics_tempsamp_iou_reward": 0.4174010008573532,
"step": 393
},
{
"completion_length": 23.640625,
"epoch": 1.6485355648535565,
"grad_norm": 3.553910255432129,
"kl": 0.20751953125,
"learning_rate": 5.889121338912134e-07,
"loss": 0.0083,
"reward": 0.5875755846500397,
"reward_std": 0.11845404282212257,
"reward_std_all": 0.10984664782881737,
"rewards-max/forensics_tempsamp_iou_reward": 0.8034631609916687,
"rewards/forensics_tempsamp_iou_reward": 0.3750755414366722,
"step": 394
},
{
"completion_length": 20.53125,
"epoch": 1.6527196652719667,
"grad_norm": 3.8198163509368896,
"kl": 0.171875,
"learning_rate": 5.878661087866108e-07,
"loss": 0.0069,
"reward": 0.7515853941440582,
"reward_std": 0.15178415924310684,
"reward_std_all": 0.14150815457105637,
"rewards-max/forensics_tempsamp_iou_reward": 0.913559764623642,
"rewards/forensics_tempsamp_iou_reward": 0.5265854001045227,
"step": 395
},
{
"completion_length": 20.796875,
"epoch": 1.6569037656903767,
"grad_norm": 4.464111328125,
"kl": 0.132568359375,
"learning_rate": 5.868200836820083e-07,
"loss": 0.0053,
"reward": 0.6798057854175568,
"reward_std": 0.10062244534492493,
"reward_std_all": 0.10073244199156761,
"rewards-max/forensics_tempsamp_iou_reward": 0.9424925446510315,
"rewards/forensics_tempsamp_iou_reward": 0.4423057436943054,
"step": 396
},
{
"completion_length": 21.6875,
"epoch": 1.6610878661087867,
"grad_norm": 4.397378921508789,
"kl": 0.1708984375,
"learning_rate": 5.857740585774058e-07,
"loss": 0.0068,
"reward": 0.6643034815788269,
"reward_std": 0.14805971831083298,
"reward_std_all": 0.14586585387587547,
"rewards-max/forensics_tempsamp_iou_reward": 0.9283192455768585,
"rewards/forensics_tempsamp_iou_reward": 0.45805343985557556,
"step": 397
},
{
"completion_length": 19.953125,
"epoch": 1.6652719665271967,
"grad_norm": 4.426388740539551,
"kl": 0.146484375,
"learning_rate": 5.847280334728033e-07,
"loss": 0.0059,
"reward": 0.6910877525806427,
"reward_std": 0.12503433227539062,
"reward_std_all": 0.13146088272333145,
"rewards-max/forensics_tempsamp_iou_reward": 0.979203999042511,
"rewards/forensics_tempsamp_iou_reward": 0.44733773171901703,
"step": 398
},
{
"completion_length": 20.390625,
"epoch": 1.6694560669456067,
"grad_norm": 3.5130980014801025,
"kl": 0.147705078125,
"learning_rate": 5.836820083682009e-07,
"loss": 0.0059,
"reward": 0.759837806224823,
"reward_std": 0.11214737221598625,
"reward_std_all": 0.09798536449670792,
"rewards-max/forensics_tempsamp_iou_reward": 0.9468344151973724,
"rewards/forensics_tempsamp_iou_reward": 0.49733781814575195,
"step": 399
},
{
"completion_length": 21.359375,
"epoch": 1.6736401673640167,
"grad_norm": 4.9252824783325195,
"kl": 0.16357421875,
"learning_rate": 5.826359832635983e-07,
"loss": 0.0065,
"reward": 0.668942779302597,
"reward_std": 0.14475005120038986,
"reward_std_all": 0.1378636620938778,
"rewards-max/forensics_tempsamp_iou_reward": 0.9356997013092041,
"rewards/forensics_tempsamp_iou_reward": 0.45019277930259705,
"step": 400
},
{
"completion_length": 19.734375,
"epoch": 1.6778242677824267,
"grad_norm": 4.085936546325684,
"kl": 0.15673828125,
"learning_rate": 5.815899581589958e-07,
"loss": 0.0063,
"reward": 0.745583713054657,
"reward_std": 0.13699695467948914,
"reward_std_all": 0.12166626006364822,
"rewards-max/forensics_tempsamp_iou_reward": 0.905004233121872,
"rewards/forensics_tempsamp_iou_reward": 0.5205836892127991,
"step": 401
},
{
"completion_length": 21.0625,
"epoch": 1.6820083682008367,
"grad_norm": 3.907957077026367,
"kl": 0.141845703125,
"learning_rate": 5.805439330543933e-07,
"loss": 0.0057,
"reward": 0.6514643728733063,
"reward_std": 0.08864112943410873,
"reward_std_all": 0.11040578782558441,
"rewards-max/forensics_tempsamp_iou_reward": 0.9702342748641968,
"rewards/forensics_tempsamp_iou_reward": 0.45146429538726807,
"step": 402
},
{
"completion_length": 20.390625,
"epoch": 1.6861924686192469,
"grad_norm": 8.717144012451172,
"kl": 0.177734375,
"learning_rate": 5.794979079497908e-07,
"loss": 0.0071,
"reward": 0.6391049921512604,
"reward_std": 0.13380855321884155,
"reward_std_all": 0.15184561908245087,
"rewards-max/forensics_tempsamp_iou_reward": 0.9202960431575775,
"rewards/forensics_tempsamp_iou_reward": 0.40785495936870575,
"step": 403
},
{
"completion_length": 19.375,
"epoch": 1.6903765690376569,
"grad_norm": 4.062610149383545,
"kl": 0.1513671875,
"learning_rate": 5.784518828451883e-07,
"loss": 0.0061,
"reward": 0.741644948720932,
"reward_std": 0.11197156086564064,
"reward_std_all": 0.13361864164471626,
"rewards-max/forensics_tempsamp_iou_reward": 0.9618658125400543,
"rewards/forensics_tempsamp_iou_reward": 0.5103949308395386,
"step": 404
},
{
"completion_length": 19.859375,
"epoch": 1.694560669456067,
"grad_norm": 4.445327281951904,
"kl": 0.1513671875,
"learning_rate": 5.774058577405857e-07,
"loss": 0.0061,
"reward": 0.6875603199005127,
"reward_std": 0.1547994241118431,
"reward_std_all": 0.14484619349241257,
"rewards-max/forensics_tempsamp_iou_reward": 0.8951769471168518,
"rewards/forensics_tempsamp_iou_reward": 0.45631030201911926,
"step": 405
},
{
"completion_length": 19.671875,
"epoch": 1.698744769874477,
"grad_norm": 6.792513370513916,
"kl": 0.19091796875,
"learning_rate": 5.763598326359832e-07,
"loss": 0.0076,
"reward": 0.7082102000713348,
"reward_std": 0.11494498699903488,
"reward_std_all": 0.11481763049960136,
"rewards-max/forensics_tempsamp_iou_reward": 0.9525451958179474,
"rewards/forensics_tempsamp_iou_reward": 0.48946017026901245,
"step": 406
},
{
"completion_length": 19.765625,
"epoch": 1.702928870292887,
"grad_norm": 4.766568183898926,
"kl": 0.16259765625,
"learning_rate": 5.753138075313807e-07,
"loss": 0.0065,
"reward": 0.6715414226055145,
"reward_std": 0.14304320886731148,
"reward_std_all": 0.15483016520738602,
"rewards-max/forensics_tempsamp_iou_reward": 0.8933103978633881,
"rewards/forensics_tempsamp_iou_reward": 0.4402913749217987,
"step": 407
},
{
"completion_length": 20.34375,
"epoch": 1.707112970711297,
"grad_norm": 3.953510046005249,
"kl": 0.271484375,
"learning_rate": 5.742677824267782e-07,
"loss": 0.0108,
"reward": 0.7983459234237671,
"reward_std": 0.12665104493498802,
"reward_std_all": 0.13351493701338768,
"rewards-max/forensics_tempsamp_iou_reward": 0.931801050901413,
"rewards/forensics_tempsamp_iou_reward": 0.523345947265625,
"step": 408
},
{
"completion_length": 20.3125,
"epoch": 1.711297071129707,
"grad_norm": 6.187833786010742,
"kl": 0.1279296875,
"learning_rate": 5.732217573221757e-07,
"loss": 0.0051,
"reward": 0.7691003978252411,
"reward_std": 0.14409371465444565,
"reward_std_all": 0.13813571259379387,
"rewards-max/forensics_tempsamp_iou_reward": 0.8829086422920227,
"rewards/forensics_tempsamp_iou_reward": 0.5316004157066345,
"step": 409
},
{
"completion_length": 20.328125,
"epoch": 1.715481171548117,
"grad_norm": 5.169047832489014,
"kl": 0.139892578125,
"learning_rate": 5.721757322175732e-07,
"loss": 0.0056,
"reward": 0.78391033411026,
"reward_std": 0.12004062905907631,
"reward_std_all": 0.11579746752977371,
"rewards-max/forensics_tempsamp_iou_reward": 0.972826212644577,
"rewards/forensics_tempsamp_iou_reward": 0.5276602804660797,
"step": 410
},
{
"completion_length": 21.359375,
"epoch": 1.719665271966527,
"grad_norm": 3.453080892562866,
"kl": 0.152099609375,
"learning_rate": 5.711297071129707e-07,
"loss": 0.0061,
"reward": 0.8528542518615723,
"reward_std": 0.08330647274851799,
"reward_std_all": 0.08108551427721977,
"rewards-max/forensics_tempsamp_iou_reward": 0.9718764126300812,
"rewards/forensics_tempsamp_iou_reward": 0.5778542160987854,
"step": 411
},
{
"completion_length": 24.46875,
"epoch": 1.723849372384937,
"grad_norm": 5.244958400726318,
"kl": 0.1630859375,
"learning_rate": 5.700836820083682e-07,
"loss": 0.0065,
"reward": 0.5416778922080994,
"reward_std": 0.10943484306335449,
"reward_std_all": 0.11760454624891281,
"rewards-max/forensics_tempsamp_iou_reward": 0.9452441930770874,
"rewards/forensics_tempsamp_iou_reward": 0.335427850484848,
"step": 412
},
{
"completion_length": 19.515625,
"epoch": 1.7280334728033473,
"grad_norm": 5.316346168518066,
"kl": 0.2412109375,
"learning_rate": 5.690376569037657e-07,
"loss": 0.0096,
"reward": 0.7208540141582489,
"reward_std": 0.13976770639419556,
"reward_std_all": 0.14901021122932434,
"rewards-max/forensics_tempsamp_iou_reward": 0.9542318880558014,
"rewards/forensics_tempsamp_iou_reward": 0.48335403203964233,
"step": 413
},
{
"completion_length": 21.078125,
"epoch": 1.7322175732217573,
"grad_norm": 4.383340358734131,
"kl": 0.15869140625,
"learning_rate": 5.679916317991632e-07,
"loss": 0.0063,
"reward": 0.8109701573848724,
"reward_std": 0.07969758659601212,
"reward_std_all": 0.10147956758737564,
"rewards-max/forensics_tempsamp_iou_reward": 0.9123515784740448,
"rewards/forensics_tempsamp_iou_reward": 0.5547201037406921,
"step": 414
},
{
"completion_length": 21.015625,
"epoch": 1.7364016736401675,
"grad_norm": 3.641148567199707,
"kl": 0.171875,
"learning_rate": 5.669456066945606e-07,
"loss": 0.0069,
"reward": 0.7207403779029846,
"reward_std": 0.07515552267432213,
"reward_std_all": 0.07417905330657959,
"rewards-max/forensics_tempsamp_iou_reward": 0.9628506600856781,
"rewards/forensics_tempsamp_iou_reward": 0.47699040174484253,
"step": 415
},
{
"completion_length": 20.578125,
"epoch": 1.7405857740585775,
"grad_norm": 4.375463485717773,
"kl": 0.2158203125,
"learning_rate": 5.65899581589958e-07,
"loss": 0.0086,
"reward": 0.6707467883825302,
"reward_std": 0.08521071448922157,
"reward_std_all": 0.09253882989287376,
"rewards-max/forensics_tempsamp_iou_reward": 0.9232481420040131,
"rewards/forensics_tempsamp_iou_reward": 0.4519967585802078,
"step": 416
},
{
"completion_length": 19.5,
"epoch": 1.7447698744769875,
"grad_norm": 3.5284013748168945,
"kl": 0.1806640625,
"learning_rate": 5.648535564853556e-07,
"loss": 0.0072,
"reward": 0.6858185529708862,
"reward_std": 0.06183498911559582,
"reward_std_all": 0.09984438493847847,
"rewards-max/forensics_tempsamp_iou_reward": 0.9613178968429565,
"rewards/forensics_tempsamp_iou_reward": 0.4795685261487961,
"step": 417
},
{
"completion_length": 20.078125,
"epoch": 1.7489539748953975,
"grad_norm": 4.590520858764648,
"kl": 0.138671875,
"learning_rate": 5.638075313807531e-07,
"loss": 0.0055,
"reward": 0.6598048210144043,
"reward_std": 0.11217747628688812,
"reward_std_all": 0.10734020173549652,
"rewards-max/forensics_tempsamp_iou_reward": 0.9497629702091217,
"rewards/forensics_tempsamp_iou_reward": 0.4285547733306885,
"step": 418
},
{
"completion_length": 19.625,
"epoch": 1.7531380753138075,
"grad_norm": 6.152802467346191,
"kl": 0.1806640625,
"learning_rate": 5.627615062761506e-07,
"loss": 0.0072,
"reward": 0.8081401884555817,
"reward_std": 0.08572977408766747,
"reward_std_all": 0.09227494150400162,
"rewards-max/forensics_tempsamp_iou_reward": 0.9558709263801575,
"rewards/forensics_tempsamp_iou_reward": 0.5518901348114014,
"step": 419
},
{
"completion_length": 19.890625,
"epoch": 1.7573221757322175,
"grad_norm": 4.699737071990967,
"kl": 0.2236328125,
"learning_rate": 5.617154811715481e-07,
"loss": 0.009,
"reward": 0.7680488228797913,
"reward_std": 0.13470616936683655,
"reward_std_all": 0.12484751641750336,
"rewards-max/forensics_tempsamp_iou_reward": 0.9413406550884247,
"rewards/forensics_tempsamp_iou_reward": 0.5305487513542175,
"step": 420
},
{
"completion_length": 19.953125,
"epoch": 1.7615062761506275,
"grad_norm": 6.765925407409668,
"kl": 0.24365234375,
"learning_rate": 5.606694560669456e-07,
"loss": 0.0098,
"reward": 0.6967054605484009,
"reward_std": 0.1415443904697895,
"reward_std_all": 0.12763941660523415,
"rewards-max/forensics_tempsamp_iou_reward": 0.9397442638874054,
"rewards/forensics_tempsamp_iou_reward": 0.484205424785614,
"step": 421
},
{
"completion_length": 19.609375,
"epoch": 1.7656903765690377,
"grad_norm": 6.17219877243042,
"kl": 0.10986328125,
"learning_rate": 5.596234309623431e-07,
"loss": 0.0044,
"reward": 0.7020261287689209,
"reward_std": 0.09228173829615116,
"reward_std_all": 0.12198314815759659,
"rewards-max/forensics_tempsamp_iou_reward": 0.9401964843273163,
"rewards/forensics_tempsamp_iou_reward": 0.47077612578868866,
"step": 422
},
{
"completion_length": 20.46875,
"epoch": 1.7698744769874477,
"grad_norm": 5.408677577972412,
"kl": 0.146728515625,
"learning_rate": 5.585774058577406e-07,
"loss": 0.0059,
"reward": 0.6512818336486816,
"reward_std": 0.13994429260492325,
"reward_std_all": 0.17609305679798126,
"rewards-max/forensics_tempsamp_iou_reward": 0.9572934508323669,
"rewards/forensics_tempsamp_iou_reward": 0.43878184258937836,
"step": 423
},
{
"completion_length": 19.875,
"epoch": 1.7740585774058577,
"grad_norm": 4.525363445281982,
"kl": 0.120361328125,
"learning_rate": 5.575313807531381e-07,
"loss": 0.0048,
"reward": 0.6665586531162262,
"reward_std": 0.15075530856847763,
"reward_std_all": 0.13754993677139282,
"rewards-max/forensics_tempsamp_iou_reward": 0.8181910514831543,
"rewards/forensics_tempsamp_iou_reward": 0.42905862629413605,
"step": 424
},
{
"completion_length": 20.453125,
"epoch": 1.778242677824268,
"grad_norm": 3.9451873302459717,
"kl": 0.20947265625,
"learning_rate": 5.564853556485354e-07,
"loss": 0.0084,
"reward": 0.6027433276176453,
"reward_std": 0.16556213796138763,
"reward_std_all": 0.16508987545967102,
"rewards-max/forensics_tempsamp_iou_reward": 0.9011151194572449,
"rewards/forensics_tempsamp_iou_reward": 0.38399331271648407,
"step": 425
},
{
"completion_length": 19.59375,
"epoch": 1.782426778242678,
"grad_norm": 4.41036319732666,
"kl": 0.232421875,
"learning_rate": 5.55439330543933e-07,
"loss": 0.0093,
"reward": 0.7905750870704651,
"reward_std": 0.11782155185937881,
"reward_std_all": 0.13790830224752426,
"rewards-max/forensics_tempsamp_iou_reward": 0.9615234732627869,
"rewards/forensics_tempsamp_iou_reward": 0.5530750751495361,
"step": 426
},
{
"completion_length": 20.671875,
"epoch": 1.786610878661088,
"grad_norm": 4.556817054748535,
"kl": 0.13818359375,
"learning_rate": 5.543933054393305e-07,
"loss": 0.0055,
"reward": 0.8201430141925812,
"reward_std": 0.0743794646114111,
"reward_std_all": 0.08874510042369366,
"rewards-max/forensics_tempsamp_iou_reward": 0.8504948019981384,
"rewards/forensics_tempsamp_iou_reward": 0.5388929843902588,
"step": 427
},
{
"completion_length": 20.953125,
"epoch": 1.790794979079498,
"grad_norm": 4.437883377075195,
"kl": 0.1484375,
"learning_rate": 5.53347280334728e-07,
"loss": 0.0059,
"reward": 0.6734131276607513,
"reward_std": 0.08554031699895859,
"reward_std_all": 0.09478030353784561,
"rewards-max/forensics_tempsamp_iou_reward": 0.767292708158493,
"rewards/forensics_tempsamp_iou_reward": 0.42966312170028687,
"step": 428
},
{
"completion_length": 20.640625,
"epoch": 1.794979079497908,
"grad_norm": 4.203531265258789,
"kl": 0.11767578125,
"learning_rate": 5.523012552301255e-07,
"loss": 0.0047,
"reward": 0.6624820828437805,
"reward_std": 0.05535481683909893,
"reward_std_all": 0.06905271485447884,
"rewards-max/forensics_tempsamp_iou_reward": 0.9642209708690643,
"rewards/forensics_tempsamp_iou_reward": 0.4499821066856384,
"step": 429
},
{
"completion_length": 20.296875,
"epoch": 1.799163179916318,
"grad_norm": 5.468527317047119,
"kl": 0.142578125,
"learning_rate": 5.51255230125523e-07,
"loss": 0.0057,
"reward": 0.6639334559440613,
"reward_std": 0.16967180371284485,
"reward_std_all": 0.17185240983963013,
"rewards-max/forensics_tempsamp_iou_reward": 0.9332874417304993,
"rewards/forensics_tempsamp_iou_reward": 0.43268343806266785,
"step": 430
},
{
"completion_length": 19.359375,
"epoch": 1.803347280334728,
"grad_norm": 3.8037455081939697,
"kl": 0.14794921875,
"learning_rate": 5.502092050209205e-07,
"loss": 0.0059,
"reward": 0.7427981495857239,
"reward_std": 0.06881360709667206,
"reward_std_all": 0.07983589172363281,
"rewards-max/forensics_tempsamp_iou_reward": 0.9282887876033783,
"rewards/forensics_tempsamp_iou_reward": 0.5177981555461884,
"step": 431
},
{
"completion_length": 19.671875,
"epoch": 1.8075313807531381,
"grad_norm": 4.110310077667236,
"kl": 0.12451171875,
"learning_rate": 5.49163179916318e-07,
"loss": 0.005,
"reward": 0.7365190386772156,
"reward_std": 0.07983171194791794,
"reward_std_all": 0.08632607758045197,
"rewards-max/forensics_tempsamp_iou_reward": 0.906836986541748,
"rewards/forensics_tempsamp_iou_reward": 0.4927690029144287,
"step": 432
},
{
"completion_length": 19.46875,
"epoch": 1.8117154811715481,
"grad_norm": 3.5568430423736572,
"kl": 0.1435546875,
"learning_rate": 5.481171548117156e-07,
"loss": 0.0057,
"reward": 0.8507027328014374,
"reward_std": 0.05915870517492294,
"reward_std_all": 0.06113400682806969,
"rewards-max/forensics_tempsamp_iou_reward": 0.9822126626968384,
"rewards/forensics_tempsamp_iou_reward": 0.5694527328014374,
"step": 433
},
{
"completion_length": 20.75,
"epoch": 1.8158995815899581,
"grad_norm": 7.631237030029297,
"kl": 0.10400390625,
"learning_rate": 5.47071129707113e-07,
"loss": 0.0042,
"reward": 0.43508394807577133,
"reward_std": 0.07766189984977245,
"reward_std_all": 0.09568028524518013,
"rewards-max/forensics_tempsamp_iou_reward": 0.5762368142604828,
"rewards/forensics_tempsamp_iou_reward": 0.25383395701646805,
"step": 434
},
{
"completion_length": 20.90625,
"epoch": 1.8200836820083683,
"grad_norm": 3.706268787384033,
"kl": 0.205078125,
"learning_rate": 5.460251046025104e-07,
"loss": 0.0082,
"reward": 0.8516391515731812,
"reward_std": 0.06919640675187111,
"reward_std_all": 0.10676006227731705,
"rewards-max/forensics_tempsamp_iou_reward": 0.9786761999130249,
"rewards/forensics_tempsamp_iou_reward": 0.5891391336917877,
"step": 435
},
{
"completion_length": 21.359375,
"epoch": 1.8242677824267783,
"grad_norm": 3.3196325302124023,
"kl": 0.20361328125,
"learning_rate": 5.449790794979079e-07,
"loss": 0.0081,
"reward": 0.6436887979507446,
"reward_std": 0.09083013236522675,
"reward_std_all": 0.10417045652866364,
"rewards-max/forensics_tempsamp_iou_reward": 0.8077735006809235,
"rewards/forensics_tempsamp_iou_reward": 0.4249388128519058,
"step": 436
},
{
"completion_length": 19.28125,
"epoch": 1.8284518828451883,
"grad_norm": 4.8309326171875,
"kl": 0.1435546875,
"learning_rate": 5.439330543933054e-07,
"loss": 0.0058,
"reward": 0.9067423343658447,
"reward_std": 0.10163940489292145,
"reward_std_all": 0.11572118103504181,
"rewards-max/forensics_tempsamp_iou_reward": 0.9398598372936249,
"rewards/forensics_tempsamp_iou_reward": 0.6129923164844513,
"step": 437
},
{
"completion_length": 19.515625,
"epoch": 1.8326359832635983,
"grad_norm": 5.852116584777832,
"kl": 0.146484375,
"learning_rate": 5.428870292887029e-07,
"loss": 0.0059,
"reward": 0.7292137145996094,
"reward_std": 0.13922729715704918,
"reward_std_all": 0.13182273134589195,
"rewards-max/forensics_tempsamp_iou_reward": 0.9502881765365601,
"rewards/forensics_tempsamp_iou_reward": 0.4792136549949646,
"step": 438
},
{
"completion_length": 21.109375,
"epoch": 1.8368200836820083,
"grad_norm": 3.743124008178711,
"kl": 0.167236328125,
"learning_rate": 5.418410041841004e-07,
"loss": 0.0067,
"reward": 0.7144513130187988,
"reward_std": 0.11778699606657028,
"reward_std_all": 0.14945252984762192,
"rewards-max/forensics_tempsamp_iou_reward": 0.9614435434341431,
"rewards/forensics_tempsamp_iou_reward": 0.47070130705833435,
"step": 439
},
{
"completion_length": 19.96875,
"epoch": 1.8410041841004183,
"grad_norm": 4.03088903427124,
"kl": 0.1396484375,
"learning_rate": 5.407949790794979e-07,
"loss": 0.0056,
"reward": 0.7124901413917542,
"reward_std": 0.11257556080818176,
"reward_std_all": 0.12474186718463898,
"rewards-max/forensics_tempsamp_iou_reward": 0.9209311306476593,
"rewards/forensics_tempsamp_iou_reward": 0.4812401086091995,
"step": 440
},
{
"completion_length": 19.859375,
"epoch": 1.8451882845188283,
"grad_norm": 4.598120212554932,
"kl": 0.171875,
"learning_rate": 5.397489539748954e-07,
"loss": 0.0069,
"reward": 0.6559779644012451,
"reward_std": 0.10927586257457733,
"reward_std_all": 0.11761520057916641,
"rewards-max/forensics_tempsamp_iou_reward": 0.9110264778137207,
"rewards/forensics_tempsamp_iou_reward": 0.43722793459892273,
"step": 441
},
{
"completion_length": 20.15625,
"epoch": 1.8493723849372385,
"grad_norm": 5.550134658813477,
"kl": 0.189453125,
"learning_rate": 5.387029288702928e-07,
"loss": 0.0076,
"reward": 0.6424511969089508,
"reward_std": 0.0779450535774231,
"reward_std_all": 0.0870528519153595,
"rewards-max/forensics_tempsamp_iou_reward": 0.9525397717952728,
"rewards/forensics_tempsamp_iou_reward": 0.4237011820077896,
"step": 442
},
{
"completion_length": 20.015625,
"epoch": 1.8535564853556485,
"grad_norm": 4.896965980529785,
"kl": 0.298828125,
"learning_rate": 5.376569037656904e-07,
"loss": 0.012,
"reward": 0.7847581803798676,
"reward_std": 0.1717252917587757,
"reward_std_all": 0.15108652412891388,
"rewards-max/forensics_tempsamp_iou_reward": 0.9514614939689636,
"rewards/forensics_tempsamp_iou_reward": 0.49725817143917084,
"step": 443
},
{
"completion_length": 20.21875,
"epoch": 1.8577405857740585,
"grad_norm": 4.739501476287842,
"kl": 0.2265625,
"learning_rate": 5.366108786610879e-07,
"loss": 0.0091,
"reward": 0.6127472221851349,
"reward_std": 0.09582417458295822,
"reward_std_all": 0.09276130795478821,
"rewards-max/forensics_tempsamp_iou_reward": 0.9443929195404053,
"rewards/forensics_tempsamp_iou_reward": 0.41274721920490265,
"step": 444
},
{
"completion_length": 20.109375,
"epoch": 1.8619246861924688,
"grad_norm": 4.267242908477783,
"kl": 0.130859375,
"learning_rate": 5.355648535564853e-07,
"loss": 0.0052,
"reward": 0.5562925040721893,
"reward_std": 0.06243395060300827,
"reward_std_all": 0.08065224438905716,
"rewards-max/forensics_tempsamp_iou_reward": 0.8880847990512848,
"rewards/forensics_tempsamp_iou_reward": 0.40629254281520844,
"step": 445
},
{
"completion_length": 20.859375,
"epoch": 1.8661087866108788,
"grad_norm": 4.213321208953857,
"kl": 0.273193359375,
"learning_rate": 5.345188284518828e-07,
"loss": 0.0109,
"reward": 0.6924384832382202,
"reward_std": 0.1242009662091732,
"reward_std_all": 0.11636490933597088,
"rewards-max/forensics_tempsamp_iou_reward": 0.9423122406005859,
"rewards/forensics_tempsamp_iou_reward": 0.47993841767311096,
"step": 446
},
{
"completion_length": 19.265625,
"epoch": 1.8702928870292888,
"grad_norm": 4.199617385864258,
"kl": 0.24755859375,
"learning_rate": 5.334728033472803e-07,
"loss": 0.0099,
"reward": 0.9171596467494965,
"reward_std": 0.10346904024481773,
"reward_std_all": 0.13632908463478088,
"rewards-max/forensics_tempsamp_iou_reward": 0.9938794672489166,
"rewards/forensics_tempsamp_iou_reward": 0.6421596109867096,
"step": 447
},
{
"completion_length": 20.40625,
"epoch": 1.8744769874476988,
"grad_norm": 6.2797160148620605,
"kl": 0.1845703125,
"learning_rate": 5.324267782426778e-07,
"loss": 0.0074,
"reward": 0.7514411211013794,
"reward_std": 0.08729777112603188,
"reward_std_all": 0.09180822223424911,
"rewards-max/forensics_tempsamp_iou_reward": 0.9271131157875061,
"rewards/forensics_tempsamp_iou_reward": 0.4951910972595215,
"step": 448
},
{
"completion_length": 20.046875,
"epoch": 1.8786610878661087,
"grad_norm": 3.1748924255371094,
"kl": 0.169921875,
"learning_rate": 5.313807531380753e-07,
"loss": 0.0068,
"reward": 0.8324241042137146,
"reward_std": 0.08243624493479729,
"reward_std_all": 0.12107046693563461,
"rewards-max/forensics_tempsamp_iou_reward": 0.9785456359386444,
"rewards/forensics_tempsamp_iou_reward": 0.5636740624904633,
"step": 449
},
{
"completion_length": 20.71875,
"epoch": 1.8828451882845187,
"grad_norm": 6.286313056945801,
"kl": 0.17626953125,
"learning_rate": 5.303347280334727e-07,
"loss": 0.0071,
"reward": 0.9500621259212494,
"reward_std": 0.16991065442562103,
"reward_std_all": 0.15105010569095612,
"rewards-max/forensics_tempsamp_iou_reward": 0.9636314809322357,
"rewards/forensics_tempsamp_iou_reward": 0.6813121736049652,
"step": 450
},
{
"completion_length": 21.96875,
"epoch": 1.8870292887029287,
"grad_norm": 8.48346996307373,
"kl": 0.1572265625,
"learning_rate": 5.292887029288703e-07,
"loss": 0.0063,
"reward": 0.8219042420387268,
"reward_std": 0.09687184169888496,
"reward_std_all": 0.09445686638355255,
"rewards-max/forensics_tempsamp_iou_reward": 0.9180921018123627,
"rewards/forensics_tempsamp_iou_reward": 0.5406542420387268,
"step": 451
},
{
"completion_length": 19.25,
"epoch": 1.891213389121339,
"grad_norm": 4.275371551513672,
"kl": 0.12744140625,
"learning_rate": 5.282426778242678e-07,
"loss": 0.0051,
"reward": 0.9466750621795654,
"reward_std": 0.09079718589782715,
"reward_std_all": 0.09029418602585793,
"rewards-max/forensics_tempsamp_iou_reward": 0.9658136069774628,
"rewards/forensics_tempsamp_iou_reward": 0.665425032377243,
"step": 452
},
{
"completion_length": 20.765625,
"epoch": 1.895397489539749,
"grad_norm": 4.400267124176025,
"kl": 0.146484375,
"learning_rate": 5.271966527196653e-07,
"loss": 0.0059,
"reward": 0.7999768257141113,
"reward_std": 0.10104827582836151,
"reward_std_all": 0.10124699771404266,
"rewards-max/forensics_tempsamp_iou_reward": 0.9248524010181427,
"rewards/forensics_tempsamp_iou_reward": 0.5249768197536469,
"step": 453
},
{
"completion_length": 20.703125,
"epoch": 1.899581589958159,
"grad_norm": 3.898847818374634,
"kl": 0.14990234375,
"learning_rate": 5.261506276150628e-07,
"loss": 0.006,
"reward": 0.7393383979797363,
"reward_std": 0.0993000902235508,
"reward_std_all": 0.09946845471858978,
"rewards-max/forensics_tempsamp_iou_reward": 0.938933789730072,
"rewards/forensics_tempsamp_iou_reward": 0.49558840692043304,
"step": 454
},
{
"completion_length": 21.296875,
"epoch": 1.9037656903765692,
"grad_norm": 4.367659568786621,
"kl": 0.15576171875,
"learning_rate": 5.251046025104602e-07,
"loss": 0.0062,
"reward": 0.7480147182941437,
"reward_std": 0.13222967833280563,
"reward_std_all": 0.13353487476706505,
"rewards-max/forensics_tempsamp_iou_reward": 0.9553551077842712,
"rewards/forensics_tempsamp_iou_reward": 0.5230146795511246,
"step": 455
},
{
"completion_length": 19.71875,
"epoch": 1.9079497907949792,
"grad_norm": 4.786739349365234,
"kl": 0.18359375,
"learning_rate": 5.240585774058577e-07,
"loss": 0.0074,
"reward": 0.7526397109031677,
"reward_std": 0.057242730632424355,
"reward_std_all": 0.06435484439134598,
"rewards-max/forensics_tempsamp_iou_reward": 0.9944644868373871,
"rewards/forensics_tempsamp_iou_reward": 0.5088897049427032,
"step": 456
},
{
"completion_length": 19.921875,
"epoch": 1.9121338912133892,
"grad_norm": 7.64514684677124,
"kl": 0.216796875,
"learning_rate": 5.230125523012552e-07,
"loss": 0.0087,
"reward": 0.9166000485420227,
"reward_std": 0.09590192884206772,
"reward_std_all": 0.08862397819757462,
"rewards-max/forensics_tempsamp_iou_reward": 0.9284242987632751,
"rewards/forensics_tempsamp_iou_reward": 0.6416000723838806,
"step": 457
},
{
"completion_length": 20.8125,
"epoch": 1.9163179916317992,
"grad_norm": 9.58333969116211,
"kl": 0.169921875,
"learning_rate": 5.219665271966526e-07,
"loss": 0.0068,
"reward": 0.5387391895055771,
"reward_std": 0.06076199188828468,
"reward_std_all": 0.05701957456767559,
"rewards-max/forensics_tempsamp_iou_reward": 0.8736865818500519,
"rewards/forensics_tempsamp_iou_reward": 0.3574891537427902,
"step": 458
},
{
"completion_length": 21.890625,
"epoch": 1.9205020920502092,
"grad_norm": 5.249431133270264,
"kl": 0.20556640625,
"learning_rate": 5.209205020920501e-07,
"loss": 0.0082,
"reward": 0.6473446786403656,
"reward_std": 0.16444868221879005,
"reward_std_all": 0.16361460089683533,
"rewards-max/forensics_tempsamp_iou_reward": 0.9723826348781586,
"rewards/forensics_tempsamp_iou_reward": 0.447344645857811,
"step": 459
},
{
"completion_length": 19.703125,
"epoch": 1.9246861924686192,
"grad_norm": 5.504107475280762,
"kl": 0.1611328125,
"learning_rate": 5.198744769874477e-07,
"loss": 0.0064,
"reward": 0.8437934219837189,
"reward_std": 0.12214842066168785,
"reward_std_all": 0.11379506438970566,
"rewards-max/forensics_tempsamp_iou_reward": 0.9323796629905701,
"rewards/forensics_tempsamp_iou_reward": 0.5750434100627899,
"step": 460
},
{
"completion_length": 20.203125,
"epoch": 1.9288702928870292,
"grad_norm": 5.2310895919799805,
"kl": 0.1767578125,
"learning_rate": 5.188284518828452e-07,
"loss": 0.0071,
"reward": 0.6221912503242493,
"reward_std": 0.13697515055537224,
"reward_std_all": 0.1258988119661808,
"rewards-max/forensics_tempsamp_iou_reward": 0.9117108881473541,
"rewards/forensics_tempsamp_iou_reward": 0.42844127118587494,
"step": 461
},
{
"completion_length": 20.703125,
"epoch": 1.9330543933054394,
"grad_norm": 4.161437511444092,
"kl": 0.20458984375,
"learning_rate": 5.177824267782427e-07,
"loss": 0.0082,
"reward": 0.8201517462730408,
"reward_std": 0.08441862463951111,
"reward_std_all": 0.09203736111521721,
"rewards-max/forensics_tempsamp_iou_reward": 0.9547144770622253,
"rewards/forensics_tempsamp_iou_reward": 0.5576517581939697,
"step": 462
},
{
"completion_length": 20.796875,
"epoch": 1.9372384937238494,
"grad_norm": 4.291947841644287,
"kl": 0.220703125,
"learning_rate": 5.167364016736402e-07,
"loss": 0.0088,
"reward": 0.7974441349506378,
"reward_std": 0.11584929004311562,
"reward_std_all": 0.10452156141400337,
"rewards-max/forensics_tempsamp_iou_reward": 0.9087044894695282,
"rewards/forensics_tempsamp_iou_reward": 0.5286940932273865,
"step": 463
},
{
"completion_length": 19.953125,
"epoch": 1.9414225941422594,
"grad_norm": 4.864584445953369,
"kl": 0.17822265625,
"learning_rate": 5.156903765690377e-07,
"loss": 0.0071,
"reward": 0.7410966753959656,
"reward_std": 0.1232747957110405,
"reward_std_all": 0.12275166809558868,
"rewards-max/forensics_tempsamp_iou_reward": 0.8933026194572449,
"rewards/forensics_tempsamp_iou_reward": 0.4848466217517853,
"step": 464
},
{
"completion_length": 22.28125,
"epoch": 1.9456066945606696,
"grad_norm": 4.2172532081604,
"kl": 0.3017578125,
"learning_rate": 5.146443514644351e-07,
"loss": 0.0121,
"reward": 0.6278038322925568,
"reward_std": 0.08161557652056217,
"reward_std_all": 0.07905913144350052,
"rewards-max/forensics_tempsamp_iou_reward": 0.8738830387592316,
"rewards/forensics_tempsamp_iou_reward": 0.40280376374721527,
"step": 465
},
{
"completion_length": 21.234375,
"epoch": 1.9497907949790796,
"grad_norm": 4.727593421936035,
"kl": 0.21630859375,
"learning_rate": 5.135983263598326e-07,
"loss": 0.0086,
"reward": 0.5474130213260651,
"reward_std": 0.09241268783807755,
"reward_std_all": 0.09237403981387615,
"rewards-max/forensics_tempsamp_iou_reward": 0.906915932893753,
"rewards/forensics_tempsamp_iou_reward": 0.37866300344467163,
"step": 466
},
{
"completion_length": 20.765625,
"epoch": 1.9539748953974896,
"grad_norm": 11.916687965393066,
"kl": 0.19970703125,
"learning_rate": 5.1255230125523e-07,
"loss": 0.008,
"reward": 0.733984649181366,
"reward_std": 0.08153478801250458,
"reward_std_all": 0.10414128005504608,
"rewards-max/forensics_tempsamp_iou_reward": 0.9116531014442444,
"rewards/forensics_tempsamp_iou_reward": 0.5089846253395081,
"step": 467
},
{
"completion_length": 20.390625,
"epoch": 1.9581589958158996,
"grad_norm": 6.949976444244385,
"kl": 0.20263671875,
"learning_rate": 5.115062761506275e-07,
"loss": 0.0081,
"reward": 0.5776670873165131,
"reward_std": 0.06543304957449436,
"reward_std_all": 0.061315424740314484,
"rewards-max/forensics_tempsamp_iou_reward": 0.6874517798423767,
"rewards/forensics_tempsamp_iou_reward": 0.35266706347465515,
"step": 468
},
{
"completion_length": 19.65625,
"epoch": 1.9623430962343096,
"grad_norm": 5.099358558654785,
"kl": 0.173828125,
"learning_rate": 5.104602510460251e-07,
"loss": 0.0069,
"reward": 0.7740083038806915,
"reward_std": 0.05054723657667637,
"reward_std_all": 0.09387540072202682,
"rewards-max/forensics_tempsamp_iou_reward": 0.9451797306537628,
"rewards/forensics_tempsamp_iou_reward": 0.530258297920227,
"step": 469
},
{
"completion_length": 19.859375,
"epoch": 1.9665271966527196,
"grad_norm": 6.082616806030273,
"kl": 0.20458984375,
"learning_rate": 5.094142259414226e-07,
"loss": 0.0082,
"reward": 0.6716298162937164,
"reward_std": 0.08644292503595352,
"reward_std_all": 0.07809855043888092,
"rewards-max/forensics_tempsamp_iou_reward": 0.9431059062480927,
"rewards/forensics_tempsamp_iou_reward": 0.48412978649139404,
"step": 470
},
{
"completion_length": 19.578125,
"epoch": 1.9707112970711296,
"grad_norm": 3.6755566596984863,
"kl": 0.15625,
"learning_rate": 5.083682008368201e-07,
"loss": 0.0062,
"reward": 0.7911820411682129,
"reward_std": 0.07231957092881203,
"reward_std_all": 0.0839817114174366,
"rewards-max/forensics_tempsamp_iou_reward": 0.8835620582103729,
"rewards/forensics_tempsamp_iou_reward": 0.5411819666624069,
"step": 471
},
{
"completion_length": 21.453125,
"epoch": 1.9748953974895398,
"grad_norm": 6.328579902648926,
"kl": 0.14111328125,
"learning_rate": 5.073221757322176e-07,
"loss": 0.0056,
"reward": 0.5331286042928696,
"reward_std": 0.11225835606455803,
"reward_std_all": 0.10888328403234482,
"rewards-max/forensics_tempsamp_iou_reward": 0.9235446751117706,
"rewards/forensics_tempsamp_iou_reward": 0.3518785983324051,
"step": 472
},
{
"completion_length": 19.65625,
"epoch": 1.9790794979079498,
"grad_norm": 6.455973148345947,
"kl": 0.123779296875,
"learning_rate": 5.062761506276151e-07,
"loss": 0.005,
"reward": 0.6150353848934174,
"reward_std": 0.061394089832901955,
"reward_std_all": 0.06223496049642563,
"rewards-max/forensics_tempsamp_iou_reward": 0.8500346541404724,
"rewards/forensics_tempsamp_iou_reward": 0.42753538489341736,
"step": 473
},
{
"completion_length": 19.71875,
"epoch": 1.9832635983263598,
"grad_norm": 4.187480926513672,
"kl": 0.15625,
"learning_rate": 5.052301255230126e-07,
"loss": 0.0063,
"reward": 0.8898547291755676,
"reward_std": 0.05991816334426403,
"reward_std_all": 0.06613104231655598,
"rewards-max/forensics_tempsamp_iou_reward": 0.9762676954269409,
"rewards/forensics_tempsamp_iou_reward": 0.60860475897789,
"step": 474
},
{
"completion_length": 19.78125,
"epoch": 1.98744769874477,
"grad_norm": 25.833904266357422,
"kl": 1.6630859375,
"learning_rate": 5.041841004184099e-07,
"loss": 0.0665,
"reward": 0.5743933022022247,
"reward_std": 0.10331925749778748,
"reward_std_all": 0.15259242057800293,
"rewards-max/forensics_tempsamp_iou_reward": 0.8552265465259552,
"rewards/forensics_tempsamp_iou_reward": 0.3806432783603668,
"step": 475
},
{
"completion_length": 20.484375,
"epoch": 1.99163179916318,
"grad_norm": 5.823233604431152,
"kl": 0.205078125,
"learning_rate": 5.031380753138074e-07,
"loss": 0.0082,
"reward": 0.7260853350162506,
"reward_std": 0.10899453982710838,
"reward_std_all": 0.09977227449417114,
"rewards-max/forensics_tempsamp_iou_reward": 0.9656015932559967,
"rewards/forensics_tempsamp_iou_reward": 0.5135852992534637,
"step": 476
},
{
"completion_length": 19.5625,
"epoch": 1.99581589958159,
"grad_norm": 7.408677577972412,
"kl": 0.6171875,
"learning_rate": 5.02092050209205e-07,
"loss": 0.0247,
"reward": 0.6892688572406769,
"reward_std": 0.13172685354948044,
"reward_std_all": 0.11816461756825447,
"rewards-max/forensics_tempsamp_iou_reward": 0.9852905571460724,
"rewards/forensics_tempsamp_iou_reward": 0.4830188453197479,
"step": 477
},
{
"completion_length": 18.928571701049805,
"epoch": 2.0,
"grad_norm": 7.964142322540283,
"kl": 0.390625,
"learning_rate": 5.010460251046025e-07,
"loss": 0.0144,
"reward": 0.8028178811073303,
"reward_std": 0.12956472113728523,
"reward_std_all": 0.10137015208601952,
"rewards-max/forensics_tempsamp_iou_reward": 0.8525775671005249,
"rewards/forensics_tempsamp_iou_reward": 0.5671035051345825,
"step": 478
},
{
"completion_length": 19.65625,
"epoch": 2.00418410041841,
"grad_norm": 8.527399063110352,
"kl": 0.2431640625,
"learning_rate": 5e-07,
"loss": 0.0097,
"reward": 0.809595376253128,
"reward_std": 0.11371487379074097,
"reward_std_all": 0.1350804716348648,
"rewards-max/forensics_tempsamp_iou_reward": 0.903954952955246,
"rewards/forensics_tempsamp_iou_reward": 0.5408453345298767,
"step": 479
},
{
"completion_length": 19.609375,
"epoch": 2.00836820083682,
"grad_norm": 6.7841572761535645,
"kl": 0.22119140625,
"learning_rate": 4.989539748953975e-07,
"loss": 0.0089,
"reward": 0.7461375594139099,
"reward_std": 0.13980356976389885,
"reward_std_all": 0.13120391219854355,
"rewards-max/forensics_tempsamp_iou_reward": 0.986119955778122,
"rewards/forensics_tempsamp_iou_reward": 0.5086375772953033,
"step": 480
},
{
"completion_length": 20.59375,
"epoch": 2.01255230125523,
"grad_norm": 3.4995598793029785,
"kl": 0.12939453125,
"learning_rate": 4.97907949790795e-07,
"loss": 0.0052,
"reward": 0.8245829343795776,
"reward_std": 0.07631887122988701,
"reward_std_all": 0.08363746106624603,
"rewards-max/forensics_tempsamp_iou_reward": 0.9482442438602448,
"rewards/forensics_tempsamp_iou_reward": 0.562082976102829,
"step": 481
},
{
"completion_length": 22.890625,
"epoch": 2.01673640167364,
"grad_norm": 4.607640743255615,
"kl": 0.2060546875,
"learning_rate": 4.968619246861924e-07,
"loss": 0.0082,
"reward": 0.803694874048233,
"reward_std": 0.10944909229874611,
"reward_std_all": 0.12096577510237694,
"rewards-max/forensics_tempsamp_iou_reward": 0.862613320350647,
"rewards/forensics_tempsamp_iou_reward": 0.5474448502063751,
"step": 482
},
{
"completion_length": 20.390625,
"epoch": 2.02092050209205,
"grad_norm": 4.900407314300537,
"kl": 0.2431640625,
"learning_rate": 4.958158995815899e-07,
"loss": 0.0097,
"reward": 0.8390975594520569,
"reward_std": 0.09010681882500648,
"reward_std_all": 0.08426626026630402,
"rewards-max/forensics_tempsamp_iou_reward": 0.9095088541507721,
"rewards/forensics_tempsamp_iou_reward": 0.5578474998474121,
"step": 483
},
{
"completion_length": 20.21875,
"epoch": 2.0251046025104604,
"grad_norm": 5.106548309326172,
"kl": 0.3515625,
"learning_rate": 4.947698744769874e-07,
"loss": 0.014,
"reward": 0.9405445456504822,
"reward_std": 0.08504585549235344,
"reward_std_all": 0.09903440997004509,
"rewards-max/forensics_tempsamp_iou_reward": 0.947987973690033,
"rewards/forensics_tempsamp_iou_reward": 0.6405445337295532,
"step": 484
},
{
"completion_length": 19.4375,
"epoch": 2.0292887029288704,
"grad_norm": 4.192861080169678,
"kl": 0.15087890625,
"learning_rate": 4.93723849372385e-07,
"loss": 0.006,
"reward": 0.8466391861438751,
"reward_std": 0.07771428301930428,
"reward_std_all": 0.07493259757757187,
"rewards-max/forensics_tempsamp_iou_reward": 0.9747673869132996,
"rewards/forensics_tempsamp_iou_reward": 0.6028891801834106,
"step": 485
},
{
"completion_length": 19.71875,
"epoch": 2.0334728033472804,
"grad_norm": 5.181811809539795,
"kl": 0.12451171875,
"learning_rate": 4.926778242677825e-07,
"loss": 0.005,
"reward": 0.6222259402275085,
"reward_std": 0.07642215490341187,
"reward_std_all": 0.09573835134506226,
"rewards-max/forensics_tempsamp_iou_reward": 0.9434074461460114,
"rewards/forensics_tempsamp_iou_reward": 0.4472258687019348,
"step": 486
},
{
"completion_length": 20.484375,
"epoch": 2.0376569037656904,
"grad_norm": 4.756209850311279,
"kl": 0.19970703125,
"learning_rate": 4.916317991631799e-07,
"loss": 0.008,
"reward": 0.7294941544532776,
"reward_std": 0.13824788853526115,
"reward_std_all": 0.15472250431776047,
"rewards-max/forensics_tempsamp_iou_reward": 0.9553029835224152,
"rewards/forensics_tempsamp_iou_reward": 0.5294941365718842,
"step": 487
},
{
"completion_length": 20.078125,
"epoch": 2.0418410041841004,
"grad_norm": 4.63997745513916,
"kl": 0.16357421875,
"learning_rate": 4.905857740585774e-07,
"loss": 0.0066,
"reward": 0.8495528101921082,
"reward_std": 0.118222925812006,
"reward_std_all": 0.13138620555400848,
"rewards-max/forensics_tempsamp_iou_reward": 0.9760923683643341,
"rewards/forensics_tempsamp_iou_reward": 0.5745528340339661,
"step": 488
},
{
"completion_length": 20.546875,
"epoch": 2.0460251046025104,
"grad_norm": 4.15028190612793,
"kl": 0.17138671875,
"learning_rate": 4.895397489539749e-07,
"loss": 0.0069,
"reward": 0.8010169267654419,
"reward_std": 0.10153650492429733,
"reward_std_all": 0.104568000882864,
"rewards-max/forensics_tempsamp_iou_reward": 0.9620760679244995,
"rewards/forensics_tempsamp_iou_reward": 0.5572669357061386,
"step": 489
},
{
"completion_length": 19.859375,
"epoch": 2.0502092050209204,
"grad_norm": 4.310617923736572,
"kl": 0.21728515625,
"learning_rate": 4.884937238493724e-07,
"loss": 0.0087,
"reward": 0.7907264232635498,
"reward_std": 0.09279168769717216,
"reward_std_all": 0.141476821154356,
"rewards-max/forensics_tempsamp_iou_reward": 0.8797867298126221,
"rewards/forensics_tempsamp_iou_reward": 0.509476363658905,
"step": 490
},
{
"completion_length": 20.078125,
"epoch": 2.0543933054393304,
"grad_norm": 3.994873046875,
"kl": 0.16943359375,
"learning_rate": 4.874476987447699e-07,
"loss": 0.0068,
"reward": 0.704792857170105,
"reward_std": 0.1001114621758461,
"reward_std_all": 0.10915351659059525,
"rewards-max/forensics_tempsamp_iou_reward": 0.854892760515213,
"rewards/forensics_tempsamp_iou_reward": 0.4610428661108017,
"step": 491
},
{
"completion_length": 20.5625,
"epoch": 2.0585774058577404,
"grad_norm": 11.48891830444336,
"kl": 0.1533203125,
"learning_rate": 4.864016736401673e-07,
"loss": 0.0061,
"reward": 0.9339205324649811,
"reward_std": 0.10262962430715561,
"reward_std_all": 0.09757392108440399,
"rewards-max/forensics_tempsamp_iou_reward": 0.9808842837810516,
"rewards/forensics_tempsamp_iou_reward": 0.6589204668998718,
"step": 492
},
{
"completion_length": 19.984375,
"epoch": 2.062761506276151,
"grad_norm": 3.5709593296051025,
"kl": 0.17138671875,
"learning_rate": 4.853556485355648e-07,
"loss": 0.0069,
"reward": 0.9610295295715332,
"reward_std": 0.09938562661409378,
"reward_std_all": 0.1348491907119751,
"rewards-max/forensics_tempsamp_iou_reward": 0.9805526733398438,
"rewards/forensics_tempsamp_iou_reward": 0.6610295474529266,
"step": 493
},
{
"completion_length": 20.09375,
"epoch": 2.066945606694561,
"grad_norm": 4.93778133392334,
"kl": 0.160400390625,
"learning_rate": 4.843096234309624e-07,
"loss": 0.0064,
"reward": 0.9590050578117371,
"reward_std": 0.08286719024181366,
"reward_std_all": 0.08054666593670845,
"rewards-max/forensics_tempsamp_iou_reward": 0.9338242411613464,
"rewards/forensics_tempsamp_iou_reward": 0.6590049862861633,
"step": 494
},
{
"completion_length": 19.265625,
"epoch": 2.071129707112971,
"grad_norm": 4.979303359985352,
"kl": 0.21923828125,
"learning_rate": 4.832635983263599e-07,
"loss": 0.0088,
"reward": 0.8057624995708466,
"reward_std": 0.12773869931697845,
"reward_std_all": 0.1482546031475067,
"rewards-max/forensics_tempsamp_iou_reward": 0.9423358738422394,
"rewards/forensics_tempsamp_iou_reward": 0.5432625114917755,
"step": 495
},
{
"completion_length": 20.453125,
"epoch": 2.075313807531381,
"grad_norm": 3.180353879928589,
"kl": 0.17041015625,
"learning_rate": 4.822175732217573e-07,
"loss": 0.0068,
"reward": 0.8550508320331573,
"reward_std": 0.04423682391643524,
"reward_std_all": 0.06391364522278309,
"rewards-max/forensics_tempsamp_iou_reward": 0.9554106891155243,
"rewards/forensics_tempsamp_iou_reward": 0.6113007664680481,
"step": 496
},
{
"completion_length": 19.4375,
"epoch": 2.079497907949791,
"grad_norm": 3.783013343811035,
"kl": 0.16455078125,
"learning_rate": 4.811715481171548e-07,
"loss": 0.0066,
"reward": 0.6608845293521881,
"reward_std": 0.11432984471321106,
"reward_std_all": 0.10176334530115128,
"rewards-max/forensics_tempsamp_iou_reward": 0.9248557388782501,
"rewards/forensics_tempsamp_iou_reward": 0.4483845382928848,
"step": 497
},
{
"completion_length": 19.125,
"epoch": 2.083682008368201,
"grad_norm": 4.897608757019043,
"kl": 0.17724609375,
"learning_rate": 4.801255230125523e-07,
"loss": 0.0071,
"reward": 0.8219216465950012,
"reward_std": 0.07148072868585587,
"reward_std_all": 0.07528561912477016,
"rewards-max/forensics_tempsamp_iou_reward": 0.9565538763999939,
"rewards/forensics_tempsamp_iou_reward": 0.5656716525554657,
"step": 498
},
{
"completion_length": 19.75,
"epoch": 2.087866108786611,
"grad_norm": 5.080911159515381,
"kl": 0.18408203125,
"learning_rate": 4.790794979079498e-07,
"loss": 0.0074,
"reward": 0.8727889657020569,
"reward_std": 0.09680312871932983,
"reward_std_all": 0.09809565171599388,
"rewards-max/forensics_tempsamp_iou_reward": 0.9712503552436829,
"rewards/forensics_tempsamp_iou_reward": 0.616538941860199,
"step": 499
},
{
"completion_length": 19.96875,
"epoch": 2.092050209205021,
"grad_norm": 4.536181926727295,
"kl": 0.2255859375,
"learning_rate": 4.780334728033473e-07,
"loss": 0.009,
"reward": 0.8211414515972137,
"reward_std": 0.11489741504192352,
"reward_std_all": 0.12471882253885269,
"rewards-max/forensics_tempsamp_iou_reward": 0.9239851236343384,
"rewards/forensics_tempsamp_iou_reward": 0.5461414158344269,
"step": 500
},
{
"completion_length": 19.75,
"epoch": 2.096234309623431,
"grad_norm": 4.730930328369141,
"kl": 0.16552734375,
"learning_rate": 4.769874476987447e-07,
"loss": 0.0066,
"reward": 0.7812789678573608,
"reward_std": 0.0760280042886734,
"reward_std_all": 0.08675938472151756,
"rewards-max/forensics_tempsamp_iou_reward": 0.90865558385849,
"rewards/forensics_tempsamp_iou_reward": 0.5625289678573608,
"step": 501
},
{
"completion_length": 19.5625,
"epoch": 2.100418410041841,
"grad_norm": 5.886802673339844,
"kl": 0.17626953125,
"learning_rate": 4.759414225941422e-07,
"loss": 0.007,
"reward": 0.6637192666530609,
"reward_std": 0.1377096027135849,
"reward_std_all": 0.14105181396007538,
"rewards-max/forensics_tempsamp_iou_reward": 0.9203345775604248,
"rewards/forensics_tempsamp_iou_reward": 0.4449692517518997,
"step": 502
},
{
"completion_length": 20.171875,
"epoch": 2.104602510460251,
"grad_norm": 5.05579137802124,
"kl": 0.2275390625,
"learning_rate": 4.748953974895397e-07,
"loss": 0.0091,
"reward": 0.6800119876861572,
"reward_std": 0.10433380678296089,
"reward_std_all": 0.11354615539312363,
"rewards-max/forensics_tempsamp_iou_reward": 0.85012087225914,
"rewards/forensics_tempsamp_iou_reward": 0.46751195192337036,
"step": 503
},
{
"completion_length": 20.734375,
"epoch": 2.1087866108786613,
"grad_norm": 6.506906986236572,
"kl": 0.203125,
"learning_rate": 4.738493723849372e-07,
"loss": 0.0081,
"reward": 0.7377215027809143,
"reward_std": 0.11270886287093163,
"reward_std_all": 0.1237868070602417,
"rewards-max/forensics_tempsamp_iou_reward": 0.8918470442295074,
"rewards/forensics_tempsamp_iou_reward": 0.5002214759588242,
"step": 504
},
{
"completion_length": 19.421875,
"epoch": 2.1129707112970713,
"grad_norm": 5.411094665527344,
"kl": 0.20361328125,
"learning_rate": 4.7280334728033473e-07,
"loss": 0.0081,
"reward": 0.6707089841365814,
"reward_std": 0.09282658621668816,
"reward_std_all": 0.09044953808188438,
"rewards-max/forensics_tempsamp_iou_reward": 0.9527122378349304,
"rewards/forensics_tempsamp_iou_reward": 0.4769589602947235,
"step": 505
},
{
"completion_length": 19.46875,
"epoch": 2.1171548117154813,
"grad_norm": 11.452067375183105,
"kl": 0.16650390625,
"learning_rate": 4.7175732217573224e-07,
"loss": 0.0067,
"reward": 0.626579612493515,
"reward_std": 0.11198870092630386,
"reward_std_all": 0.11675965413451195,
"rewards-max/forensics_tempsamp_iou_reward": 0.8809859156608582,
"rewards/forensics_tempsamp_iou_reward": 0.4140796363353729,
"step": 506
},
{
"completion_length": 20.71875,
"epoch": 2.1213389121338913,
"grad_norm": 5.145151615142822,
"kl": 0.20751953125,
"learning_rate": 4.7071129707112965e-07,
"loss": 0.0083,
"reward": 0.802668422460556,
"reward_std": 0.05302160978317261,
"reward_std_all": 0.08912575244903564,
"rewards-max/forensics_tempsamp_iou_reward": 0.9275757670402527,
"rewards/forensics_tempsamp_iou_reward": 0.5589184761047363,
"step": 507
},
{
"completion_length": 19.46875,
"epoch": 2.1255230125523012,
"grad_norm": 6.483147621154785,
"kl": 0.154541015625,
"learning_rate": 4.6966527196652716e-07,
"loss": 0.0062,
"reward": 0.5916315913200378,
"reward_std": 0.052650026977062225,
"reward_std_all": 0.051696885377168655,
"rewards-max/forensics_tempsamp_iou_reward": 0.9478002488613129,
"rewards/forensics_tempsamp_iou_reward": 0.40413157641887665,
"step": 508
},
{
"completion_length": 20.203125,
"epoch": 2.1297071129707112,
"grad_norm": 5.414334774017334,
"kl": 0.19873046875,
"learning_rate": 4.686192468619247e-07,
"loss": 0.008,
"reward": 0.7742977440357208,
"reward_std": 0.08255920931696892,
"reward_std_all": 0.11936867237091064,
"rewards-max/forensics_tempsamp_iou_reward": 0.9094044864177704,
"rewards/forensics_tempsamp_iou_reward": 0.530547708272934,
"step": 509
},
{
"completion_length": 19.734375,
"epoch": 2.1338912133891212,
"grad_norm": 4.198781490325928,
"kl": 0.1689453125,
"learning_rate": 4.675732217573222e-07,
"loss": 0.0068,
"reward": 0.7340344786643982,
"reward_std": 0.10258784145116806,
"reward_std_all": 0.10217270255088806,
"rewards-max/forensics_tempsamp_iou_reward": 0.9034367799758911,
"rewards/forensics_tempsamp_iou_reward": 0.5090344250202179,
"step": 510
},
{
"completion_length": 20.015625,
"epoch": 2.1380753138075312,
"grad_norm": 8.994231224060059,
"kl": 0.1591796875,
"learning_rate": 4.6652719665271965e-07,
"loss": 0.0064,
"reward": 0.9149849712848663,
"reward_std": 0.06796960160136223,
"reward_std_all": 0.07320309989154339,
"rewards-max/forensics_tempsamp_iou_reward": 0.9887359738349915,
"rewards/forensics_tempsamp_iou_reward": 0.6337349712848663,
"step": 511
},
{
"completion_length": 19.5625,
"epoch": 2.1422594142259412,
"grad_norm": 11.76638126373291,
"kl": 0.21826171875,
"learning_rate": 4.654811715481171e-07,
"loss": 0.0087,
"reward": 0.8072271943092346,
"reward_std": 0.11938677355647087,
"reward_std_all": 0.1098562628030777,
"rewards-max/forensics_tempsamp_iou_reward": 0.9847364723682404,
"rewards/forensics_tempsamp_iou_reward": 0.557227224111557,
"step": 512
},
{
"completion_length": 20.9375,
"epoch": 2.1464435146443517,
"grad_norm": 4.416269779205322,
"kl": 0.205078125,
"learning_rate": 4.644351464435146e-07,
"loss": 0.0082,
"reward": 0.7171244323253632,
"reward_std": 0.0624085608869791,
"reward_std_all": 0.09868739545345306,
"rewards-max/forensics_tempsamp_iou_reward": 0.8991140723228455,
"rewards/forensics_tempsamp_iou_reward": 0.4858744591474533,
"step": 513
},
{
"completion_length": 20.890625,
"epoch": 2.1506276150627617,
"grad_norm": 5.458975315093994,
"kl": 0.20361328125,
"learning_rate": 4.6338912133891213e-07,
"loss": 0.0082,
"reward": 0.723075807094574,
"reward_std": 0.11257071793079376,
"reward_std_all": 0.11600349098443985,
"rewards-max/forensics_tempsamp_iou_reward": 0.8371534645557404,
"rewards/forensics_tempsamp_iou_reward": 0.46682579815387726,
"step": 514
},
{
"completion_length": 19.375,
"epoch": 2.1548117154811717,
"grad_norm": 6.706021308898926,
"kl": 0.29541015625,
"learning_rate": 4.623430962343096e-07,
"loss": 0.0118,
"reward": 0.7678598463535309,
"reward_std": 0.13144999742507935,
"reward_std_all": 0.1357121393084526,
"rewards-max/forensics_tempsamp_iou_reward": 0.9510038197040558,
"rewards/forensics_tempsamp_iou_reward": 0.524109810590744,
"step": 515
},
{
"completion_length": 20.1875,
"epoch": 2.1589958158995817,
"grad_norm": 7.274849891662598,
"kl": 0.20458984375,
"learning_rate": 4.612970711297071e-07,
"loss": 0.0082,
"reward": 0.719489723443985,
"reward_std": 0.09699538722634315,
"reward_std_all": 0.10317607782781124,
"rewards-max/forensics_tempsamp_iou_reward": 0.9032253324985504,
"rewards/forensics_tempsamp_iou_reward": 0.4694896936416626,
"step": 516
},
{
"completion_length": 19.34375,
"epoch": 2.1631799163179917,
"grad_norm": 5.715125560760498,
"kl": 0.384765625,
"learning_rate": 4.6025104602510457e-07,
"loss": 0.0154,
"reward": 0.8385485112667084,
"reward_std": 0.05112417973577976,
"reward_std_all": 0.06385861337184906,
"rewards-max/forensics_tempsamp_iou_reward": 0.9944179654121399,
"rewards/forensics_tempsamp_iou_reward": 0.5760485827922821,
"step": 517
},
{
"completion_length": 19.796875,
"epoch": 2.1673640167364017,
"grad_norm": 4.573932647705078,
"kl": 0.24609375,
"learning_rate": 4.592050209205021e-07,
"loss": 0.0099,
"reward": 0.6209633052349091,
"reward_std": 0.09164205193519592,
"reward_std_all": 0.08499329537153244,
"rewards-max/forensics_tempsamp_iou_reward": 0.8839658200740814,
"rewards/forensics_tempsamp_iou_reward": 0.4147133082151413,
"step": 518
},
{
"completion_length": 19.109375,
"epoch": 2.1715481171548117,
"grad_norm": 3.72192645072937,
"kl": 0.193359375,
"learning_rate": 4.5815899581589954e-07,
"loss": 0.0077,
"reward": 0.8566508591175079,
"reward_std": 0.10162849351763725,
"reward_std_all": 0.10203078016638756,
"rewards-max/forensics_tempsamp_iou_reward": 0.9631470143795013,
"rewards/forensics_tempsamp_iou_reward": 0.5816508233547211,
"step": 519
},
{
"completion_length": 19.03125,
"epoch": 2.1757322175732217,
"grad_norm": 5.132172107696533,
"kl": 0.14404296875,
"learning_rate": 4.5711297071129705e-07,
"loss": 0.0058,
"reward": 0.808495283126831,
"reward_std": 0.050763217732310295,
"reward_std_all": 0.05160978622734547,
"rewards-max/forensics_tempsamp_iou_reward": 0.9806326627731323,
"rewards/forensics_tempsamp_iou_reward": 0.5834953188896179,
"step": 520
},
{
"completion_length": 19.921875,
"epoch": 2.1799163179916317,
"grad_norm": 5.0079755783081055,
"kl": 0.1982421875,
"learning_rate": 4.5606694560669456e-07,
"loss": 0.0079,
"reward": 0.7725296318531036,
"reward_std": 0.12858223915100098,
"reward_std_all": 0.11155817657709122,
"rewards-max/forensics_tempsamp_iou_reward": 0.9341751337051392,
"rewards/forensics_tempsamp_iou_reward": 0.5350296050310135,
"step": 521
},
{
"completion_length": 19.40625,
"epoch": 2.1841004184100417,
"grad_norm": 5.127690315246582,
"kl": 0.22265625,
"learning_rate": 4.55020920502092e-07,
"loss": 0.0089,
"reward": 0.6189653277397156,
"reward_std": 0.14874126017093658,
"reward_std_all": 0.1357589103281498,
"rewards-max/forensics_tempsamp_iou_reward": 0.9838048815727234,
"rewards/forensics_tempsamp_iou_reward": 0.42521530389785767,
"step": 522
},
{
"completion_length": 20.125,
"epoch": 2.1882845188284517,
"grad_norm": 4.704853534698486,
"kl": 0.20068359375,
"learning_rate": 4.539748953974895e-07,
"loss": 0.008,
"reward": 0.7245790064334869,
"reward_std": 0.10194584354758263,
"reward_std_all": 0.09717009589076042,
"rewards-max/forensics_tempsamp_iou_reward": 0.9112019836902618,
"rewards/forensics_tempsamp_iou_reward": 0.46832896769046783,
"step": 523
},
{
"completion_length": 19.625,
"epoch": 2.192468619246862,
"grad_norm": 5.7402496337890625,
"kl": 0.19970703125,
"learning_rate": 4.52928870292887e-07,
"loss": 0.008,
"reward": 0.6898003220558167,
"reward_std": 0.08877832442522049,
"reward_std_all": 0.08128324151039124,
"rewards-max/forensics_tempsamp_iou_reward": 0.9284935891628265,
"rewards/forensics_tempsamp_iou_reward": 0.4523002952337265,
"step": 524
},
{
"completion_length": 19.859375,
"epoch": 2.196652719665272,
"grad_norm": 4.778491973876953,
"kl": 0.16357421875,
"learning_rate": 4.518828451882845e-07,
"loss": 0.0065,
"reward": 0.8065979778766632,
"reward_std": 0.08821172080934048,
"reward_std_all": 0.09869246929883957,
"rewards-max/forensics_tempsamp_iou_reward": 0.9445652067661285,
"rewards/forensics_tempsamp_iou_reward": 0.5315980315208435,
"step": 525
},
{
"completion_length": 19.765625,
"epoch": 2.200836820083682,
"grad_norm": 5.042792320251465,
"kl": 0.2236328125,
"learning_rate": 4.50836820083682e-07,
"loss": 0.0089,
"reward": 0.8504658937454224,
"reward_std": 0.12204217910766602,
"reward_std_all": 0.11429278180003166,
"rewards-max/forensics_tempsamp_iou_reward": 0.9473364353179932,
"rewards/forensics_tempsamp_iou_reward": 0.5567159056663513,
"step": 526
},
{
"completion_length": 19.65625,
"epoch": 2.205020920502092,
"grad_norm": 6.5350518226623535,
"kl": 0.189453125,
"learning_rate": 4.497907949790795e-07,
"loss": 0.0076,
"reward": 0.787993997335434,
"reward_std": 0.09967102110385895,
"reward_std_all": 0.1358715258538723,
"rewards-max/forensics_tempsamp_iou_reward": 0.9469552934169769,
"rewards/forensics_tempsamp_iou_reward": 0.5129939615726471,
"step": 527
},
{
"completion_length": 19.5625,
"epoch": 2.209205020920502,
"grad_norm": 7.786820411682129,
"kl": 0.2080078125,
"learning_rate": 4.4874476987447694e-07,
"loss": 0.0083,
"reward": 0.6216503381729126,
"reward_std": 0.10071680322289467,
"reward_std_all": 0.11645929142832756,
"rewards-max/forensics_tempsamp_iou_reward": 0.8982471525669098,
"rewards/forensics_tempsamp_iou_reward": 0.4279002845287323,
"step": 528
},
{
"completion_length": 20.296875,
"epoch": 2.213389121338912,
"grad_norm": 14.009378433227539,
"kl": 0.2353515625,
"learning_rate": 4.4769874476987445e-07,
"loss": 0.0094,
"reward": 0.8867015242576599,
"reward_std": 0.11796678230166435,
"reward_std_all": 0.13490141928195953,
"rewards-max/forensics_tempsamp_iou_reward": 0.9903475046157837,
"rewards/forensics_tempsamp_iou_reward": 0.5992015600204468,
"step": 529
},
{
"completion_length": 20.0625,
"epoch": 2.217573221757322,
"grad_norm": 5.874517917633057,
"kl": 0.154541015625,
"learning_rate": 4.4665271966527197e-07,
"loss": 0.0062,
"reward": 1.0525118708610535,
"reward_std": 0.09618723765015602,
"reward_std_all": 0.08980529755353928,
"rewards-max/forensics_tempsamp_iou_reward": 0.9778302013874054,
"rewards/forensics_tempsamp_iou_reward": 0.7525118291378021,
"step": 530
},
{
"completion_length": 19.84375,
"epoch": 2.221757322175732,
"grad_norm": 3.768432855606079,
"kl": 0.23779296875,
"learning_rate": 4.456066945606695e-07,
"loss": 0.0095,
"reward": 0.8768711686134338,
"reward_std": 0.08216530457139015,
"reward_std_all": 0.07300118170678616,
"rewards-max/forensics_tempsamp_iou_reward": 0.8720933198928833,
"rewards/forensics_tempsamp_iou_reward": 0.5768711566925049,
"step": 531
},
{
"completion_length": 19.875,
"epoch": 2.225941422594142,
"grad_norm": 5.778804779052734,
"kl": 0.20556640625,
"learning_rate": 4.445606694560669e-07,
"loss": 0.0082,
"reward": 0.7191672921180725,
"reward_std": 0.09999847412109375,
"reward_std_all": 0.10893122851848602,
"rewards-max/forensics_tempsamp_iou_reward": 0.925102710723877,
"rewards/forensics_tempsamp_iou_reward": 0.4691672623157501,
"step": 532
},
{
"completion_length": 20.4375,
"epoch": 2.2301255230125525,
"grad_norm": 5.832368850708008,
"kl": 0.3974609375,
"learning_rate": 4.435146443514644e-07,
"loss": 0.0159,
"reward": 0.7466718554496765,
"reward_std": 0.12515771389007568,
"reward_std_all": 0.15022336691617966,
"rewards-max/forensics_tempsamp_iou_reward": 0.892082542181015,
"rewards/forensics_tempsamp_iou_reward": 0.5091718137264252,
"step": 533
},
{
"completion_length": 18.96875,
"epoch": 2.2343096234309625,
"grad_norm": 7.27992057800293,
"kl": 0.23291015625,
"learning_rate": 4.424686192468619e-07,
"loss": 0.0093,
"reward": 0.7808465957641602,
"reward_std": 0.1326676681637764,
"reward_std_all": 0.12548920884728432,
"rewards-max/forensics_tempsamp_iou_reward": 0.940140575170517,
"rewards/forensics_tempsamp_iou_reward": 0.5308466255664825,
"step": 534
},
{
"completion_length": 19.65625,
"epoch": 2.2384937238493725,
"grad_norm": 6.759598255157471,
"kl": 0.1787109375,
"learning_rate": 4.414225941422594e-07,
"loss": 0.0072,
"reward": 0.7316485941410065,
"reward_std": 0.08184535708278418,
"reward_std_all": 0.10136428102850914,
"rewards-max/forensics_tempsamp_iou_reward": 0.9100067317485809,
"rewards/forensics_tempsamp_iou_reward": 0.5128985345363617,
"step": 535
},
{
"completion_length": 19.296875,
"epoch": 2.2426778242677825,
"grad_norm": 5.016279220581055,
"kl": 0.4345703125,
"learning_rate": 4.403765690376569e-07,
"loss": 0.0174,
"reward": 0.9253139495849609,
"reward_std": 0.08172677457332611,
"reward_std_all": 0.08561087772250175,
"rewards-max/forensics_tempsamp_iou_reward": 0.9447246193885803,
"rewards/forensics_tempsamp_iou_reward": 0.625313937664032,
"step": 536
},
{
"completion_length": 19.296875,
"epoch": 2.2468619246861925,
"grad_norm": 5.523147106170654,
"kl": 0.29638671875,
"learning_rate": 4.3933054393305435e-07,
"loss": 0.0118,
"reward": 0.7101505696773529,
"reward_std": 0.10702027752995491,
"reward_std_all": 0.10659291967749596,
"rewards-max/forensics_tempsamp_iou_reward": 0.9773317575454712,
"rewards/forensics_tempsamp_iou_reward": 0.47890058159828186,
"step": 537
},
{
"completion_length": 19.953125,
"epoch": 2.2510460251046025,
"grad_norm": 5.156852722167969,
"kl": 0.14794921875,
"learning_rate": 4.3828451882845186e-07,
"loss": 0.0059,
"reward": 0.7346127927303314,
"reward_std": 0.06807839870452881,
"reward_std_all": 0.06359209306538105,
"rewards-max/forensics_tempsamp_iou_reward": 0.9117133021354675,
"rewards/forensics_tempsamp_iou_reward": 0.5096127539873123,
"step": 538
},
{
"completion_length": 20.125,
"epoch": 2.2552301255230125,
"grad_norm": 5.17818546295166,
"kl": 0.25048828125,
"learning_rate": 4.3723849372384937e-07,
"loss": 0.01,
"reward": 0.7027948796749115,
"reward_std": 0.05474289506673813,
"reward_std_all": 0.08011150360107422,
"rewards-max/forensics_tempsamp_iou_reward": 0.8666802048683167,
"rewards/forensics_tempsamp_iou_reward": 0.47779491543769836,
"step": 539
},
{
"completion_length": 20.40625,
"epoch": 2.2594142259414225,
"grad_norm": 5.241072654724121,
"kl": 0.197265625,
"learning_rate": 4.3619246861924683e-07,
"loss": 0.0079,
"reward": 0.7040524482727051,
"reward_std": 0.12208626419305801,
"reward_std_all": 0.11887601017951965,
"rewards-max/forensics_tempsamp_iou_reward": 0.939929723739624,
"rewards/forensics_tempsamp_iou_reward": 0.47905245423316956,
"step": 540
},
{
"completion_length": 19.328125,
"epoch": 2.2635983263598325,
"grad_norm": 6.168412685394287,
"kl": 0.169921875,
"learning_rate": 4.3514644351464434e-07,
"loss": 0.0068,
"reward": 0.7878838777542114,
"reward_std": 0.10685522109270096,
"reward_std_all": 0.10557439178228378,
"rewards-max/forensics_tempsamp_iou_reward": 0.9014396965503693,
"rewards/forensics_tempsamp_iou_reward": 0.5503838360309601,
"step": 541
},
{
"completion_length": 20.328125,
"epoch": 2.2677824267782425,
"grad_norm": 4.762511253356934,
"kl": 0.21923828125,
"learning_rate": 4.341004184100418e-07,
"loss": 0.0088,
"reward": 0.8740364909172058,
"reward_std": 0.08191834762692451,
"reward_std_all": 0.08230091258883476,
"rewards-max/forensics_tempsamp_iou_reward": 0.9185419082641602,
"rewards/forensics_tempsamp_iou_reward": 0.6177864670753479,
"step": 542
},
{
"completion_length": 19.59375,
"epoch": 2.2719665271966525,
"grad_norm": 6.142400741577148,
"kl": 0.17333984375,
"learning_rate": 4.330543933054393e-07,
"loss": 0.0069,
"reward": 0.8098754286766052,
"reward_std": 0.08240043744444847,
"reward_std_all": 0.11004289239645004,
"rewards-max/forensics_tempsamp_iou_reward": 0.9959999918937683,
"rewards/forensics_tempsamp_iou_reward": 0.5661254972219467,
"step": 543
},
{
"completion_length": 19.59375,
"epoch": 2.276150627615063,
"grad_norm": 4.876639366149902,
"kl": 0.21240234375,
"learning_rate": 4.3200836820083683e-07,
"loss": 0.0085,
"reward": 0.864603191614151,
"reward_std": 0.07770978659391403,
"reward_std_all": 0.08939583413302898,
"rewards-max/forensics_tempsamp_iou_reward": 0.9530367255210876,
"rewards/forensics_tempsamp_iou_reward": 0.5896031260490417,
"step": 544
},
{
"completion_length": 19.578125,
"epoch": 2.280334728033473,
"grad_norm": 5.268033027648926,
"kl": 0.1796875,
"learning_rate": 4.309623430962343e-07,
"loss": 0.0072,
"reward": 0.8277930617332458,
"reward_std": 0.1291949711740017,
"reward_std_all": 0.15010645985603333,
"rewards-max/forensics_tempsamp_iou_reward": 0.9294543862342834,
"rewards/forensics_tempsamp_iou_reward": 0.5715430378913879,
"step": 545
},
{
"completion_length": 19.78125,
"epoch": 2.284518828451883,
"grad_norm": 5.861181259155273,
"kl": 0.203125,
"learning_rate": 4.299163179916318e-07,
"loss": 0.0081,
"reward": 0.7808391749858856,
"reward_std": 0.1460546851158142,
"reward_std_all": 0.129144087433815,
"rewards-max/forensics_tempsamp_iou_reward": 0.9608834981918335,
"rewards/forensics_tempsamp_iou_reward": 0.5245891511440277,
"step": 546
},
{
"completion_length": 19.796875,
"epoch": 2.288702928870293,
"grad_norm": 3.7409873008728027,
"kl": 0.248046875,
"learning_rate": 4.2887029288702926e-07,
"loss": 0.0099,
"reward": 0.6471245884895325,
"reward_std": 0.10059045627713203,
"reward_std_all": 0.0955934226512909,
"rewards-max/forensics_tempsamp_iou_reward": 0.8655905723571777,
"rewards/forensics_tempsamp_iou_reward": 0.42212456464767456,
"step": 547
},
{
"completion_length": 19.609375,
"epoch": 2.292887029288703,
"grad_norm": 4.890139579772949,
"kl": 0.2294921875,
"learning_rate": 4.278242677824268e-07,
"loss": 0.0092,
"reward": 0.885291576385498,
"reward_std": 0.07640947215259075,
"reward_std_all": 0.08268946036696434,
"rewards-max/forensics_tempsamp_iou_reward": 0.9510392546653748,
"rewards/forensics_tempsamp_iou_reward": 0.6102915704250336,
"step": 548
},
{
"completion_length": 20.953125,
"epoch": 2.297071129707113,
"grad_norm": 3.94885516166687,
"kl": 0.2294921875,
"learning_rate": 4.2677824267782424e-07,
"loss": 0.0092,
"reward": 0.8409311473369598,
"reward_std": 0.07836669310927391,
"reward_std_all": 0.0918048806488514,
"rewards-max/forensics_tempsamp_iou_reward": 0.890780121088028,
"rewards/forensics_tempsamp_iou_reward": 0.5471811294555664,
"step": 549
},
{
"completion_length": 20.046875,
"epoch": 2.301255230125523,
"grad_norm": 4.000836372375488,
"kl": 0.22314453125,
"learning_rate": 4.2573221757322175e-07,
"loss": 0.0089,
"reward": 0.7646938562393188,
"reward_std": 0.11859513074159622,
"reward_std_all": 0.13401080667972565,
"rewards-max/forensics_tempsamp_iou_reward": 0.9495051205158234,
"rewards/forensics_tempsamp_iou_reward": 0.533443883061409,
"step": 550
},
{
"completion_length": 20.328125,
"epoch": 2.305439330543933,
"grad_norm": 3.825608253479004,
"kl": 0.22119140625,
"learning_rate": 4.2468619246861926e-07,
"loss": 0.0088,
"reward": 0.6713233888149261,
"reward_std": 0.11670936644077301,
"reward_std_all": 0.10766930133104324,
"rewards-max/forensics_tempsamp_iou_reward": 0.8399813771247864,
"rewards/forensics_tempsamp_iou_reward": 0.4400733411312103,
"step": 551
},
{
"completion_length": 19.59375,
"epoch": 2.309623430962343,
"grad_norm": 4.409998416900635,
"kl": 0.15185546875,
"learning_rate": 4.236401673640167e-07,
"loss": 0.0061,
"reward": 0.8542134165763855,
"reward_std": 0.08677715063095093,
"reward_std_all": 0.09328364580869675,
"rewards-max/forensics_tempsamp_iou_reward": 0.9046052694320679,
"rewards/forensics_tempsamp_iou_reward": 0.5542134046554565,
"step": 552
},
{
"completion_length": 19.46875,
"epoch": 2.3138075313807533,
"grad_norm": 5.811684608459473,
"kl": 0.1416015625,
"learning_rate": 4.225941422594142e-07,
"loss": 0.0057,
"reward": 0.8311557173728943,
"reward_std": 0.06526858173310757,
"reward_std_all": 0.06528631411492825,
"rewards-max/forensics_tempsamp_iou_reward": 0.9306145012378693,
"rewards/forensics_tempsamp_iou_reward": 0.5499057173728943,
"step": 553
},
{
"completion_length": 19.53125,
"epoch": 2.3179916317991633,
"grad_norm": 7.476200580596924,
"kl": 0.49560546875,
"learning_rate": 4.215481171548117e-07,
"loss": 0.0198,
"reward": 0.7298004925251007,
"reward_std": 0.09055406227707863,
"reward_std_all": 0.10191579535603523,
"rewards-max/forensics_tempsamp_iou_reward": 0.94111368060112,
"rewards/forensics_tempsamp_iou_reward": 0.4735504984855652,
"step": 554
},
{
"completion_length": 19.90625,
"epoch": 2.3221757322175733,
"grad_norm": 5.128610134124756,
"kl": 0.149169921875,
"learning_rate": 4.205020920502092e-07,
"loss": 0.006,
"reward": 0.812649130821228,
"reward_std": 0.06565347127616405,
"reward_std_all": 0.08100682124495506,
"rewards-max/forensics_tempsamp_iou_reward": 0.9841265678405762,
"rewards/forensics_tempsamp_iou_reward": 0.5688991248607635,
"step": 555
},
{
"completion_length": 21.359375,
"epoch": 2.3263598326359833,
"grad_norm": 6.845063209533691,
"kl": 0.20751953125,
"learning_rate": 4.194560669456067e-07,
"loss": 0.0083,
"reward": 0.7230862975120544,
"reward_std": 0.1262136697769165,
"reward_std_all": 0.1669256016612053,
"rewards-max/forensics_tempsamp_iou_reward": 0.84506556391716,
"rewards/forensics_tempsamp_iou_reward": 0.4543362557888031,
"step": 556
},
{
"completion_length": 19.3125,
"epoch": 2.3305439330543933,
"grad_norm": 6.990042686462402,
"kl": 0.15673828125,
"learning_rate": 4.184100418410041e-07,
"loss": 0.0063,
"reward": 0.8625707030296326,
"reward_std": 0.05226464197039604,
"reward_std_all": 0.05368735454976559,
"rewards-max/forensics_tempsamp_iou_reward": 0.9334782958030701,
"rewards/forensics_tempsamp_iou_reward": 0.5625706613063812,
"step": 557
},
{
"completion_length": 20.265625,
"epoch": 2.3347280334728033,
"grad_norm": 6.682015895843506,
"kl": 0.2109375,
"learning_rate": 4.1736401673640164e-07,
"loss": 0.0084,
"reward": 0.8650551736354828,
"reward_std": 0.11513014137744904,
"reward_std_all": 0.11226258054375648,
"rewards-max/forensics_tempsamp_iou_reward": 0.9742611348628998,
"rewards/forensics_tempsamp_iou_reward": 0.5963051915168762,
"step": 558
},
{
"completion_length": 19.671875,
"epoch": 2.3389121338912133,
"grad_norm": 8.14515209197998,
"kl": 0.15869140625,
"learning_rate": 4.1631799163179915e-07,
"loss": 0.0064,
"reward": 0.832645833492279,
"reward_std": 0.17803804948925972,
"reward_std_all": 0.16703474521636963,
"rewards-max/forensics_tempsamp_iou_reward": 0.9801188707351685,
"rewards/forensics_tempsamp_iou_reward": 0.582645833492279,
"step": 559
},
{
"completion_length": 19.34375,
"epoch": 2.3430962343096233,
"grad_norm": 5.212371826171875,
"kl": 0.20263671875,
"learning_rate": 4.1527196652719666e-07,
"loss": 0.0081,
"reward": 0.9322017729282379,
"reward_std": 0.06757988967001438,
"reward_std_all": 0.08183667063713074,
"rewards-max/forensics_tempsamp_iou_reward": 0.8882275819778442,
"rewards/forensics_tempsamp_iou_reward": 0.6322017908096313,
"step": 560
},
{
"completion_length": 19.40625,
"epoch": 2.3472803347280333,
"grad_norm": 3.9458134174346924,
"kl": 0.17041015625,
"learning_rate": 4.142259414225941e-07,
"loss": 0.0068,
"reward": 0.7893003523349762,
"reward_std": 0.04496923554688692,
"reward_std_all": 0.05517719127237797,
"rewards-max/forensics_tempsamp_iou_reward": 0.9378100335597992,
"rewards/forensics_tempsamp_iou_reward": 0.5268003046512604,
"step": 561
},
{
"completion_length": 19.921875,
"epoch": 2.3514644351464433,
"grad_norm": 3.9698948860168457,
"kl": 0.15185546875,
"learning_rate": 4.131799163179916e-07,
"loss": 0.0061,
"reward": 0.7105948626995087,
"reward_std": 0.057975493371486664,
"reward_std_all": 0.07538148015737534,
"rewards-max/forensics_tempsamp_iou_reward": 0.8888473212718964,
"rewards/forensics_tempsamp_iou_reward": 0.48559486865997314,
"step": 562
},
{
"completion_length": 19.296875,
"epoch": 2.3556485355648533,
"grad_norm": 5.163781642913818,
"kl": 0.1396484375,
"learning_rate": 4.121338912133891e-07,
"loss": 0.0056,
"reward": 0.8613614737987518,
"reward_std": 0.10432932898402214,
"reward_std_all": 0.09365060739219189,
"rewards-max/forensics_tempsamp_iou_reward": 0.9518457651138306,
"rewards/forensics_tempsamp_iou_reward": 0.6113614737987518,
"step": 563
},
{
"completion_length": 20.90625,
"epoch": 2.3598326359832638,
"grad_norm": 4.371069431304932,
"kl": 0.287109375,
"learning_rate": 4.110878661087866e-07,
"loss": 0.0115,
"reward": 0.8573383986949921,
"reward_std": 0.09024647995829582,
"reward_std_all": 0.09084992855787277,
"rewards-max/forensics_tempsamp_iou_reward": 0.9331088066101074,
"rewards/forensics_tempsamp_iou_reward": 0.5635884404182434,
"step": 564
},
{
"completion_length": 19.421875,
"epoch": 2.3640167364016738,
"grad_norm": 4.907522201538086,
"kl": 0.390625,
"learning_rate": 4.100418410041841e-07,
"loss": 0.0157,
"reward": 0.8723192811012268,
"reward_std": 0.11656848341226578,
"reward_std_all": 0.12971869483590126,
"rewards-max/forensics_tempsamp_iou_reward": 0.9440198540687561,
"rewards/forensics_tempsamp_iou_reward": 0.5785692632198334,
"step": 565
},
{
"completion_length": 19.953125,
"epoch": 2.3682008368200838,
"grad_norm": 5.485681056976318,
"kl": 0.3095703125,
"learning_rate": 4.089958158995816e-07,
"loss": 0.0124,
"reward": 0.8542315661907196,
"reward_std": 0.08538665249943733,
"reward_std_all": 0.08191237226128578,
"rewards-max/forensics_tempsamp_iou_reward": 0.9544620513916016,
"rewards/forensics_tempsamp_iou_reward": 0.5792315304279327,
"step": 566
},
{
"completion_length": 19.53125,
"epoch": 2.3723849372384938,
"grad_norm": 4.2162089347839355,
"kl": 0.1611328125,
"learning_rate": 4.0794979079497904e-07,
"loss": 0.0064,
"reward": 0.8082980215549469,
"reward_std": 0.05821315012872219,
"reward_std_all": 0.06996244937181473,
"rewards-max/forensics_tempsamp_iou_reward": 0.9591994285583496,
"rewards/forensics_tempsamp_iou_reward": 0.5645479559898376,
"step": 567
},
{
"completion_length": 20.125,
"epoch": 2.3765690376569037,
"grad_norm": 3.7707202434539795,
"kl": 0.21484375,
"learning_rate": 4.0690376569037656e-07,
"loss": 0.0086,
"reward": 0.6380931735038757,
"reward_std": 0.1328081078827381,
"reward_std_all": 0.13952016830444336,
"rewards-max/forensics_tempsamp_iou_reward": 0.9348475933074951,
"rewards/forensics_tempsamp_iou_reward": 0.45059312880039215,
"step": 568
},
{
"completion_length": 19.59375,
"epoch": 2.3807531380753137,
"grad_norm": 5.106783866882324,
"kl": 0.162109375,
"learning_rate": 4.0585774058577407e-07,
"loss": 0.0065,
"reward": 0.8622868061065674,
"reward_std": 0.07060226425528526,
"reward_std_all": 0.06886870600283146,
"rewards-max/forensics_tempsamp_iou_reward": 0.960615485906601,
"rewards/forensics_tempsamp_iou_reward": 0.5810368210077286,
"step": 569
},
{
"completion_length": 19.609375,
"epoch": 2.3849372384937237,
"grad_norm": 4.372280597686768,
"kl": 0.2158203125,
"learning_rate": 4.0481171548117153e-07,
"loss": 0.0086,
"reward": 0.8909682035446167,
"reward_std": 0.07443519309163094,
"reward_std_all": 0.09059130772948265,
"rewards-max/forensics_tempsamp_iou_reward": 0.9438815116882324,
"rewards/forensics_tempsamp_iou_reward": 0.6159682273864746,
"step": 570
},
{
"completion_length": 20.1875,
"epoch": 2.3891213389121337,
"grad_norm": 8.104747772216797,
"kl": 0.17431640625,
"learning_rate": 4.0376569037656904e-07,
"loss": 0.007,
"reward": 0.9544511437416077,
"reward_std": 0.11320538073778152,
"reward_std_all": 0.13755165040493011,
"rewards-max/forensics_tempsamp_iou_reward": 0.9500020146369934,
"rewards/forensics_tempsamp_iou_reward": 0.6857011020183563,
"step": 571
},
{
"completion_length": 19.71875,
"epoch": 2.393305439330544,
"grad_norm": 4.293045997619629,
"kl": 0.139404296875,
"learning_rate": 4.027196652719665e-07,
"loss": 0.0056,
"reward": 0.7631777822971344,
"reward_std": 0.06937412731349468,
"reward_std_all": 0.07395971938967705,
"rewards-max/forensics_tempsamp_iou_reward": 0.9202623665332794,
"rewards/forensics_tempsamp_iou_reward": 0.49442777037620544,
"step": 572
},
{
"completion_length": 20.359375,
"epoch": 2.397489539748954,
"grad_norm": 5.100405216217041,
"kl": 0.2158203125,
"learning_rate": 4.01673640167364e-07,
"loss": 0.0086,
"reward": 0.7666550576686859,
"reward_std": 0.11047893762588501,
"reward_std_all": 0.10917370021343231,
"rewards-max/forensics_tempsamp_iou_reward": 0.9838695526123047,
"rewards/forensics_tempsamp_iou_reward": 0.5291550755500793,
"step": 573
},
{
"completion_length": 19.515625,
"epoch": 2.401673640167364,
"grad_norm": 3.7552223205566406,
"kl": 0.1982421875,
"learning_rate": 4.006276150627615e-07,
"loss": 0.0079,
"reward": 0.7181138396263123,
"reward_std": 0.09349928051233292,
"reward_std_all": 0.08834211900830269,
"rewards-max/forensics_tempsamp_iou_reward": 0.9623788595199585,
"rewards/forensics_tempsamp_iou_reward": 0.5056138038635254,
"step": 574
},
{
"completion_length": 19.4375,
"epoch": 2.405857740585774,
"grad_norm": 5.230954647064209,
"kl": 0.19384765625,
"learning_rate": 3.99581589958159e-07,
"loss": 0.0077,
"reward": 0.8990738391876221,
"reward_std": 0.13239216059446335,
"reward_std_all": 0.12016239389777184,
"rewards-max/forensics_tempsamp_iou_reward": 0.9189765751361847,
"rewards/forensics_tempsamp_iou_reward": 0.6053238213062286,
"step": 575
},
{
"completion_length": 20.5625,
"epoch": 2.410041841004184,
"grad_norm": 4.598206996917725,
"kl": 0.15673828125,
"learning_rate": 3.985355648535565e-07,
"loss": 0.0063,
"reward": 0.8858963251113892,
"reward_std": 0.08882933109998703,
"reward_std_all": 0.0925118625164032,
"rewards-max/forensics_tempsamp_iou_reward": 0.9669237434864044,
"rewards/forensics_tempsamp_iou_reward": 0.6046462953090668,
"step": 576
},
{
"completion_length": 21.15625,
"epoch": 2.414225941422594,
"grad_norm": 5.4895524978637695,
"kl": 0.19384765625,
"learning_rate": 3.9748953974895396e-07,
"loss": 0.0078,
"reward": 0.8014865219593048,
"reward_std": 0.09231236204504967,
"reward_std_all": 0.09697947651147842,
"rewards-max/forensics_tempsamp_iou_reward": 0.9042695462703705,
"rewards/forensics_tempsamp_iou_reward": 0.5452365279197693,
"step": 577
},
{
"completion_length": 19.421875,
"epoch": 2.418410041841004,
"grad_norm": 5.188279151916504,
"kl": 0.19287109375,
"learning_rate": 3.964435146443514e-07,
"loss": 0.0077,
"reward": 0.8085197806358337,
"reward_std": 0.07207491248846054,
"reward_std_all": 0.07466640323400497,
"rewards-max/forensics_tempsamp_iou_reward": 0.9679206013679504,
"rewards/forensics_tempsamp_iou_reward": 0.5647697746753693,
"step": 578
},
{
"completion_length": 19.5,
"epoch": 2.422594142259414,
"grad_norm": 4.916874885559082,
"kl": 0.30029296875,
"learning_rate": 3.9539748953974893e-07,
"loss": 0.012,
"reward": 0.682951420545578,
"reward_std": 0.08943436481058598,
"reward_std_all": 0.09052596800029278,
"rewards-max/forensics_tempsamp_iou_reward": 0.9392091631889343,
"rewards/forensics_tempsamp_iou_reward": 0.432951420545578,
"step": 579
},
{
"completion_length": 19.921875,
"epoch": 2.426778242677824,
"grad_norm": 5.267099380493164,
"kl": 0.23193359375,
"learning_rate": 3.9435146443514645e-07,
"loss": 0.0093,
"reward": 0.8710204064846039,
"reward_std": 0.09700940549373627,
"reward_std_all": 0.10794921219348907,
"rewards-max/forensics_tempsamp_iou_reward": 0.897548496723175,
"rewards/forensics_tempsamp_iou_reward": 0.596020370721817,
"step": 580
},
{
"completion_length": 19.921875,
"epoch": 2.430962343096234,
"grad_norm": 4.047446250915527,
"kl": 0.23583984375,
"learning_rate": 3.9330543933054396e-07,
"loss": 0.0094,
"reward": 0.7757426202297211,
"reward_std": 0.1113538108766079,
"reward_std_all": 0.1311514638364315,
"rewards-max/forensics_tempsamp_iou_reward": 0.9249296188354492,
"rewards/forensics_tempsamp_iou_reward": 0.5257425904273987,
"step": 581
},
{
"completion_length": 19.546875,
"epoch": 2.435146443514644,
"grad_norm": 5.110191822052002,
"kl": 0.162109375,
"learning_rate": 3.9225941422594137e-07,
"loss": 0.0065,
"reward": 0.7864111959934235,
"reward_std": 0.07453197613358498,
"reward_std_all": 0.07102247327566147,
"rewards-max/forensics_tempsamp_iou_reward": 0.9715700149536133,
"rewards/forensics_tempsamp_iou_reward": 0.5489111542701721,
"step": 582
},
{
"completion_length": 21.421875,
"epoch": 2.439330543933054,
"grad_norm": 5.129436016082764,
"kl": 0.205078125,
"learning_rate": 3.912133891213389e-07,
"loss": 0.0082,
"reward": 0.8897488117218018,
"reward_std": 0.09283800423145294,
"reward_std_all": 0.09641597419977188,
"rewards-max/forensics_tempsamp_iou_reward": 0.9036122858524323,
"rewards/forensics_tempsamp_iou_reward": 0.6209987699985504,
"step": 583
},
{
"completion_length": 19.6875,
"epoch": 2.4435146443514646,
"grad_norm": 4.854077339172363,
"kl": 0.1923828125,
"learning_rate": 3.901673640167364e-07,
"loss": 0.0077,
"reward": 0.7576445937156677,
"reward_std": 0.11118551343679428,
"reward_std_all": 0.1175193153321743,
"rewards-max/forensics_tempsamp_iou_reward": 0.8803848028182983,
"rewards/forensics_tempsamp_iou_reward": 0.507644534111023,
"step": 584
},
{
"completion_length": 19.828125,
"epoch": 2.4476987447698746,
"grad_norm": 7.1031365394592285,
"kl": 0.2060546875,
"learning_rate": 3.891213389121339e-07,
"loss": 0.0082,
"reward": 0.813186764717102,
"reward_std": 0.09125462546944618,
"reward_std_all": 0.08182576671242714,
"rewards-max/forensics_tempsamp_iou_reward": 0.9592089653015137,
"rewards/forensics_tempsamp_iou_reward": 0.5319367796182632,
"step": 585
},
{
"completion_length": 21.671875,
"epoch": 2.4518828451882846,
"grad_norm": 6.807741165161133,
"kl": 0.2001953125,
"learning_rate": 3.880753138075314e-07,
"loss": 0.008,
"reward": 0.869337260723114,
"reward_std": 0.10792866721749306,
"reward_std_all": 0.10664808005094528,
"rewards-max/forensics_tempsamp_iou_reward": 0.9574138522148132,
"rewards/forensics_tempsamp_iou_reward": 0.5943372845649719,
"step": 586
},
{
"completion_length": 20.703125,
"epoch": 2.4560669456066946,
"grad_norm": 5.297536373138428,
"kl": 0.2021484375,
"learning_rate": 3.870292887029288e-07,
"loss": 0.0081,
"reward": 0.5532055497169495,
"reward_std": 0.06435329467058182,
"reward_std_all": 0.09175803884863853,
"rewards-max/forensics_tempsamp_iou_reward": 0.8096730709075928,
"rewards/forensics_tempsamp_iou_reward": 0.36570554226636887,
"step": 587
},
{
"completion_length": 21.171875,
"epoch": 2.4602510460251046,
"grad_norm": 4.840020179748535,
"kl": 0.27197265625,
"learning_rate": 3.8598326359832634e-07,
"loss": 0.0109,
"reward": 1.0008462369441986,
"reward_std": 0.0966714359819889,
"reward_std_all": 0.12925242260098457,
"rewards-max/forensics_tempsamp_iou_reward": 0.9885169267654419,
"rewards/forensics_tempsamp_iou_reward": 0.700846254825592,
"step": 588
},
{
"completion_length": 20.8125,
"epoch": 2.4644351464435146,
"grad_norm": 10.385025024414062,
"kl": 0.1806640625,
"learning_rate": 3.8493723849372385e-07,
"loss": 0.0072,
"reward": 0.9371341168880463,
"reward_std": 0.08863060176372528,
"reward_std_all": 0.09413032606244087,
"rewards-max/forensics_tempsamp_iou_reward": 0.9765430688858032,
"rewards/forensics_tempsamp_iou_reward": 0.6558841168880463,
"step": 589
},
{
"completion_length": 19.84375,
"epoch": 2.4686192468619246,
"grad_norm": 5.582331657409668,
"kl": 0.19921875,
"learning_rate": 3.8389121338912136e-07,
"loss": 0.008,
"reward": 0.7858001589775085,
"reward_std": 0.1182388886809349,
"reward_std_all": 0.11343411728739738,
"rewards-max/forensics_tempsamp_iou_reward": 0.927475243806839,
"rewards/forensics_tempsamp_iou_reward": 0.5295500755310059,
"step": 590
},
{
"completion_length": 19.21875,
"epoch": 2.4728033472803346,
"grad_norm": 6.032072067260742,
"kl": 0.20947265625,
"learning_rate": 3.828451882845188e-07,
"loss": 0.0084,
"reward": 0.8761445879936218,
"reward_std": 0.05198107846081257,
"reward_std_all": 0.05460864678025246,
"rewards-max/forensics_tempsamp_iou_reward": 0.9547987580299377,
"rewards/forensics_tempsamp_iou_reward": 0.6136445701122284,
"step": 591
},
{
"completion_length": 21.515625,
"epoch": 2.476987447698745,
"grad_norm": 7.293477535247803,
"kl": 0.2705078125,
"learning_rate": 3.817991631799163e-07,
"loss": 0.0108,
"reward": 0.8566181063652039,
"reward_std": 0.06676855683326721,
"reward_std_all": 0.06915068812668324,
"rewards-max/forensics_tempsamp_iou_reward": 0.9622986018657684,
"rewards/forensics_tempsamp_iou_reward": 0.594118058681488,
"step": 592
},
{
"completion_length": 21.53125,
"epoch": 2.481171548117155,
"grad_norm": 5.644343852996826,
"kl": 0.18212890625,
"learning_rate": 3.807531380753138e-07,
"loss": 0.0073,
"reward": 0.6878427863121033,
"reward_std": 0.10340671241283417,
"reward_std_all": 0.09769491478800774,
"rewards-max/forensics_tempsamp_iou_reward": 0.9253195226192474,
"rewards/forensics_tempsamp_iou_reward": 0.4565928280353546,
"step": 593
},
{
"completion_length": 19.921875,
"epoch": 2.485355648535565,
"grad_norm": 7.202208518981934,
"kl": 0.16748046875,
"learning_rate": 3.797071129707113e-07,
"loss": 0.0067,
"reward": 0.8490453362464905,
"reward_std": 0.1473342850804329,
"reward_std_all": 0.15578225255012512,
"rewards-max/forensics_tempsamp_iou_reward": 0.9490000009536743,
"rewards/forensics_tempsamp_iou_reward": 0.5552953034639359,
"step": 594
},
{
"completion_length": 20.28125,
"epoch": 2.489539748953975,
"grad_norm": 5.737064838409424,
"kl": 0.1904296875,
"learning_rate": 3.7866108786610877e-07,
"loss": 0.0076,
"reward": 0.7760838568210602,
"reward_std": 0.15826310589909554,
"reward_std_all": 0.15485338121652603,
"rewards-max/forensics_tempsamp_iou_reward": 0.9563879370689392,
"rewards/forensics_tempsamp_iou_reward": 0.5510838031768799,
"step": 595
},
{
"completion_length": 19.265625,
"epoch": 2.493723849372385,
"grad_norm": 10.077260971069336,
"kl": 0.20849609375,
"learning_rate": 3.776150627615063e-07,
"loss": 0.0084,
"reward": 0.8298962116241455,
"reward_std": 0.08680163137614727,
"reward_std_all": 0.09436693787574768,
"rewards-max/forensics_tempsamp_iou_reward": 0.9880990982055664,
"rewards/forensics_tempsamp_iou_reward": 0.6048962473869324,
"step": 596
},
{
"completion_length": 21.171875,
"epoch": 2.497907949790795,
"grad_norm": 5.528036594390869,
"kl": 0.16064453125,
"learning_rate": 3.7656903765690374e-07,
"loss": 0.0064,
"reward": 0.8958442211151123,
"reward_std": 0.06914406642317772,
"reward_std_all": 0.0927613265812397,
"rewards-max/forensics_tempsamp_iou_reward": 0.9227914810180664,
"rewards/forensics_tempsamp_iou_reward": 0.6020942032337189,
"step": 597
},
{
"completion_length": 20.71875,
"epoch": 2.502092050209205,
"grad_norm": 5.894502639770508,
"kl": 0.19921875,
"learning_rate": 3.7552301255230125e-07,
"loss": 0.008,
"reward": 0.822025716304779,
"reward_std": 0.1019892767071724,
"reward_std_all": 0.11128396540880203,
"rewards-max/forensics_tempsamp_iou_reward": 0.9721780717372894,
"rewards/forensics_tempsamp_iou_reward": 0.5532757639884949,
"step": 598
},
{
"completion_length": 19.75,
"epoch": 2.506276150627615,
"grad_norm": 4.2805280685424805,
"kl": 0.1708984375,
"learning_rate": 3.744769874476987e-07,
"loss": 0.0068,
"reward": 0.8491278290748596,
"reward_std": 0.06643960438668728,
"reward_std_all": 0.06574018858373165,
"rewards-max/forensics_tempsamp_iou_reward": 0.946080893278122,
"rewards/forensics_tempsamp_iou_reward": 0.5803778171539307,
"step": 599
},
{
"completion_length": 21.390625,
"epoch": 2.510460251046025,
"grad_norm": 5.597010612487793,
"kl": 0.189453125,
"learning_rate": 3.734309623430962e-07,
"loss": 0.0076,
"reward": 0.8626092672348022,
"reward_std": 0.09721511602401733,
"reward_std_all": 0.09235209971666336,
"rewards-max/forensics_tempsamp_iou_reward": 0.9898760616779327,
"rewards/forensics_tempsamp_iou_reward": 0.6063593029975891,
"step": 600
},
{
"completion_length": 20.640625,
"epoch": 2.514644351464435,
"grad_norm": 6.800095558166504,
"kl": 0.193359375,
"learning_rate": 3.723849372384937e-07,
"loss": 0.0077,
"reward": 0.7336352169513702,
"reward_std": 0.07328333891928196,
"reward_std_all": 0.0702967718243599,
"rewards-max/forensics_tempsamp_iou_reward": 0.9615329205989838,
"rewards/forensics_tempsamp_iou_reward": 0.5086352229118347,
"step": 601
},
{
"completion_length": 20.1875,
"epoch": 2.518828451882845,
"grad_norm": 6.013426780700684,
"kl": 0.1982421875,
"learning_rate": 3.713389121338912e-07,
"loss": 0.0079,
"reward": 0.9344053864479065,
"reward_std": 0.11202744022011757,
"reward_std_all": 0.12034991011023521,
"rewards-max/forensics_tempsamp_iou_reward": 0.936537355184555,
"rewards/forensics_tempsamp_iou_reward": 0.6656553149223328,
"step": 602
},
{
"completion_length": 20.21875,
"epoch": 2.523012552301255,
"grad_norm": 7.081243515014648,
"kl": 0.1669921875,
"learning_rate": 3.7029288702928866e-07,
"loss": 0.0067,
"reward": 0.714651495218277,
"reward_std": 0.06446066685020924,
"reward_std_all": 0.06905527412891388,
"rewards-max/forensics_tempsamp_iou_reward": 0.9587455987930298,
"rewards/forensics_tempsamp_iou_reward": 0.4709014296531677,
"step": 603
},
{
"completion_length": 22.484375,
"epoch": 2.5271966527196654,
"grad_norm": 4.424902439117432,
"kl": 0.3203125,
"learning_rate": 3.6924686192468617e-07,
"loss": 0.0128,
"reward": 0.7480040192604065,
"reward_std": 0.09537188708782196,
"reward_std_all": 0.11255637928843498,
"rewards-max/forensics_tempsamp_iou_reward": 0.984334796667099,
"rewards/forensics_tempsamp_iou_reward": 0.5292540490627289,
"step": 604
},
{
"completion_length": 20.921875,
"epoch": 2.5313807531380754,
"grad_norm": 5.03400993347168,
"kl": 0.16943359375,
"learning_rate": 3.682008368200837e-07,
"loss": 0.0068,
"reward": 0.783243328332901,
"reward_std": 0.07590292580425739,
"reward_std_all": 0.07346416264772415,
"rewards-max/forensics_tempsamp_iou_reward": 0.9473426640033722,
"rewards/forensics_tempsamp_iou_reward": 0.5269933044910431,
"step": 605
},
{
"completion_length": 20.46875,
"epoch": 2.5355648535564854,
"grad_norm": 5.1441330909729,
"kl": 0.21484375,
"learning_rate": 3.6715481171548114e-07,
"loss": 0.0086,
"reward": 0.7621039152145386,
"reward_std": 0.06339849531650543,
"reward_std_all": 0.059872861951589584,
"rewards-max/forensics_tempsamp_iou_reward": 0.8585434257984161,
"rewards/forensics_tempsamp_iou_reward": 0.4871039390563965,
"step": 606
},
{
"completion_length": 20.703125,
"epoch": 2.5397489539748954,
"grad_norm": 6.551379680633545,
"kl": 0.2939453125,
"learning_rate": 3.661087866108786e-07,
"loss": 0.0118,
"reward": 0.8612671196460724,
"reward_std": 0.09236548468470573,
"reward_std_all": 0.08007372543215752,
"rewards-max/forensics_tempsamp_iou_reward": 0.9575724303722382,
"rewards/forensics_tempsamp_iou_reward": 0.58001708984375,
"step": 607
},
{
"completion_length": 20.125,
"epoch": 2.5439330543933054,
"grad_norm": 4.165430545806885,
"kl": 0.1611328125,
"learning_rate": 3.650627615062761e-07,
"loss": 0.0064,
"reward": 0.8536880612373352,
"reward_std": 0.0786689855158329,
"reward_std_all": 0.0731958020478487,
"rewards-max/forensics_tempsamp_iou_reward": 0.9994988739490509,
"rewards/forensics_tempsamp_iou_reward": 0.591187983751297,
"step": 608
},
{
"completion_length": 20.0,
"epoch": 2.5481171548117154,
"grad_norm": 4.324161529541016,
"kl": 0.19140625,
"learning_rate": 3.6401673640167363e-07,
"loss": 0.0077,
"reward": 0.8701904714107513,
"reward_std": 0.11419408768415451,
"reward_std_all": 0.1115323156118393,
"rewards-max/forensics_tempsamp_iou_reward": 0.9252484142780304,
"rewards/forensics_tempsamp_iou_reward": 0.588940441608429,
"step": 609
},
{
"completion_length": 21.234375,
"epoch": 2.5523012552301254,
"grad_norm": 4.926645755767822,
"kl": 0.234375,
"learning_rate": 3.6297071129707114e-07,
"loss": 0.0094,
"reward": 0.8271781802177429,
"reward_std": 0.07373835518956184,
"reward_std_all": 0.09626222401857376,
"rewards-max/forensics_tempsamp_iou_reward": 0.919380396604538,
"rewards/forensics_tempsamp_iou_reward": 0.5646781325340271,
"step": 610
},
{
"completion_length": 24.25,
"epoch": 2.5564853556485354,
"grad_norm": 4.502800464630127,
"kl": 0.2607421875,
"learning_rate": 3.6192468619246855e-07,
"loss": 0.0104,
"reward": 0.7110764384269714,
"reward_std": 0.09917126968502998,
"reward_std_all": 0.11369679123163223,
"rewards-max/forensics_tempsamp_iou_reward": 0.9481720626354218,
"rewards/forensics_tempsamp_iou_reward": 0.5048263818025589,
"step": 611
},
{
"completion_length": 21.125,
"epoch": 2.560669456066946,
"grad_norm": 4.876762866973877,
"kl": 0.18115234375,
"learning_rate": 3.6087866108786606e-07,
"loss": 0.0072,
"reward": 0.7069624364376068,
"reward_std": 0.0793485939502716,
"reward_std_all": 0.08729111962020397,
"rewards-max/forensics_tempsamp_iou_reward": 0.9680996835231781,
"rewards/forensics_tempsamp_iou_reward": 0.4757124036550522,
"step": 612
},
{
"completion_length": 22.078125,
"epoch": 2.564853556485356,
"grad_norm": 6.057740688323975,
"kl": 0.19091796875,
"learning_rate": 3.598326359832636e-07,
"loss": 0.0076,
"reward": 0.7884946763515472,
"reward_std": 0.11094876751303673,
"reward_std_all": 0.10364798828959465,
"rewards-max/forensics_tempsamp_iou_reward": 0.9240118861198425,
"rewards/forensics_tempsamp_iou_reward": 0.5322446823120117,
"step": 613
},
{
"completion_length": 19.859375,
"epoch": 2.569037656903766,
"grad_norm": 4.535978317260742,
"kl": 0.2041015625,
"learning_rate": 3.587866108786611e-07,
"loss": 0.0082,
"reward": 0.6738980710506439,
"reward_std": 0.06042826175689697,
"reward_std_all": 0.05442884378135204,
"rewards-max/forensics_tempsamp_iou_reward": 0.8403000235557556,
"rewards/forensics_tempsamp_iou_reward": 0.4488980770111084,
"step": 614
},
{
"completion_length": 20.984375,
"epoch": 2.573221757322176,
"grad_norm": 4.1841230392456055,
"kl": 0.234375,
"learning_rate": 3.577405857740586e-07,
"loss": 0.0094,
"reward": 0.9022273421287537,
"reward_std": 0.058929141610860825,
"reward_std_all": 0.05621855519711971,
"rewards-max/forensics_tempsamp_iou_reward": 0.9523241519927979,
"rewards/forensics_tempsamp_iou_reward": 0.6022273898124695,
"step": 615
},
{
"completion_length": 20.9375,
"epoch": 2.577405857740586,
"grad_norm": 8.850273132324219,
"kl": 0.1796875,
"learning_rate": 3.56694560669456e-07,
"loss": 0.0072,
"reward": 0.7279095351696014,
"reward_std": 0.040770155377686024,
"reward_std_all": 0.07458714209496975,
"rewards-max/forensics_tempsamp_iou_reward": 0.9522014856338501,
"rewards/forensics_tempsamp_iou_reward": 0.5029094964265823,
"step": 616
},
{
"completion_length": 19.515625,
"epoch": 2.581589958158996,
"grad_norm": 4.493823528289795,
"kl": 0.162109375,
"learning_rate": 3.556485355648535e-07,
"loss": 0.0065,
"reward": 0.8975645899772644,
"reward_std": 0.06195509806275368,
"reward_std_all": 0.0584196038544178,
"rewards-max/forensics_tempsamp_iou_reward": 0.9644642770290375,
"rewards/forensics_tempsamp_iou_reward": 0.6288145780563354,
"step": 617
},
{
"completion_length": 20.15625,
"epoch": 2.585774058577406,
"grad_norm": 6.58408784866333,
"kl": 0.24462890625,
"learning_rate": 3.5460251046025103e-07,
"loss": 0.0098,
"reward": 0.7495608031749725,
"reward_std": 0.08294487744569778,
"reward_std_all": 0.08340511471033096,
"rewards-max/forensics_tempsamp_iou_reward": 0.83553746342659,
"rewards/forensics_tempsamp_iou_reward": 0.4870608150959015,
"step": 618
},
{
"completion_length": 20.0,
"epoch": 2.589958158995816,
"grad_norm": 5.926811218261719,
"kl": 0.1884765625,
"learning_rate": 3.5355648535564855e-07,
"loss": 0.0075,
"reward": 0.9292153418064117,
"reward_std": 0.11061679199337959,
"reward_std_all": 0.11015992239117622,
"rewards-max/forensics_tempsamp_iou_reward": 0.9342105388641357,
"rewards/forensics_tempsamp_iou_reward": 0.6354653537273407,
"step": 619
},
{
"completion_length": 20.078125,
"epoch": 2.594142259414226,
"grad_norm": 4.82370138168335,
"kl": 0.26611328125,
"learning_rate": 3.52510460251046e-07,
"loss": 0.0106,
"reward": 0.8984701335430145,
"reward_std": 0.06605037674307823,
"reward_std_all": 0.06083778664469719,
"rewards-max/forensics_tempsamp_iou_reward": 0.9907528460025787,
"rewards/forensics_tempsamp_iou_reward": 0.6172201633453369,
"step": 620
},
{
"completion_length": 22.984375,
"epoch": 2.598326359832636,
"grad_norm": 5.012823104858398,
"kl": 0.25537109375,
"learning_rate": 3.5146443514644347e-07,
"loss": 0.0102,
"reward": 0.8242910802364349,
"reward_std": 0.09483649954199791,
"reward_std_all": 0.10143988579511642,
"rewards-max/forensics_tempsamp_iou_reward": 0.9206887781620026,
"rewards/forensics_tempsamp_iou_reward": 0.5805411040782928,
"step": 621
},
{
"completion_length": 20.4375,
"epoch": 2.602510460251046,
"grad_norm": 4.235166072845459,
"kl": 0.20263671875,
"learning_rate": 3.50418410041841e-07,
"loss": 0.0081,
"reward": 0.8475816249847412,
"reward_std": 0.13514988124370575,
"reward_std_all": 0.12416186556220055,
"rewards-max/forensics_tempsamp_iou_reward": 0.9675457179546356,
"rewards/forensics_tempsamp_iou_reward": 0.5788316130638123,
"step": 622
},
{
"completion_length": 21.40625,
"epoch": 2.606694560669456,
"grad_norm": 5.572633266448975,
"kl": 0.2080078125,
"learning_rate": 3.493723849372385e-07,
"loss": 0.0083,
"reward": 0.7043496370315552,
"reward_std": 0.09166326746344566,
"reward_std_all": 0.09888366237282753,
"rewards-max/forensics_tempsamp_iou_reward": 0.906641811132431,
"rewards/forensics_tempsamp_iou_reward": 0.4543496370315552,
"step": 623
},
{
"completion_length": 20.046875,
"epoch": 2.6108786610878663,
"grad_norm": 9.217677116394043,
"kl": 0.607421875,
"learning_rate": 3.4832635983263595e-07,
"loss": 0.0243,
"reward": 0.7020125985145569,
"reward_std": 0.11061655730009079,
"reward_std_all": 0.10439477115869522,
"rewards-max/forensics_tempsamp_iou_reward": 0.9043394327163696,
"rewards/forensics_tempsamp_iou_reward": 0.477012574672699,
"step": 624
},
{
"completion_length": 20.09375,
"epoch": 2.6150627615062763,
"grad_norm": 9.843174934387207,
"kl": 0.24267578125,
"learning_rate": 3.4728033472803346e-07,
"loss": 0.0097,
"reward": 0.8030969798564911,
"reward_std": 0.07424535602331161,
"reward_std_all": 0.07552865892648697,
"rewards-max/forensics_tempsamp_iou_reward": 0.9621820151805878,
"rewards/forensics_tempsamp_iou_reward": 0.54059699177742,
"step": 625
},
{
"completion_length": 20.234375,
"epoch": 2.6192468619246863,
"grad_norm": 7.325879096984863,
"kl": 0.24169921875,
"learning_rate": 3.462343096234309e-07,
"loss": 0.0097,
"reward": 0.5704999268054962,
"reward_std": 0.09133920073509216,
"reward_std_all": 0.10718759149312973,
"rewards-max/forensics_tempsamp_iou_reward": 0.8883001208305359,
"rewards/forensics_tempsamp_iou_reward": 0.370499923825264,
"step": 626
},
{
"completion_length": 20.640625,
"epoch": 2.6234309623430963,
"grad_norm": 5.273987293243408,
"kl": 0.1875,
"learning_rate": 3.4518828451882844e-07,
"loss": 0.0075,
"reward": 0.64668408036232,
"reward_std": 0.09427374973893166,
"reward_std_all": 0.10709137469530106,
"rewards-max/forensics_tempsamp_iou_reward": 0.8889458775520325,
"rewards/forensics_tempsamp_iou_reward": 0.440434068441391,
"step": 627
},
{
"completion_length": 21.125,
"epoch": 2.6276150627615062,
"grad_norm": 4.34000825881958,
"kl": 0.17041015625,
"learning_rate": 3.441422594142259e-07,
"loss": 0.0068,
"reward": 0.8214268386363983,
"reward_std": 0.06510229036211967,
"reward_std_all": 0.07313190773129463,
"rewards-max/forensics_tempsamp_iou_reward": 0.970091313123703,
"rewards/forensics_tempsamp_iou_reward": 0.5776768326759338,
"step": 628
},
{
"completion_length": 21.984375,
"epoch": 2.6317991631799162,
"grad_norm": 8.422666549682617,
"kl": 0.1943359375,
"learning_rate": 3.430962343096234e-07,
"loss": 0.0078,
"reward": 0.9003874957561493,
"reward_std": 0.12358853220939636,
"reward_std_all": 0.1360606998205185,
"rewards-max/forensics_tempsamp_iou_reward": 0.9679976105690002,
"rewards/forensics_tempsamp_iou_reward": 0.6003874242305756,
"step": 629
},
{
"completion_length": 21.734375,
"epoch": 2.6359832635983262,
"grad_norm": 4.427485942840576,
"kl": 0.23193359375,
"learning_rate": 3.420502092050209e-07,
"loss": 0.0093,
"reward": 0.7696331143379211,
"reward_std": 0.12153154611587524,
"reward_std_all": 0.11035013198852539,
"rewards-max/forensics_tempsamp_iou_reward": 0.9326688051223755,
"rewards/forensics_tempsamp_iou_reward": 0.5321330726146698,
"step": 630
},
{
"completion_length": 21.609375,
"epoch": 2.6401673640167362,
"grad_norm": 5.2793869972229,
"kl": 0.20751953125,
"learning_rate": 3.410041841004184e-07,
"loss": 0.0083,
"reward": 0.8190011978149414,
"reward_std": 0.05983816273510456,
"reward_std_all": 0.08828002959489822,
"rewards-max/forensics_tempsamp_iou_reward": 0.9158983528614044,
"rewards/forensics_tempsamp_iou_reward": 0.5565011501312256,
"step": 631
},
{
"completion_length": 20.625,
"epoch": 2.6443514644351467,
"grad_norm": 4.723827838897705,
"kl": 0.27490234375,
"learning_rate": 3.3995815899581584e-07,
"loss": 0.011,
"reward": 0.7426750659942627,
"reward_std": 0.08816316910088062,
"reward_std_all": 0.08822921570390463,
"rewards-max/forensics_tempsamp_iou_reward": 0.9138504266738892,
"rewards/forensics_tempsamp_iou_reward": 0.5114250183105469,
"step": 632
},
{
"completion_length": 21.515625,
"epoch": 2.6485355648535567,
"grad_norm": 4.957258224487305,
"kl": 0.18017578125,
"learning_rate": 3.3891213389121336e-07,
"loss": 0.0072,
"reward": 0.8494804501533508,
"reward_std": 0.09078509360551834,
"reward_std_all": 0.09325071796774864,
"rewards-max/forensics_tempsamp_iou_reward": 0.9577869474887848,
"rewards/forensics_tempsamp_iou_reward": 0.5744803845882416,
"step": 633
},
{
"completion_length": 20.03125,
"epoch": 2.6527196652719667,
"grad_norm": 6.898361682891846,
"kl": 0.23095703125,
"learning_rate": 3.3786610878661087e-07,
"loss": 0.0092,
"reward": 0.8426689505577087,
"reward_std": 0.07706702128052711,
"reward_std_all": 0.10311945527791977,
"rewards-max/forensics_tempsamp_iou_reward": 0.9737820327281952,
"rewards/forensics_tempsamp_iou_reward": 0.5989188849925995,
"step": 634
},
{
"completion_length": 19.9375,
"epoch": 2.6569037656903767,
"grad_norm": 4.805290222167969,
"kl": 0.25927734375,
"learning_rate": 3.368200836820084e-07,
"loss": 0.0104,
"reward": 0.839143306016922,
"reward_std": 0.11297395080327988,
"reward_std_all": 0.10797487199306488,
"rewards-max/forensics_tempsamp_iou_reward": 0.9515917301177979,
"rewards/forensics_tempsamp_iou_reward": 0.5578932762145996,
"step": 635
},
{
"completion_length": 20.546875,
"epoch": 2.6610878661087867,
"grad_norm": 4.485482215881348,
"kl": 0.17236328125,
"learning_rate": 3.3577405857740584e-07,
"loss": 0.0069,
"reward": 0.8373636901378632,
"reward_std": 0.10960780829191208,
"reward_std_all": 0.09901289269328117,
"rewards-max/forensics_tempsamp_iou_reward": 0.926111102104187,
"rewards/forensics_tempsamp_iou_reward": 0.5936136841773987,
"step": 636
},
{
"completion_length": 21.5,
"epoch": 2.6652719665271967,
"grad_norm": 7.634530067443848,
"kl": 0.30322265625,
"learning_rate": 3.347280334728033e-07,
"loss": 0.0121,
"reward": 0.7936932444572449,
"reward_std": 0.091767068952322,
"reward_std_all": 0.11422779783606529,
"rewards-max/forensics_tempsamp_iou_reward": 0.9599959850311279,
"rewards/forensics_tempsamp_iou_reward": 0.5874432325363159,
"step": 637
},
{
"completion_length": 19.953125,
"epoch": 2.6694560669456067,
"grad_norm": 5.258272647857666,
"kl": 0.2041015625,
"learning_rate": 3.336820083682008e-07,
"loss": 0.0082,
"reward": 0.7050544321537018,
"reward_std": 0.06951872259378433,
"reward_std_all": 0.0872751995921135,
"rewards-max/forensics_tempsamp_iou_reward": 0.916875422000885,
"rewards/forensics_tempsamp_iou_reward": 0.4613044112920761,
"step": 638
},
{
"completion_length": 20.8125,
"epoch": 2.6736401673640167,
"grad_norm": 4.905300140380859,
"kl": 0.2080078125,
"learning_rate": 3.3263598326359833e-07,
"loss": 0.0083,
"reward": 0.798465758562088,
"reward_std": 0.058457160368561745,
"reward_std_all": 0.06220287084579468,
"rewards-max/forensics_tempsamp_iou_reward": 0.9043493270874023,
"rewards/forensics_tempsamp_iou_reward": 0.5359657108783722,
"step": 639
},
{
"completion_length": 22.359375,
"epoch": 2.6778242677824267,
"grad_norm": 4.6229047775268555,
"kl": 0.17333984375,
"learning_rate": 3.3158995815899584e-07,
"loss": 0.007,
"reward": 0.596031129360199,
"reward_std": 0.10093620419502258,
"reward_std_all": 0.09891637414693832,
"rewards-max/forensics_tempsamp_iou_reward": 0.9051241278648376,
"rewards/forensics_tempsamp_iou_reward": 0.3835311383008957,
"step": 640
},
{
"completion_length": 19.484375,
"epoch": 2.6820083682008367,
"grad_norm": 14.480425834655762,
"kl": 0.25830078125,
"learning_rate": 3.3054393305439325e-07,
"loss": 0.0103,
"reward": 0.6596346497535706,
"reward_std": 0.08615501597523689,
"reward_std_all": 0.08259401097893715,
"rewards-max/forensics_tempsamp_iou_reward": 0.9132545590400696,
"rewards/forensics_tempsamp_iou_reward": 0.43463467061519623,
"step": 641
},
{
"completion_length": 20.40625,
"epoch": 2.6861924686192467,
"grad_norm": 6.0682053565979,
"kl": 0.19140625,
"learning_rate": 3.2949790794979076e-07,
"loss": 0.0077,
"reward": 0.8071572482585907,
"reward_std": 0.07460752129554749,
"reward_std_all": 0.07550089806318283,
"rewards-max/forensics_tempsamp_iou_reward": 0.9023326337337494,
"rewards/forensics_tempsamp_iou_reward": 0.5446572601795197,
"step": 642
},
{
"completion_length": 20.765625,
"epoch": 2.6903765690376567,
"grad_norm": 6.108468532562256,
"kl": 0.17919921875,
"learning_rate": 3.2845188284518827e-07,
"loss": 0.0072,
"reward": 0.5610159635543823,
"reward_std": 0.07382835447788239,
"reward_std_all": 0.09118270874023438,
"rewards-max/forensics_tempsamp_iou_reward": 0.8884581029415131,
"rewards/forensics_tempsamp_iou_reward": 0.37976595759391785,
"step": 643
},
{
"completion_length": 19.484375,
"epoch": 2.694560669456067,
"grad_norm": 5.757806777954102,
"kl": 0.3212890625,
"learning_rate": 3.274058577405858e-07,
"loss": 0.0128,
"reward": 0.859315812587738,
"reward_std": 0.058715641498565674,
"reward_std_all": 0.07311167940497398,
"rewards-max/forensics_tempsamp_iou_reward": 0.9841897189617157,
"rewards/forensics_tempsamp_iou_reward": 0.584315836429596,
"step": 644
},
{
"completion_length": 19.46875,
"epoch": 2.698744769874477,
"grad_norm": 5.88788366317749,
"kl": 0.20751953125,
"learning_rate": 3.2635983263598325e-07,
"loss": 0.0083,
"reward": 0.7330654561519623,
"reward_std": 0.06878883577883244,
"reward_std_all": 0.061190079897642136,
"rewards-max/forensics_tempsamp_iou_reward": 0.9752644002437592,
"rewards/forensics_tempsamp_iou_reward": 0.5080654919147491,
"step": 645
},
{
"completion_length": 20.234375,
"epoch": 2.702928870292887,
"grad_norm": 6.159724235534668,
"kl": 0.4423828125,
"learning_rate": 3.253138075313807e-07,
"loss": 0.0177,
"reward": 0.7100460529327393,
"reward_std": 0.15580304712057114,
"reward_std_all": 0.14070936292409897,
"rewards-max/forensics_tempsamp_iou_reward": 0.9013502299785614,
"rewards/forensics_tempsamp_iou_reward": 0.46004606783390045,
"step": 646
},
{
"completion_length": 20.90625,
"epoch": 2.707112970711297,
"grad_norm": 5.436916828155518,
"kl": 0.22265625,
"learning_rate": 3.242677824267782e-07,
"loss": 0.0089,
"reward": 0.774897426366806,
"reward_std": 0.05176848452538252,
"reward_std_all": 0.08982807397842407,
"rewards-max/forensics_tempsamp_iou_reward": 0.8752261698246002,
"rewards/forensics_tempsamp_iou_reward": 0.5123974084854126,
"step": 647
},
{
"completion_length": 20.234375,
"epoch": 2.711297071129707,
"grad_norm": 6.8205671310424805,
"kl": 0.1796875,
"learning_rate": 3.2322175732217573e-07,
"loss": 0.0072,
"reward": 0.9104834496974945,
"reward_std": 0.07327147573232651,
"reward_std_all": 0.07940509170293808,
"rewards-max/forensics_tempsamp_iou_reward": 0.9836128056049347,
"rewards/forensics_tempsamp_iou_reward": 0.6292334794998169,
"step": 648
},
{
"completion_length": 21.109375,
"epoch": 2.715481171548117,
"grad_norm": 6.515774250030518,
"kl": 0.22705078125,
"learning_rate": 3.221757322175732e-07,
"loss": 0.0091,
"reward": 0.8052845597267151,
"reward_std": 0.0781508032232523,
"reward_std_all": 0.08435656502842903,
"rewards-max/forensics_tempsamp_iou_reward": 0.9026113152503967,
"rewards/forensics_tempsamp_iou_reward": 0.5240345597267151,
"step": 649
},
{
"completion_length": 21.484375,
"epoch": 2.719665271966527,
"grad_norm": 7.112102031707764,
"kl": 0.287109375,
"learning_rate": 3.211297071129707e-07,
"loss": 0.0115,
"reward": 0.6703790128231049,
"reward_std": 0.12505191192030907,
"reward_std_all": 0.11880094558000565,
"rewards-max/forensics_tempsamp_iou_reward": 0.9202735424041748,
"rewards/forensics_tempsamp_iou_reward": 0.45162904262542725,
"step": 650
},
{
"completion_length": 19.21875,
"epoch": 2.723849372384937,
"grad_norm": 4.605976581573486,
"kl": 0.169921875,
"learning_rate": 3.2008368200836816e-07,
"loss": 0.0068,
"reward": 0.8180746734142303,
"reward_std": 0.04606065712869167,
"reward_std_all": 0.04407161846756935,
"rewards-max/forensics_tempsamp_iou_reward": 0.8950508534908295,
"rewards/forensics_tempsamp_iou_reward": 0.5555746555328369,
"step": 651
},
{
"completion_length": 20.578125,
"epoch": 2.7280334728033475,
"grad_norm": 5.283288478851318,
"kl": 0.18115234375,
"learning_rate": 3.190376569037657e-07,
"loss": 0.0072,
"reward": 0.8629083335399628,
"reward_std": 0.11024075746536255,
"reward_std_all": 0.12581834197044373,
"rewards-max/forensics_tempsamp_iou_reward": 0.891412615776062,
"rewards/forensics_tempsamp_iou_reward": 0.5941582918167114,
"step": 652
},
{
"completion_length": 19.859375,
"epoch": 2.7322175732217575,
"grad_norm": 6.839631080627441,
"kl": 0.1943359375,
"learning_rate": 3.1799163179916314e-07,
"loss": 0.0078,
"reward": 0.6190532445907593,
"reward_std": 0.10794354975223541,
"reward_std_all": 0.12137752026319504,
"rewards-max/forensics_tempsamp_iou_reward": 0.9572031497955322,
"rewards/forensics_tempsamp_iou_reward": 0.42530323565006256,
"step": 653
},
{
"completion_length": 19.46875,
"epoch": 2.7364016736401675,
"grad_norm": 3.7283411026000977,
"kl": 0.168701171875,
"learning_rate": 3.1694560669456065e-07,
"loss": 0.0068,
"reward": 0.8129734396934509,
"reward_std": 0.05251630861312151,
"reward_std_all": 0.059141991659998894,
"rewards-max/forensics_tempsamp_iou_reward": 0.9487444460391998,
"rewards/forensics_tempsamp_iou_reward": 0.5692234337329865,
"step": 654
},
{
"completion_length": 19.71875,
"epoch": 2.7405857740585775,
"grad_norm": 8.533520698547363,
"kl": 0.1689453125,
"learning_rate": 3.1589958158995816e-07,
"loss": 0.0068,
"reward": 0.74663046002388,
"reward_std": 0.07409488037228584,
"reward_std_all": 0.08063921332359314,
"rewards-max/forensics_tempsamp_iou_reward": 0.9009277522563934,
"rewards/forensics_tempsamp_iou_reward": 0.49663046002388,
"step": 655
},
{
"completion_length": 19.375,
"epoch": 2.7447698744769875,
"grad_norm": 8.89717960357666,
"kl": 0.1943359375,
"learning_rate": 3.148535564853556e-07,
"loss": 0.0078,
"reward": 1.0098183751106262,
"reward_std": 0.08751573041081429,
"reward_std_all": 0.08299465477466583,
"rewards-max/forensics_tempsamp_iou_reward": 0.9795015752315521,
"rewards/forensics_tempsamp_iou_reward": 0.7098183929920197,
"step": 656
},
{
"completion_length": 20.578125,
"epoch": 2.7489539748953975,
"grad_norm": 6.493984222412109,
"kl": 0.17529296875,
"learning_rate": 3.1380753138075313e-07,
"loss": 0.007,
"reward": 0.832367330789566,
"reward_std": 0.05137710087001324,
"reward_std_all": 0.04847449250519276,
"rewards-max/forensics_tempsamp_iou_reward": 0.9727492332458496,
"rewards/forensics_tempsamp_iou_reward": 0.5886172950267792,
"step": 657
},
{
"completion_length": 20.125,
"epoch": 2.7531380753138075,
"grad_norm": 7.991951942443848,
"kl": 0.20068359375,
"learning_rate": 3.127615062761506e-07,
"loss": 0.008,
"reward": 0.7834153175354004,
"reward_std": 0.11800814419984818,
"reward_std_all": 0.10761496052145958,
"rewards-max/forensics_tempsamp_iou_reward": 0.9681159555912018,
"rewards/forensics_tempsamp_iou_reward": 0.520915299654007,
"step": 658
},
{
"completion_length": 22.953125,
"epoch": 2.7573221757322175,
"grad_norm": 6.719840049743652,
"kl": 0.27490234375,
"learning_rate": 3.117154811715481e-07,
"loss": 0.011,
"reward": 0.5680880695581436,
"reward_std": 0.06133674085140228,
"reward_std_all": 0.07554551586508751,
"rewards-max/forensics_tempsamp_iou_reward": 0.8720759451389313,
"rewards/forensics_tempsamp_iou_reward": 0.4055880606174469,
"step": 659
},
{
"completion_length": 20.1875,
"epoch": 2.7615062761506275,
"grad_norm": 4.893772125244141,
"kl": 0.18017578125,
"learning_rate": 3.106694560669456e-07,
"loss": 0.0072,
"reward": 0.9399865865707397,
"reward_std": 0.13855689764022827,
"reward_std_all": 0.17552970349788666,
"rewards-max/forensics_tempsamp_iou_reward": 0.9732608795166016,
"rewards/forensics_tempsamp_iou_reward": 0.6587366163730621,
"step": 660
},
{
"completion_length": 21.21875,
"epoch": 2.7656903765690375,
"grad_norm": 5.507148742675781,
"kl": 0.24365234375,
"learning_rate": 3.096234309623431e-07,
"loss": 0.0097,
"reward": 0.8607252836227417,
"reward_std": 0.10168198123574257,
"reward_std_all": 0.10735760629177094,
"rewards-max/forensics_tempsamp_iou_reward": 0.9324597120285034,
"rewards/forensics_tempsamp_iou_reward": 0.5794753134250641,
"step": 661
},
{
"completion_length": 19.59375,
"epoch": 2.7698744769874475,
"grad_norm": 6.862271308898926,
"kl": 0.197265625,
"learning_rate": 3.0857740585774054e-07,
"loss": 0.0079,
"reward": 0.7904066741466522,
"reward_std": 0.05316418409347534,
"reward_std_all": 0.10023283958435059,
"rewards-max/forensics_tempsamp_iou_reward": 0.8844965696334839,
"rewards/forensics_tempsamp_iou_reward": 0.52790667116642,
"step": 662
},
{
"completion_length": 20.796875,
"epoch": 2.7740585774058575,
"grad_norm": 5.585855484008789,
"kl": 0.19287109375,
"learning_rate": 3.0753138075313805e-07,
"loss": 0.0077,
"reward": 0.884100079536438,
"reward_std": 0.07227933406829834,
"reward_std_all": 0.06735477969050407,
"rewards-max/forensics_tempsamp_iou_reward": 0.9985198974609375,
"rewards/forensics_tempsamp_iou_reward": 0.634100079536438,
"step": 663
},
{
"completion_length": 22.34375,
"epoch": 2.778242677824268,
"grad_norm": 4.370343208312988,
"kl": 0.205078125,
"learning_rate": 3.0648535564853557e-07,
"loss": 0.0082,
"reward": 0.7263499200344086,
"reward_std": 0.11490900814533234,
"reward_std_all": 0.10681981965899467,
"rewards-max/forensics_tempsamp_iou_reward": 0.9555803537368774,
"rewards/forensics_tempsamp_iou_reward": 0.4950999319553375,
"step": 664
},
{
"completion_length": 19.03125,
"epoch": 2.782426778242678,
"grad_norm": 6.070370197296143,
"kl": 0.2353515625,
"learning_rate": 3.054393305439331e-07,
"loss": 0.0094,
"reward": 0.6544511616230011,
"reward_std": 0.10051285475492477,
"reward_std_all": 0.11844871565699577,
"rewards-max/forensics_tempsamp_iou_reward": 0.8563385009765625,
"rewards/forensics_tempsamp_iou_reward": 0.4419512003660202,
"step": 665
},
{
"completion_length": 20.328125,
"epoch": 2.786610878661088,
"grad_norm": 8.899542808532715,
"kl": 0.5595703125,
"learning_rate": 3.043933054393305e-07,
"loss": 0.0223,
"reward": 0.7879388928413391,
"reward_std": 0.050643229857087135,
"reward_std_all": 0.0522985365241766,
"rewards-max/forensics_tempsamp_iou_reward": 0.9641894102096558,
"rewards/forensics_tempsamp_iou_reward": 0.5254389047622681,
"step": 666
},
{
"completion_length": 21.703125,
"epoch": 2.790794979079498,
"grad_norm": 6.804049968719482,
"kl": 0.224609375,
"learning_rate": 3.03347280334728e-07,
"loss": 0.009,
"reward": 0.828371524810791,
"reward_std": 0.10462157800793648,
"reward_std_all": 0.10606299340724945,
"rewards-max/forensics_tempsamp_iou_reward": 0.9318304061889648,
"rewards/forensics_tempsamp_iou_reward": 0.5846214592456818,
"step": 667
},
{
"completion_length": 19.734375,
"epoch": 2.794979079497908,
"grad_norm": 6.283775806427002,
"kl": 0.19140625,
"learning_rate": 3.023012552301255e-07,
"loss": 0.0077,
"reward": 0.8446761965751648,
"reward_std": 0.09537205100059509,
"reward_std_all": 0.09544693306088448,
"rewards-max/forensics_tempsamp_iou_reward": 0.954266756772995,
"rewards/forensics_tempsamp_iou_reward": 0.56342613697052,
"step": 668
},
{
"completion_length": 21.59375,
"epoch": 2.799163179916318,
"grad_norm": 4.281744956970215,
"kl": 0.1982421875,
"learning_rate": 3.01255230125523e-07,
"loss": 0.0079,
"reward": 0.6420599818229675,
"reward_std": 0.07078928872942924,
"reward_std_all": 0.0664643794298172,
"rewards-max/forensics_tempsamp_iou_reward": 0.9187992513179779,
"rewards/forensics_tempsamp_iou_reward": 0.45455998182296753,
"step": 669
},
{
"completion_length": 21.59375,
"epoch": 2.803347280334728,
"grad_norm": 5.387828350067139,
"kl": 0.18408203125,
"learning_rate": 3.002092050209205e-07,
"loss": 0.0074,
"reward": 0.7419160008430481,
"reward_std": 0.1364210695028305,
"reward_std_all": 0.12995852902531624,
"rewards-max/forensics_tempsamp_iou_reward": 0.9563265740871429,
"rewards/forensics_tempsamp_iou_reward": 0.5106659978628159,
"step": 670
},
{
"completion_length": 19.75,
"epoch": 2.8075313807531384,
"grad_norm": 9.6547269821167,
"kl": 0.2255859375,
"learning_rate": 2.9916317991631794e-07,
"loss": 0.009,
"reward": 0.8135649561882019,
"reward_std": 0.16392094641923904,
"reward_std_all": 0.14236608892679214,
"rewards-max/forensics_tempsamp_iou_reward": 0.9571601748466492,
"rewards/forensics_tempsamp_iou_reward": 0.5698149651288986,
"step": 671
},
{
"completion_length": 20.953125,
"epoch": 2.8117154811715483,
"grad_norm": 6.220807075500488,
"kl": 0.314453125,
"learning_rate": 2.9811715481171546e-07,
"loss": 0.0126,
"reward": 0.8401525914669037,
"reward_std": 0.12776606529951096,
"reward_std_all": 0.13955529779195786,
"rewards-max/forensics_tempsamp_iou_reward": 0.9827375113964081,
"rewards/forensics_tempsamp_iou_reward": 0.5714024901390076,
"step": 672
},
{
"completion_length": 20.171875,
"epoch": 2.8158995815899583,
"grad_norm": 5.489406108856201,
"kl": 0.236328125,
"learning_rate": 2.9707112970711297e-07,
"loss": 0.0095,
"reward": 0.6501046419143677,
"reward_std": 0.10842456668615341,
"reward_std_all": 0.10385278984904289,
"rewards-max/forensics_tempsamp_iou_reward": 0.8640714287757874,
"rewards/forensics_tempsamp_iou_reward": 0.4313546270132065,
"step": 673
},
{
"completion_length": 20.390625,
"epoch": 2.8200836820083683,
"grad_norm": 6.209896564483643,
"kl": 0.20703125,
"learning_rate": 2.960251046025105e-07,
"loss": 0.0083,
"reward": 0.6572745144367218,
"reward_std": 0.0798945389688015,
"reward_std_all": 0.07768618315458298,
"rewards-max/forensics_tempsamp_iou_reward": 0.9008100032806396,
"rewards/forensics_tempsamp_iou_reward": 0.438524529337883,
"step": 674
},
{
"completion_length": 20.140625,
"epoch": 2.8242677824267783,
"grad_norm": 6.897103786468506,
"kl": 0.20263671875,
"learning_rate": 2.9497907949790794e-07,
"loss": 0.0081,
"reward": 0.7665424644947052,
"reward_std": 0.06630626507103443,
"reward_std_all": 0.0757555365562439,
"rewards-max/forensics_tempsamp_iou_reward": 0.8984314203262329,
"rewards/forensics_tempsamp_iou_reward": 0.5227924883365631,
"step": 675
},
{
"completion_length": 20.203125,
"epoch": 2.8284518828451883,
"grad_norm": 4.448201656341553,
"kl": 0.20556640625,
"learning_rate": 2.939330543933054e-07,
"loss": 0.0082,
"reward": 0.8943449854850769,
"reward_std": 0.04770939610898495,
"reward_std_all": 0.05902347341179848,
"rewards-max/forensics_tempsamp_iou_reward": 0.903565376996994,
"rewards/forensics_tempsamp_iou_reward": 0.6318449974060059,
"step": 676
},
{
"completion_length": 21.625,
"epoch": 2.8326359832635983,
"grad_norm": 5.329812049865723,
"kl": 0.212890625,
"learning_rate": 2.928870292887029e-07,
"loss": 0.0085,
"reward": 0.7921278774738312,
"reward_std": 0.15443918108940125,
"reward_std_all": 0.17045144364237785,
"rewards-max/forensics_tempsamp_iou_reward": 0.9001910388469696,
"rewards/forensics_tempsamp_iou_reward": 0.5296278297901154,
"step": 677
},
{
"completion_length": 20.265625,
"epoch": 2.8368200836820083,
"grad_norm": 5.359128952026367,
"kl": 0.2138671875,
"learning_rate": 2.9184100418410043e-07,
"loss": 0.0085,
"reward": 0.6792440414428711,
"reward_std": 0.13021454215049744,
"reward_std_all": 0.131755530834198,
"rewards-max/forensics_tempsamp_iou_reward": 0.9230764508247375,
"rewards/forensics_tempsamp_iou_reward": 0.47924405336380005,
"step": 678
},
{
"completion_length": 22.0,
"epoch": 2.8410041841004183,
"grad_norm": 8.606685638427734,
"kl": 0.3935546875,
"learning_rate": 2.907949790794979e-07,
"loss": 0.0157,
"reward": 0.7374584376811981,
"reward_std": 0.07957269251346588,
"reward_std_all": 0.08472098223865032,
"rewards-max/forensics_tempsamp_iou_reward": 0.9391049146652222,
"rewards/forensics_tempsamp_iou_reward": 0.49370837211608887,
"step": 679
},
{
"completion_length": 20.359375,
"epoch": 2.8451882845188283,
"grad_norm": 4.6378631591796875,
"kl": 0.22607421875,
"learning_rate": 2.897489539748954e-07,
"loss": 0.009,
"reward": 0.7617925405502319,
"reward_std": 0.09122407622635365,
"reward_std_all": 0.09477687440812588,
"rewards-max/forensics_tempsamp_iou_reward": 0.9039513468742371,
"rewards/forensics_tempsamp_iou_reward": 0.505542516708374,
"step": 680
},
{
"completion_length": 19.703125,
"epoch": 2.8493723849372383,
"grad_norm": 5.1421589851379395,
"kl": 0.15576171875,
"learning_rate": 2.8870292887029286e-07,
"loss": 0.0062,
"reward": 0.6708229184150696,
"reward_std": 0.11410325393080711,
"reward_std_all": 0.11171428859233856,
"rewards-max/forensics_tempsamp_iou_reward": 0.8638278245925903,
"rewards/forensics_tempsamp_iou_reward": 0.44582289457321167,
"step": 681
},
{
"completion_length": 20.53125,
"epoch": 2.8535564853556483,
"grad_norm": 5.024061679840088,
"kl": 0.18115234375,
"learning_rate": 2.8765690376569037e-07,
"loss": 0.0072,
"reward": 0.8894773423671722,
"reward_std": 0.11371540278196335,
"reward_std_all": 0.10908025875687599,
"rewards-max/forensics_tempsamp_iou_reward": 0.942296028137207,
"rewards/forensics_tempsamp_iou_reward": 0.6144773662090302,
"step": 682
},
{
"completion_length": 20.890625,
"epoch": 2.8577405857740583,
"grad_norm": 6.196515083312988,
"kl": 0.345703125,
"learning_rate": 2.8661087866108783e-07,
"loss": 0.0138,
"reward": 0.7909137308597565,
"reward_std": 0.1310274563729763,
"reward_std_all": 0.11746442317962646,
"rewards-max/forensics_tempsamp_iou_reward": 0.9188385009765625,
"rewards/forensics_tempsamp_iou_reward": 0.5534137487411499,
"step": 683
},
{
"completion_length": 20.296875,
"epoch": 2.8619246861924688,
"grad_norm": 5.27137565612793,
"kl": 0.18408203125,
"learning_rate": 2.8556485355648535e-07,
"loss": 0.0074,
"reward": 0.7997726798057556,
"reward_std": 0.08165958896279335,
"reward_std_all": 0.08775199577212334,
"rewards-max/forensics_tempsamp_iou_reward": 0.9313131272792816,
"rewards/forensics_tempsamp_iou_reward": 0.518522709608078,
"step": 684
},
{
"completion_length": 19.828125,
"epoch": 2.8661087866108788,
"grad_norm": 5.087715148925781,
"kl": 0.17529296875,
"learning_rate": 2.8451882845188286e-07,
"loss": 0.007,
"reward": 0.8373432159423828,
"reward_std": 0.06878370232880116,
"reward_std_all": 0.07628358900547028,
"rewards-max/forensics_tempsamp_iou_reward": 0.9379313588142395,
"rewards/forensics_tempsamp_iou_reward": 0.5623432695865631,
"step": 685
},
{
"completion_length": 20.4375,
"epoch": 2.8702928870292888,
"grad_norm": 4.829983234405518,
"kl": 0.1591796875,
"learning_rate": 2.834728033472803e-07,
"loss": 0.0064,
"reward": 0.8639207780361176,
"reward_std": 0.0892842598259449,
"reward_std_all": 0.0863950103521347,
"rewards-max/forensics_tempsamp_iou_reward": 0.9349645972251892,
"rewards/forensics_tempsamp_iou_reward": 0.626420795917511,
"step": 686
},
{
"completion_length": 19.6875,
"epoch": 2.8744769874476988,
"grad_norm": 7.180575370788574,
"kl": 0.21142578125,
"learning_rate": 2.824267782426778e-07,
"loss": 0.0084,
"reward": 0.8784083127975464,
"reward_std": 0.08926041424274445,
"reward_std_all": 0.0827992632985115,
"rewards-max/forensics_tempsamp_iou_reward": 0.9280132353305817,
"rewards/forensics_tempsamp_iou_reward": 0.6159083545207977,
"step": 687
},
{
"completion_length": 21.734375,
"epoch": 2.8786610878661087,
"grad_norm": 9.717437744140625,
"kl": 0.1650390625,
"learning_rate": 2.813807531380753e-07,
"loss": 0.0066,
"reward": 0.8028751015663147,
"reward_std": 0.1297800987958908,
"reward_std_all": 0.1440691128373146,
"rewards-max/forensics_tempsamp_iou_reward": 0.9385727047920227,
"rewards/forensics_tempsamp_iou_reward": 0.5591250956058502,
"step": 688
},
{
"completion_length": 19.671875,
"epoch": 2.8828451882845187,
"grad_norm": 4.042797088623047,
"kl": 0.18212890625,
"learning_rate": 2.803347280334728e-07,
"loss": 0.0073,
"reward": 0.7660133242607117,
"reward_std": 0.052032630890607834,
"reward_std_all": 0.05342792719602585,
"rewards-max/forensics_tempsamp_iou_reward": 0.9503845274448395,
"rewards/forensics_tempsamp_iou_reward": 0.5222632884979248,
"step": 689
},
{
"completion_length": 21.5,
"epoch": 2.8870292887029287,
"grad_norm": 5.120216369628906,
"kl": 0.21240234375,
"learning_rate": 2.792887029288703e-07,
"loss": 0.0085,
"reward": 0.5288089215755463,
"reward_std": 0.09583986736834049,
"reward_std_all": 0.0998966172337532,
"rewards-max/forensics_tempsamp_iou_reward": 0.9719983637332916,
"rewards/forensics_tempsamp_iou_reward": 0.328808918595314,
"step": 690
},
{
"completion_length": 21.78125,
"epoch": 2.891213389121339,
"grad_norm": 7.655664443969727,
"kl": 0.271484375,
"learning_rate": 2.782426778242677e-07,
"loss": 0.0108,
"reward": 0.7209093868732452,
"reward_std": 0.08090070635080338,
"reward_std_all": 0.0878806784749031,
"rewards-max/forensics_tempsamp_iou_reward": 0.9203901290893555,
"rewards/forensics_tempsamp_iou_reward": 0.47715936601161957,
"step": 691
},
{
"completion_length": 20.765625,
"epoch": 2.895397489539749,
"grad_norm": 5.8181891441345215,
"kl": 0.26123046875,
"learning_rate": 2.7719665271966524e-07,
"loss": 0.0104,
"reward": 0.9000170528888702,
"reward_std": 0.09588447958230972,
"reward_std_all": 0.09411587938666344,
"rewards-max/forensics_tempsamp_iou_reward": 0.9593351483345032,
"rewards/forensics_tempsamp_iou_reward": 0.6375170350074768,
"step": 692
},
{
"completion_length": 19.5625,
"epoch": 2.899581589958159,
"grad_norm": 5.789057731628418,
"kl": 0.18798828125,
"learning_rate": 2.7615062761506275e-07,
"loss": 0.0075,
"reward": 0.5565265864133835,
"reward_std": 0.0925302542746067,
"reward_std_all": 0.08289150893688202,
"rewards-max/forensics_tempsamp_iou_reward": 0.7644661068916321,
"rewards/forensics_tempsamp_iou_reward": 0.3440265953540802,
"step": 693
},
{
"completion_length": 20.671875,
"epoch": 2.903765690376569,
"grad_norm": 5.704459190368652,
"kl": 0.32421875,
"learning_rate": 2.7510460251046026e-07,
"loss": 0.0129,
"reward": 0.7902655303478241,
"reward_std": 0.10102225467562675,
"reward_std_all": 0.09907383471727371,
"rewards-max/forensics_tempsamp_iou_reward": 0.918194442987442,
"rewards/forensics_tempsamp_iou_reward": 0.5402655303478241,
"step": 694
},
{
"completion_length": 19.859375,
"epoch": 2.907949790794979,
"grad_norm": 4.814507961273193,
"kl": 0.3564453125,
"learning_rate": 2.740585774058578e-07,
"loss": 0.0142,
"reward": 0.8664440810680389,
"reward_std": 0.07617324590682983,
"reward_std_all": 0.12271608412265778,
"rewards-max/forensics_tempsamp_iou_reward": 0.9706824421882629,
"rewards/forensics_tempsamp_iou_reward": 0.6039440631866455,
"step": 695
},
{
"completion_length": 20.171875,
"epoch": 2.912133891213389,
"grad_norm": 6.204622268676758,
"kl": 0.2412109375,
"learning_rate": 2.730125523012552e-07,
"loss": 0.0096,
"reward": 0.9463446140289307,
"reward_std": 0.07390572316944599,
"reward_std_all": 0.08129706047475338,
"rewards-max/forensics_tempsamp_iou_reward": 0.9437605142593384,
"rewards/forensics_tempsamp_iou_reward": 0.6525946259498596,
"step": 696
},
{
"completion_length": 21.09375,
"epoch": 2.916317991631799,
"grad_norm": 5.802903652191162,
"kl": 0.2255859375,
"learning_rate": 2.719665271966527e-07,
"loss": 0.009,
"reward": 0.5980451107025146,
"reward_std": 0.07963968068361282,
"reward_std_all": 0.07547495886683464,
"rewards-max/forensics_tempsamp_iou_reward": 0.794983983039856,
"rewards/forensics_tempsamp_iou_reward": 0.38554510474205017,
"step": 697
},
{
"completion_length": 19.71875,
"epoch": 2.920502092050209,
"grad_norm": 6.234683513641357,
"kl": 0.22802734375,
"learning_rate": 2.709205020920502e-07,
"loss": 0.0091,
"reward": 0.8451353907585144,
"reward_std": 0.11079662293195724,
"reward_std_all": 0.12119964510202408,
"rewards-max/forensics_tempsamp_iou_reward": 0.9536739587783813,
"rewards/forensics_tempsamp_iou_reward": 0.5888854265213013,
"step": 698
},
{
"completion_length": 21.171875,
"epoch": 2.924686192468619,
"grad_norm": 5.7004194259643555,
"kl": 0.18505859375,
"learning_rate": 2.698744769874477e-07,
"loss": 0.0074,
"reward": 0.9438608586788177,
"reward_std": 0.10949746519327164,
"reward_std_all": 0.10750070959329605,
"rewards-max/forensics_tempsamp_iou_reward": 0.9719131588935852,
"rewards/forensics_tempsamp_iou_reward": 0.6688607931137085,
"step": 699
},
{
"completion_length": 20.0,
"epoch": 2.928870292887029,
"grad_norm": 6.923460960388184,
"kl": 0.26708984375,
"learning_rate": 2.688284518828452e-07,
"loss": 0.0107,
"reward": 0.8327723145484924,
"reward_std": 0.1544271856546402,
"reward_std_all": 0.14830826222896576,
"rewards-max/forensics_tempsamp_iou_reward": 0.930269181728363,
"rewards/forensics_tempsamp_iou_reward": 0.582772359251976,
"step": 700
},
{
"completion_length": 22.734375,
"epoch": 2.933054393305439,
"grad_norm": 4.861824035644531,
"kl": 0.21630859375,
"learning_rate": 2.6778242677824264e-07,
"loss": 0.0087,
"reward": 0.7729401290416718,
"reward_std": 0.08518625237047672,
"reward_std_all": 0.07869824022054672,
"rewards-max/forensics_tempsamp_iou_reward": 0.9584306180477142,
"rewards/forensics_tempsamp_iou_reward": 0.5479401499032974,
"step": 701
},
{
"completion_length": 20.21875,
"epoch": 2.937238493723849,
"grad_norm": 4.143327713012695,
"kl": 0.189453125,
"learning_rate": 2.6673640167364015e-07,
"loss": 0.0076,
"reward": 0.8248704373836517,
"reward_std": 0.05236225761473179,
"reward_std_all": 0.06903576478362083,
"rewards-max/forensics_tempsamp_iou_reward": 0.9852542281150818,
"rewards/forensics_tempsamp_iou_reward": 0.5623704195022583,
"step": 702
},
{
"completion_length": 20.296875,
"epoch": 2.941422594142259,
"grad_norm": 6.271547317504883,
"kl": 0.20751953125,
"learning_rate": 2.6569037656903767e-07,
"loss": 0.0083,
"reward": 0.9084902703762054,
"reward_std": 0.06739876046776772,
"reward_std_all": 0.09681582450866699,
"rewards-max/forensics_tempsamp_iou_reward": 0.9301114976406097,
"rewards/forensics_tempsamp_iou_reward": 0.6272402405738831,
"step": 703
},
{
"completion_length": 19.21875,
"epoch": 2.9456066945606696,
"grad_norm": 5.853221416473389,
"kl": 0.20166015625,
"learning_rate": 2.6464435146443513e-07,
"loss": 0.0081,
"reward": 0.7933639883995056,
"reward_std": 0.07602764293551445,
"reward_std_all": 0.0970185399055481,
"rewards-max/forensics_tempsamp_iou_reward": 0.8581653833389282,
"rewards/forensics_tempsamp_iou_reward": 0.5183639824390411,
"step": 704
},
{
"completion_length": 20.328125,
"epoch": 2.9497907949790796,
"grad_norm": 5.773459434509277,
"kl": 0.18603515625,
"learning_rate": 2.6359832635983264e-07,
"loss": 0.0074,
"reward": 0.7987222969532013,
"reward_std": 0.08778952807188034,
"reward_std_all": 0.09454626590013504,
"rewards-max/forensics_tempsamp_iou_reward": 0.9806002080440521,
"rewards/forensics_tempsamp_iou_reward": 0.5487222671508789,
"step": 705
},
{
"completion_length": 22.421875,
"epoch": 2.9539748953974896,
"grad_norm": 6.463435649871826,
"kl": 0.24609375,
"learning_rate": 2.625523012552301e-07,
"loss": 0.0099,
"reward": 0.6932802796363831,
"reward_std": 0.08957156166434288,
"reward_std_all": 0.10846394300460815,
"rewards-max/forensics_tempsamp_iou_reward": 0.9643307030200958,
"rewards/forensics_tempsamp_iou_reward": 0.4745302349328995,
"step": 706
},
{
"completion_length": 19.25,
"epoch": 2.9581589958158996,
"grad_norm": 6.46225118637085,
"kl": 0.197265625,
"learning_rate": 2.615062761506276e-07,
"loss": 0.0079,
"reward": 0.8364826142787933,
"reward_std": 0.07196669653058052,
"reward_std_all": 0.07731226459145546,
"rewards-max/forensics_tempsamp_iou_reward": 0.912805050611496,
"rewards/forensics_tempsamp_iou_reward": 0.5927326679229736,
"step": 707
},
{
"completion_length": 19.40625,
"epoch": 2.9623430962343096,
"grad_norm": 6.942245960235596,
"kl": 0.16943359375,
"learning_rate": 2.6046025104602507e-07,
"loss": 0.0068,
"reward": 0.8107547163963318,
"reward_std": 0.11749846488237381,
"reward_std_all": 0.12125910073518753,
"rewards-max/forensics_tempsamp_iou_reward": 0.9387772083282471,
"rewards/forensics_tempsamp_iou_reward": 0.5732547640800476,
"step": 708
},
{
"completion_length": 21.953125,
"epoch": 2.9665271966527196,
"grad_norm": 4.652448654174805,
"kl": 0.16552734375,
"learning_rate": 2.594142259414226e-07,
"loss": 0.0066,
"reward": 0.9375881254673004,
"reward_std": 0.08257414028048515,
"reward_std_all": 0.09764857217669487,
"rewards-max/forensics_tempsamp_iou_reward": 0.9886102378368378,
"rewards/forensics_tempsamp_iou_reward": 0.6750881373882294,
"step": 709
},
{
"completion_length": 19.84375,
"epoch": 2.9707112970711296,
"grad_norm": 7.108526229858398,
"kl": 0.1787109375,
"learning_rate": 2.583682008368201e-07,
"loss": 0.0071,
"reward": 0.7864548563957214,
"reward_std": 0.12511669844388962,
"reward_std_all": 0.13195226714015007,
"rewards-max/forensics_tempsamp_iou_reward": 0.9114585518836975,
"rewards/forensics_tempsamp_iou_reward": 0.5302048623561859,
"step": 710
},
{
"completion_length": 20.171875,
"epoch": 2.97489539748954,
"grad_norm": 5.77980899810791,
"kl": 0.19775390625,
"learning_rate": 2.5732217573221756e-07,
"loss": 0.0079,
"reward": 0.9101278781890869,
"reward_std": 0.06364578381180763,
"reward_std_all": 0.09150517545640469,
"rewards-max/forensics_tempsamp_iou_reward": 0.9543131589889526,
"rewards/forensics_tempsamp_iou_reward": 0.6288778781890869,
"step": 711
},
{
"completion_length": 20.53125,
"epoch": 2.97907949790795,
"grad_norm": 5.0295329093933105,
"kl": 0.20361328125,
"learning_rate": 2.56276150627615e-07,
"loss": 0.0081,
"reward": 0.8207555711269379,
"reward_std": 0.10951259732246399,
"reward_std_all": 0.09521184489130974,
"rewards-max/forensics_tempsamp_iou_reward": 0.9537962079048157,
"rewards/forensics_tempsamp_iou_reward": 0.558255523443222,
"step": 712
},
{
"completion_length": 20.375,
"epoch": 2.98326359832636,
"grad_norm": 4.987839221954346,
"kl": 0.197265625,
"learning_rate": 2.5523012552301253e-07,
"loss": 0.0079,
"reward": 0.9361535310745239,
"reward_std": 0.058900900185108185,
"reward_std_all": 0.07805773802101612,
"rewards-max/forensics_tempsamp_iou_reward": 0.9383465647697449,
"rewards/forensics_tempsamp_iou_reward": 0.6549035310745239,
"step": 713
},
{
"completion_length": 19.84375,
"epoch": 2.98744769874477,
"grad_norm": 5.199387550354004,
"kl": 0.18359375,
"learning_rate": 2.5418410041841004e-07,
"loss": 0.0074,
"reward": 0.866901159286499,
"reward_std": 0.05953018553555012,
"reward_std_all": 0.06067510508000851,
"rewards-max/forensics_tempsamp_iou_reward": 0.9385381639003754,
"rewards/forensics_tempsamp_iou_reward": 0.585651159286499,
"step": 714
},
{
"completion_length": 21.671875,
"epoch": 2.99163179916318,
"grad_norm": 4.446652412414551,
"kl": 0.267578125,
"learning_rate": 2.5313807531380756e-07,
"loss": 0.0107,
"reward": 0.7073298692703247,
"reward_std": 0.15546973049640656,
"reward_std_all": 0.14854849874973297,
"rewards-max/forensics_tempsamp_iou_reward": 0.9701530635356903,
"rewards/forensics_tempsamp_iou_reward": 0.46982985734939575,
"step": 715
},
{
"completion_length": 20.59375,
"epoch": 2.99581589958159,
"grad_norm": 8.857404708862305,
"kl": 0.15673828125,
"learning_rate": 2.5209205020920496e-07,
"loss": 0.0063,
"reward": 0.7630570232868195,
"reward_std": 0.11868229508399963,
"reward_std_all": 0.10827688500285149,
"rewards-max/forensics_tempsamp_iou_reward": 0.964849054813385,
"rewards/forensics_tempsamp_iou_reward": 0.531807005405426,
"step": 716
},
{
"completion_length": 18.57142925262451,
"epoch": 3.0,
"grad_norm": 5.361730575561523,
"kl": 0.1943359375,
"learning_rate": 2.510460251046025e-07,
"loss": 0.0075,
"reward": 0.7199587821960449,
"reward_std": 0.09415345825254917,
"reward_std_all": 0.08498435188084841,
"rewards-max/forensics_tempsamp_iou_reward": 0.5665490627288818,
"rewards/forensics_tempsamp_iou_reward": 0.4413873255252838,
"step": 717
},
{
"completion_length": 20.25,
"epoch": 3.00418410041841,
"grad_norm": 6.056734085083008,
"kl": 0.197265625,
"learning_rate": 2.5e-07,
"loss": 0.0079,
"reward": 0.703175812959671,
"reward_std": 0.07657746225595474,
"reward_std_all": 0.07231422327458858,
"rewards-max/forensics_tempsamp_iou_reward": 0.8118607699871063,
"rewards/forensics_tempsamp_iou_reward": 0.46567580103874207,
"step": 718
},
{
"completion_length": 20.34375,
"epoch": 3.00836820083682,
"grad_norm": 4.9904351234436035,
"kl": 0.2841796875,
"learning_rate": 2.489539748953975e-07,
"loss": 0.0114,
"reward": 0.845964640378952,
"reward_std": 0.08386020734906197,
"reward_std_all": 0.08575496450066566,
"rewards-max/forensics_tempsamp_iou_reward": 0.9191222190856934,
"rewards/forensics_tempsamp_iou_reward": 0.583464652299881,
"step": 719
},
{
"completion_length": 20.25,
"epoch": 3.01255230125523,
"grad_norm": 4.4714837074279785,
"kl": 0.22412109375,
"learning_rate": 2.4790794979079496e-07,
"loss": 0.009,
"reward": 0.9670320153236389,
"reward_std": 0.0682367067784071,
"reward_std_all": 0.06651223078370094,
"rewards-max/forensics_tempsamp_iou_reward": 0.9264285862445831,
"rewards/forensics_tempsamp_iou_reward": 0.6857819557189941,
"step": 720
},
{
"completion_length": 19.8125,
"epoch": 3.01673640167364,
"grad_norm": 3.9778921604156494,
"kl": 0.30517578125,
"learning_rate": 2.468619246861925e-07,
"loss": 0.0122,
"reward": 0.6773166954517365,
"reward_std": 0.04791809991002083,
"reward_std_all": 0.05028712935745716,
"rewards-max/forensics_tempsamp_iou_reward": 0.871739000082016,
"rewards/forensics_tempsamp_iou_reward": 0.4710666537284851,
"step": 721
},
{
"completion_length": 21.046875,
"epoch": 3.02092050209205,
"grad_norm": 6.777620315551758,
"kl": 0.2646484375,
"learning_rate": 2.4581589958158993e-07,
"loss": 0.0106,
"reward": 0.9042037427425385,
"reward_std": 0.07581309974193573,
"reward_std_all": 0.07276582717895508,
"rewards-max/forensics_tempsamp_iou_reward": 0.9878203272819519,
"rewards/forensics_tempsamp_iou_reward": 0.660453736782074,
"step": 722
},
{
"completion_length": 19.359375,
"epoch": 3.0251046025104604,
"grad_norm": 11.469575881958008,
"kl": 0.19970703125,
"learning_rate": 2.4476987447698745e-07,
"loss": 0.008,
"reward": 0.8620355427265167,
"reward_std": 0.06453295424580574,
"reward_std_all": 0.06886700540781021,
"rewards-max/forensics_tempsamp_iou_reward": 0.9444900155067444,
"rewards/forensics_tempsamp_iou_reward": 0.5807855427265167,
"step": 723
},
{
"completion_length": 19.28125,
"epoch": 3.0292887029288704,
"grad_norm": 5.186528205871582,
"kl": 0.18408203125,
"learning_rate": 2.4372384937238496e-07,
"loss": 0.0073,
"reward": 0.7708061039447784,
"reward_std": 0.08190446346998215,
"reward_std_all": 0.10528890043497086,
"rewards-max/forensics_tempsamp_iou_reward": 0.9612206518650055,
"rewards/forensics_tempsamp_iou_reward": 0.5270560681819916,
"step": 724
},
{
"completion_length": 22.625,
"epoch": 3.0334728033472804,
"grad_norm": 5.074063777923584,
"kl": 0.19873046875,
"learning_rate": 2.426778242677824e-07,
"loss": 0.0079,
"reward": 0.9350147545337677,
"reward_std": 0.10842945985496044,
"reward_std_all": 0.11500584706664085,
"rewards-max/forensics_tempsamp_iou_reward": 0.9784989058971405,
"rewards/forensics_tempsamp_iou_reward": 0.6350147724151611,
"step": 725
},
{
"completion_length": 21.859375,
"epoch": 3.0376569037656904,
"grad_norm": 5.220991611480713,
"kl": 0.3046875,
"learning_rate": 2.4163179916317993e-07,
"loss": 0.0122,
"reward": 0.726564884185791,
"reward_std": 0.1494440734386444,
"reward_std_all": 0.14544426649808884,
"rewards-max/forensics_tempsamp_iou_reward": 0.9779797196388245,
"rewards/forensics_tempsamp_iou_reward": 0.4765648692846298,
"step": 726
},
{
"completion_length": 20.734375,
"epoch": 3.0418410041841004,
"grad_norm": 7.556947708129883,
"kl": 0.23095703125,
"learning_rate": 2.405857740585774e-07,
"loss": 0.0093,
"reward": 0.8769797086715698,
"reward_std": 0.04629150778055191,
"reward_std_all": 0.05436225049197674,
"rewards-max/forensics_tempsamp_iou_reward": 0.8968475461006165,
"rewards/forensics_tempsamp_iou_reward": 0.5957297384738922,
"step": 727
},
{
"completion_length": 20.890625,
"epoch": 3.0460251046025104,
"grad_norm": 8.094561576843262,
"kl": 0.2236328125,
"learning_rate": 2.395397489539749e-07,
"loss": 0.0089,
"reward": 0.790871262550354,
"reward_std": 0.05956508778035641,
"reward_std_all": 0.06013432890176773,
"rewards-max/forensics_tempsamp_iou_reward": 0.9660676717758179,
"rewards/forensics_tempsamp_iou_reward": 0.553371250629425,
"step": 728
},
{
"completion_length": 19.5625,
"epoch": 3.0502092050209204,
"grad_norm": 5.002279281616211,
"kl": 0.23291015625,
"learning_rate": 2.3849372384937237e-07,
"loss": 0.0093,
"reward": 0.9204050004482269,
"reward_std": 0.05883530154824257,
"reward_std_all": 0.08891299366950989,
"rewards-max/forensics_tempsamp_iou_reward": 0.9428889751434326,
"rewards/forensics_tempsamp_iou_reward": 0.6391550302505493,
"step": 729
},
{
"completion_length": 22.4375,
"epoch": 3.0543933054393304,
"grad_norm": 5.1088385581970215,
"kl": 0.23583984375,
"learning_rate": 2.3744769874476985e-07,
"loss": 0.0095,
"reward": 0.9221574664115906,
"reward_std": 0.0974840335547924,
"reward_std_all": 0.09944169223308563,
"rewards-max/forensics_tempsamp_iou_reward": 0.9521090686321259,
"rewards/forensics_tempsamp_iou_reward": 0.6284074187278748,
"step": 730
},
{
"completion_length": 20.359375,
"epoch": 3.0585774058577404,
"grad_norm": 4.8770036697387695,
"kl": 0.3310546875,
"learning_rate": 2.3640167364016736e-07,
"loss": 0.0132,
"reward": 0.8051502108573914,
"reward_std": 0.09055405668914318,
"reward_std_all": 0.10071960091590881,
"rewards-max/forensics_tempsamp_iou_reward": 0.9081403911113739,
"rewards/forensics_tempsamp_iou_reward": 0.5614001601934433,
"step": 731
},
{
"completion_length": 20.09375,
"epoch": 3.062761506276151,
"grad_norm": 5.8657355308532715,
"kl": 0.26416015625,
"learning_rate": 2.3535564853556482e-07,
"loss": 0.0106,
"reward": 0.8161037564277649,
"reward_std": 0.11244608834385872,
"reward_std_all": 0.10733787342905998,
"rewards-max/forensics_tempsamp_iou_reward": 0.8941503763198853,
"rewards/forensics_tempsamp_iou_reward": 0.5598537623882294,
"step": 732
},
{
"completion_length": 20.5,
"epoch": 3.066945606694561,
"grad_norm": 5.04122257232666,
"kl": 0.20703125,
"learning_rate": 2.3430962343096234e-07,
"loss": 0.0083,
"reward": 0.7088159322738647,
"reward_std": 0.06820591352880001,
"reward_std_all": 0.07030550017952919,
"rewards-max/forensics_tempsamp_iou_reward": 0.9050395488739014,
"rewards/forensics_tempsamp_iou_reward": 0.45881590247154236,
"step": 733
},
{
"completion_length": 20.8125,
"epoch": 3.071129707112971,
"grad_norm": 4.007349014282227,
"kl": 0.23974609375,
"learning_rate": 2.3326359832635982e-07,
"loss": 0.0096,
"reward": 0.8275973200798035,
"reward_std": 0.12072183191776276,
"reward_std_all": 0.11189211159944534,
"rewards-max/forensics_tempsamp_iou_reward": 0.9520366489887238,
"rewards/forensics_tempsamp_iou_reward": 0.6025973260402679,
"step": 734
},
{
"completion_length": 21.03125,
"epoch": 3.075313807531381,
"grad_norm": 7.96356725692749,
"kl": 0.2060546875,
"learning_rate": 2.322175732217573e-07,
"loss": 0.0082,
"reward": 0.9876367449760437,
"reward_std": 0.05319070257246494,
"reward_std_all": 0.07044857367873192,
"rewards-max/forensics_tempsamp_iou_reward": 0.9748293161392212,
"rewards/forensics_tempsamp_iou_reward": 0.68763667345047,
"step": 735
},
{
"completion_length": 19.15625,
"epoch": 3.079497907949791,
"grad_norm": 4.377989292144775,
"kl": 0.205078125,
"learning_rate": 2.311715481171548e-07,
"loss": 0.0082,
"reward": 0.8890906572341919,
"reward_std": 0.07735788077116013,
"reward_std_all": 0.07544879242777824,
"rewards-max/forensics_tempsamp_iou_reward": 0.9532166421413422,
"rewards/forensics_tempsamp_iou_reward": 0.614090621471405,
"step": 736
},
{
"completion_length": 21.0625,
"epoch": 3.083682008368201,
"grad_norm": 5.190249919891357,
"kl": 0.16943359375,
"learning_rate": 2.3012552301255228e-07,
"loss": 0.0068,
"reward": 0.7381824851036072,
"reward_std": 0.07101620361208916,
"reward_std_all": 0.06756236776709557,
"rewards-max/forensics_tempsamp_iou_reward": 0.9865187704563141,
"rewards/forensics_tempsamp_iou_reward": 0.5069324672222137,
"step": 737
},
{
"completion_length": 22.25,
"epoch": 3.087866108786611,
"grad_norm": 6.038300514221191,
"kl": 0.18408203125,
"learning_rate": 2.2907949790794977e-07,
"loss": 0.0074,
"reward": 0.8015361428260803,
"reward_std": 0.0500495508313179,
"reward_std_all": 0.04928083345293999,
"rewards-max/forensics_tempsamp_iou_reward": 0.9438438415527344,
"rewards/forensics_tempsamp_iou_reward": 0.5202861130237579,
"step": 738
},
{
"completion_length": 20.328125,
"epoch": 3.092050209205021,
"grad_norm": 12.062382698059082,
"kl": 0.1328125,
"learning_rate": 2.2803347280334728e-07,
"loss": 0.0053,
"reward": 0.9215608835220337,
"reward_std": 0.091040950268507,
"reward_std_all": 0.11010073125362396,
"rewards-max/forensics_tempsamp_iou_reward": 0.9708083868026733,
"rewards/forensics_tempsamp_iou_reward": 0.6403108537197113,
"step": 739
},
{
"completion_length": 21.9375,
"epoch": 3.096234309623431,
"grad_norm": 6.594792366027832,
"kl": 0.18701171875,
"learning_rate": 2.2698744769874474e-07,
"loss": 0.0075,
"reward": 0.9156423211097717,
"reward_std": 0.10631881095468998,
"reward_std_all": 0.11131194233894348,
"rewards-max/forensics_tempsamp_iou_reward": 0.9696827828884125,
"rewards/forensics_tempsamp_iou_reward": 0.6218923628330231,
"step": 740
},
{
"completion_length": 23.03125,
"epoch": 3.100418410041841,
"grad_norm": 5.875819206237793,
"kl": 0.2880859375,
"learning_rate": 2.2594142259414225e-07,
"loss": 0.0115,
"reward": 0.7594561874866486,
"reward_std": 0.11647286266088486,
"reward_std_all": 0.10765723511576653,
"rewards-max/forensics_tempsamp_iou_reward": 0.9251113533973694,
"rewards/forensics_tempsamp_iou_reward": 0.503206193447113,
"step": 741
},
{
"completion_length": 20.703125,
"epoch": 3.104602510460251,
"grad_norm": 8.71704387664795,
"kl": 0.17333984375,
"learning_rate": 2.2489539748953974e-07,
"loss": 0.0069,
"reward": 0.5728912800550461,
"reward_std": 0.07828766480088234,
"reward_std_all": 0.07346133328974247,
"rewards-max/forensics_tempsamp_iou_reward": 0.9356975257396698,
"rewards/forensics_tempsamp_iou_reward": 0.4166412353515625,
"step": 742
},
{
"completion_length": 23.59375,
"epoch": 3.1087866108786613,
"grad_norm": 6.375553131103516,
"kl": 0.228515625,
"learning_rate": 2.2384937238493723e-07,
"loss": 0.0091,
"reward": 0.6069704592227936,
"reward_std": 0.09228260442614555,
"reward_std_all": 0.090528704226017,
"rewards-max/forensics_tempsamp_iou_reward": 0.9282583296298981,
"rewards/forensics_tempsamp_iou_reward": 0.40697038173675537,
"step": 743
},
{
"completion_length": 21.171875,
"epoch": 3.1129707112970713,
"grad_norm": 5.035192012786865,
"kl": 0.1591796875,
"learning_rate": 2.2280334728033474e-07,
"loss": 0.0064,
"reward": 0.7432001233100891,
"reward_std": 0.13399680703878403,
"reward_std_all": 0.12008876353502274,
"rewards-max/forensics_tempsamp_iou_reward": 0.901996374130249,
"rewards/forensics_tempsamp_iou_reward": 0.5057001113891602,
"step": 744
},
{
"completion_length": 19.875,
"epoch": 3.1171548117154813,
"grad_norm": 5.485283851623535,
"kl": 0.18017578125,
"learning_rate": 2.217573221757322e-07,
"loss": 0.0072,
"reward": 0.8414407670497894,
"reward_std": 0.09788095951080322,
"reward_std_all": 0.10507724434137344,
"rewards-max/forensics_tempsamp_iou_reward": 0.9494464993476868,
"rewards/forensics_tempsamp_iou_reward": 0.6039407551288605,
"step": 745
},
{
"completion_length": 21.703125,
"epoch": 3.1213389121338913,
"grad_norm": 6.052818298339844,
"kl": 0.2412109375,
"learning_rate": 2.207112970711297e-07,
"loss": 0.0096,
"reward": 0.6593214869499207,
"reward_std": 0.09868484362959862,
"reward_std_all": 0.1053774505853653,
"rewards-max/forensics_tempsamp_iou_reward": 0.8301438987255096,
"rewards/forensics_tempsamp_iou_reward": 0.4280714988708496,
"step": 746
},
{
"completion_length": 22.09375,
"epoch": 3.1255230125523012,
"grad_norm": 6.264915466308594,
"kl": 0.1767578125,
"learning_rate": 2.1966527196652717e-07,
"loss": 0.0071,
"reward": 0.7429472506046295,
"reward_std": 0.06600623577833176,
"reward_std_all": 0.06675854325294495,
"rewards-max/forensics_tempsamp_iou_reward": 0.9413203299045563,
"rewards/forensics_tempsamp_iou_reward": 0.486697182059288,
"step": 747
},
{
"completion_length": 22.46875,
"epoch": 3.1297071129707112,
"grad_norm": 7.352994918823242,
"kl": 0.20751953125,
"learning_rate": 2.1861924686192469e-07,
"loss": 0.0083,
"reward": 0.8090952932834625,
"reward_std": 0.0844905748963356,
"reward_std_all": 0.08155005052685738,
"rewards-max/forensics_tempsamp_iou_reward": 0.9706997275352478,
"rewards/forensics_tempsamp_iou_reward": 0.5715952813625336,
"step": 748
},
{
"completion_length": 20.125,
"epoch": 3.1338912133891212,
"grad_norm": 4.5902099609375,
"kl": 0.26123046875,
"learning_rate": 2.1757322175732217e-07,
"loss": 0.0105,
"reward": 0.7823827862739563,
"reward_std": 0.13183855265378952,
"reward_std_all": 0.1186293214559555,
"rewards-max/forensics_tempsamp_iou_reward": 0.9128716886043549,
"rewards/forensics_tempsamp_iou_reward": 0.5198827385902405,
"step": 749
},
{
"completion_length": 21.765625,
"epoch": 3.1380753138075312,
"grad_norm": 5.821239948272705,
"kl": 0.29345703125,
"learning_rate": 2.1652719665271966e-07,
"loss": 0.0117,
"reward": 0.8765433430671692,
"reward_std": 0.07699605077505112,
"reward_std_all": 0.09103529155254364,
"rewards-max/forensics_tempsamp_iou_reward": 0.9578403532505035,
"rewards/forensics_tempsamp_iou_reward": 0.6015432924032211,
"step": 750
},
{
"completion_length": 19.65625,
"epoch": 3.1422594142259412,
"grad_norm": 6.06210994720459,
"kl": 0.2705078125,
"learning_rate": 2.1548117154811714e-07,
"loss": 0.0108,
"reward": 0.9185025095939636,
"reward_std": 0.1196555383503437,
"reward_std_all": 0.11287214234471321,
"rewards-max/forensics_tempsamp_iou_reward": 0.9565050899982452,
"rewards/forensics_tempsamp_iou_reward": 0.6622524857521057,
"step": 751
},
{
"completion_length": 19.828125,
"epoch": 3.1464435146443517,
"grad_norm": 5.97732400894165,
"kl": 0.19287109375,
"learning_rate": 2.1443514644351463e-07,
"loss": 0.0077,
"reward": 0.8300184309482574,
"reward_std": 0.05804705619812012,
"reward_std_all": 0.0898471362888813,
"rewards-max/forensics_tempsamp_iou_reward": 0.9368889927864075,
"rewards/forensics_tempsamp_iou_reward": 0.586268424987793,
"step": 752
},
{
"completion_length": 20.546875,
"epoch": 3.1506276150627617,
"grad_norm": 4.381422519683838,
"kl": 0.19384765625,
"learning_rate": 2.1338912133891212e-07,
"loss": 0.0077,
"reward": 0.903377890586853,
"reward_std": 0.07665875740349293,
"reward_std_all": 0.07349810190498829,
"rewards-max/forensics_tempsamp_iou_reward": 0.9501050412654877,
"rewards/forensics_tempsamp_iou_reward": 0.6033779382705688,
"step": 753
},
{
"completion_length": 20.328125,
"epoch": 3.1548117154811717,
"grad_norm": 6.221925735473633,
"kl": 0.1591796875,
"learning_rate": 2.1234309623430963e-07,
"loss": 0.0064,
"reward": 0.8606559038162231,
"reward_std": 0.048881202936172485,
"reward_std_all": 0.05027577839791775,
"rewards-max/forensics_tempsamp_iou_reward": 0.9380278885364532,
"rewards/forensics_tempsamp_iou_reward": 0.5606559067964554,
"step": 754
},
{
"completion_length": 20.046875,
"epoch": 3.1589958158995817,
"grad_norm": 9.414359092712402,
"kl": 0.3056640625,
"learning_rate": 2.112970711297071e-07,
"loss": 0.0123,
"reward": 0.9784266650676727,
"reward_std": 0.11571022495627403,
"reward_std_all": 0.12305610626935959,
"rewards-max/forensics_tempsamp_iou_reward": 0.9892515540122986,
"rewards/forensics_tempsamp_iou_reward": 0.6846766769886017,
"step": 755
},
{
"completion_length": 19.21875,
"epoch": 3.1631799163179917,
"grad_norm": 7.136001110076904,
"kl": 0.234375,
"learning_rate": 2.102510460251046e-07,
"loss": 0.0094,
"reward": 0.6677416265010834,
"reward_std": 0.07790414057672024,
"reward_std_all": 0.08935071527957916,
"rewards-max/forensics_tempsamp_iou_reward": 0.8424946963787079,
"rewards/forensics_tempsamp_iou_reward": 0.46149155497550964,
"step": 756
},
{
"completion_length": 20.171875,
"epoch": 3.1673640167364017,
"grad_norm": 7.160689353942871,
"kl": 0.25048828125,
"learning_rate": 2.0920502092050206e-07,
"loss": 0.01,
"reward": 0.905765026807785,
"reward_std": 0.08457842096686363,
"reward_std_all": 0.07898369431495667,
"rewards-max/forensics_tempsamp_iou_reward": 0.9812675714492798,
"rewards/forensics_tempsamp_iou_reward": 0.6245149970054626,
"step": 757
},
{
"completion_length": 22.359375,
"epoch": 3.1715481171548117,
"grad_norm": 18.148296356201172,
"kl": 0.6962890625,
"learning_rate": 2.0815899581589958e-07,
"loss": 0.028,
"reward": 0.8101158142089844,
"reward_std": 0.07972484640777111,
"reward_std_all": 0.08169779554009438,
"rewards-max/forensics_tempsamp_iou_reward": 0.9360426366329193,
"rewards/forensics_tempsamp_iou_reward": 0.5476158261299133,
"step": 758
},
{
"completion_length": 20.078125,
"epoch": 3.1757322175732217,
"grad_norm": 6.920512676239014,
"kl": 0.1953125,
"learning_rate": 2.0711297071129706e-07,
"loss": 0.0078,
"reward": 0.6321363747119904,
"reward_std": 0.08189406991004944,
"reward_std_all": 0.114983469247818,
"rewards-max/forensics_tempsamp_iou_reward": 0.8898427188396454,
"rewards/forensics_tempsamp_iou_reward": 0.4196363836526871,
"step": 759
},
{
"completion_length": 20.734375,
"epoch": 3.1799163179916317,
"grad_norm": 6.907042980194092,
"kl": 0.197265625,
"learning_rate": 2.0606694560669455e-07,
"loss": 0.0079,
"reward": 0.7394852936267853,
"reward_std": 0.056560641154646873,
"reward_std_all": 0.05692926608026028,
"rewards-max/forensics_tempsamp_iou_reward": 0.8991656601428986,
"rewards/forensics_tempsamp_iou_reward": 0.4957352578639984,
"step": 760
},
{
"completion_length": 19.8125,
"epoch": 3.1841004184100417,
"grad_norm": 6.005832672119141,
"kl": 0.17529296875,
"learning_rate": 2.0502092050209206e-07,
"loss": 0.007,
"reward": 0.7467350661754608,
"reward_std": 0.11306877061724663,
"reward_std_all": 0.1076328381896019,
"rewards-max/forensics_tempsamp_iou_reward": 0.8814981877803802,
"rewards/forensics_tempsamp_iou_reward": 0.5154851078987122,
"step": 761
},
{
"completion_length": 20.453125,
"epoch": 3.1882845188284517,
"grad_norm": 4.352340221405029,
"kl": 0.2568359375,
"learning_rate": 2.0397489539748952e-07,
"loss": 0.0102,
"reward": 0.9260869920253754,
"reward_std": 0.07819454744458199,
"reward_std_all": 0.08039950579404831,
"rewards-max/forensics_tempsamp_iou_reward": 0.9689799845218658,
"rewards/forensics_tempsamp_iou_reward": 0.6573370695114136,
"step": 762
},
{
"completion_length": 19.3125,
"epoch": 3.192468619246862,
"grad_norm": 8.70594310760498,
"kl": 0.27880859375,
"learning_rate": 2.0292887029288703e-07,
"loss": 0.0112,
"reward": 0.8119338750839233,
"reward_std": 0.08889476582407951,
"reward_std_all": 0.08801047131419182,
"rewards-max/forensics_tempsamp_iou_reward": 0.9314233362674713,
"rewards/forensics_tempsamp_iou_reward": 0.5431838035583496,
"step": 763
},
{
"completion_length": 20.140625,
"epoch": 3.196652719665272,
"grad_norm": 7.321717739105225,
"kl": 0.2041015625,
"learning_rate": 2.0188284518828452e-07,
"loss": 0.0082,
"reward": 0.8465913534164429,
"reward_std": 0.05658190883696079,
"reward_std_all": 0.0686459094285965,
"rewards-max/forensics_tempsamp_iou_reward": 0.9751397371292114,
"rewards/forensics_tempsamp_iou_reward": 0.5965912938117981,
"step": 764
},
{
"completion_length": 19.984375,
"epoch": 3.200836820083682,
"grad_norm": 6.6663641929626465,
"kl": 0.16650390625,
"learning_rate": 2.00836820083682e-07,
"loss": 0.0067,
"reward": 0.7342880368232727,
"reward_std": 0.09130077064037323,
"reward_std_all": 0.09235081449151039,
"rewards-max/forensics_tempsamp_iou_reward": 0.8561607003211975,
"rewards/forensics_tempsamp_iou_reward": 0.5217880606651306,
"step": 765
},
{
"completion_length": 20.75,
"epoch": 3.205020920502092,
"grad_norm": 5.201741695404053,
"kl": 0.2255859375,
"learning_rate": 1.997907949790795e-07,
"loss": 0.009,
"reward": 0.7697692215442657,
"reward_std": 0.07570295035839081,
"reward_std_all": 0.07661169022321701,
"rewards-max/forensics_tempsamp_iou_reward": 0.9129528999328613,
"rewards/forensics_tempsamp_iou_reward": 0.5447691977024078,
"step": 766
},
{
"completion_length": 21.109375,
"epoch": 3.209205020920502,
"grad_norm": 5.284005165100098,
"kl": 0.43359375,
"learning_rate": 1.9874476987447698e-07,
"loss": 0.0173,
"reward": 0.749329149723053,
"reward_std": 0.13423029333353043,
"reward_std_all": 0.1264910027384758,
"rewards-max/forensics_tempsamp_iou_reward": 0.8924019634723663,
"rewards/forensics_tempsamp_iou_reward": 0.49932919442653656,
"step": 767
},
{
"completion_length": 20.34375,
"epoch": 3.213389121338912,
"grad_norm": 9.929222106933594,
"kl": 0.1806640625,
"learning_rate": 1.9769874476987447e-07,
"loss": 0.0072,
"reward": 0.9089229702949524,
"reward_std": 0.08400377258658409,
"reward_std_all": 0.1350308582186699,
"rewards-max/forensics_tempsamp_iou_reward": 0.9591150879859924,
"rewards/forensics_tempsamp_iou_reward": 0.6339228749275208,
"step": 768
},
{
"completion_length": 20.921875,
"epoch": 3.217573221757322,
"grad_norm": 7.449955463409424,
"kl": 0.22412109375,
"learning_rate": 1.9665271966527198e-07,
"loss": 0.009,
"reward": 0.8637288808822632,
"reward_std": 0.12144440039992332,
"reward_std_all": 0.1645219698548317,
"rewards-max/forensics_tempsamp_iou_reward": 0.9342342913150787,
"rewards/forensics_tempsamp_iou_reward": 0.5887289047241211,
"step": 769
},
{
"completion_length": 21.015625,
"epoch": 3.221757322175732,
"grad_norm": 8.551521301269531,
"kl": 0.1826171875,
"learning_rate": 1.9560669456066944e-07,
"loss": 0.0073,
"reward": 0.7702149450778961,
"reward_std": 0.07296909764409065,
"reward_std_all": 0.08229687809944153,
"rewards-max/forensics_tempsamp_iou_reward": 0.931447833776474,
"rewards/forensics_tempsamp_iou_reward": 0.5639649629592896,
"step": 770
},
{
"completion_length": 20.703125,
"epoch": 3.225941422594142,
"grad_norm": 5.065396785736084,
"kl": 0.22216796875,
"learning_rate": 1.9456066945606695e-07,
"loss": 0.0089,
"reward": 0.9210221171379089,
"reward_std": 0.06051299348473549,
"reward_std_all": 0.06313078850507736,
"rewards-max/forensics_tempsamp_iou_reward": 0.9557015895843506,
"rewards/forensics_tempsamp_iou_reward": 0.6585221290588379,
"step": 771
},
{
"completion_length": 19.6875,
"epoch": 3.2301255230125525,
"grad_norm": 5.596981525421143,
"kl": 0.21826171875,
"learning_rate": 1.935146443514644e-07,
"loss": 0.0087,
"reward": 0.8153601586818695,
"reward_std": 0.0984622910618782,
"reward_std_all": 0.09113691560924053,
"rewards-max/forensics_tempsamp_iou_reward": 0.9429479241371155,
"rewards/forensics_tempsamp_iou_reward": 0.559110164642334,
"step": 772
},
{
"completion_length": 20.453125,
"epoch": 3.2343096234309625,
"grad_norm": 7.036488056182861,
"kl": 0.21728515625,
"learning_rate": 1.9246861924686192e-07,
"loss": 0.0087,
"reward": 0.789393424987793,
"reward_std": 0.07022463530302048,
"reward_std_all": 0.07044101506471634,
"rewards-max/forensics_tempsamp_iou_reward": 0.9178018569946289,
"rewards/forensics_tempsamp_iou_reward": 0.5268934369087219,
"step": 773
},
{
"completion_length": 19.734375,
"epoch": 3.2384937238493725,
"grad_norm": 5.659024715423584,
"kl": 0.236328125,
"learning_rate": 1.914225941422594e-07,
"loss": 0.0095,
"reward": 0.9716691076755524,
"reward_std": 0.055400919169187546,
"reward_std_all": 0.0578254759311676,
"rewards-max/forensics_tempsamp_iou_reward": 0.9436567723751068,
"rewards/forensics_tempsamp_iou_reward": 0.671669065952301,
"step": 774
},
{
"completion_length": 20.21875,
"epoch": 3.2426778242677825,
"grad_norm": 7.51660680770874,
"kl": 0.20654296875,
"learning_rate": 1.903765690376569e-07,
"loss": 0.0083,
"reward": 0.8589926660060883,
"reward_std": 0.09055192768573761,
"reward_std_all": 0.11658091098070145,
"rewards-max/forensics_tempsamp_iou_reward": 0.9667032063007355,
"rewards/forensics_tempsamp_iou_reward": 0.6027426421642303,
"step": 775
},
{
"completion_length": 19.75,
"epoch": 3.2468619246861925,
"grad_norm": 7.03537654876709,
"kl": 0.16015625,
"learning_rate": 1.8933054393305438e-07,
"loss": 0.0064,
"reward": 0.9454669058322906,
"reward_std": 0.07916749268770218,
"reward_std_all": 0.08459703624248505,
"rewards-max/forensics_tempsamp_iou_reward": 0.9386895596981049,
"rewards/forensics_tempsamp_iou_reward": 0.6642168760299683,
"step": 776
},
{
"completion_length": 19.234375,
"epoch": 3.2510460251046025,
"grad_norm": 5.070270538330078,
"kl": 0.19287109375,
"learning_rate": 1.8828451882845187e-07,
"loss": 0.0077,
"reward": 0.9119013845920563,
"reward_std": 0.14743946865200996,
"reward_std_all": 0.15193580836057663,
"rewards-max/forensics_tempsamp_iou_reward": 0.9467938542366028,
"rewards/forensics_tempsamp_iou_reward": 0.6306513845920563,
"step": 777
},
{
"completion_length": 20.25,
"epoch": 3.2552301255230125,
"grad_norm": 5.74210262298584,
"kl": 0.17724609375,
"learning_rate": 1.8723849372384936e-07,
"loss": 0.0071,
"reward": 0.6658553183078766,
"reward_std": 0.11327797919511795,
"reward_std_all": 0.13432613760232925,
"rewards-max/forensics_tempsamp_iou_reward": 0.9100730717182159,
"rewards/forensics_tempsamp_iou_reward": 0.46585530042648315,
"step": 778
},
{
"completion_length": 20.890625,
"epoch": 3.2594142259414225,
"grad_norm": 5.806168079376221,
"kl": 0.15283203125,
"learning_rate": 1.8619246861924684e-07,
"loss": 0.0061,
"reward": 0.7966260313987732,
"reward_std": 0.043096983805298805,
"reward_std_all": 0.04853084869682789,
"rewards-max/forensics_tempsamp_iou_reward": 0.97401362657547,
"rewards/forensics_tempsamp_iou_reward": 0.5341260135173798,
"step": 779
},
{
"completion_length": 21.125,
"epoch": 3.2635983263598325,
"grad_norm": 4.122276306152344,
"kl": 0.1884765625,
"learning_rate": 1.8514644351464433e-07,
"loss": 0.0075,
"reward": 0.9713796377182007,
"reward_std": 0.10321496427059174,
"reward_std_all": 0.1464422270655632,
"rewards-max/forensics_tempsamp_iou_reward": 0.9731252491474152,
"rewards/forensics_tempsamp_iou_reward": 0.6776296496391296,
"step": 780
},
{
"completion_length": 21.078125,
"epoch": 3.2677824267782425,
"grad_norm": 8.665406227111816,
"kl": 0.1728515625,
"learning_rate": 1.8410041841004184e-07,
"loss": 0.0069,
"reward": 0.9905488193035126,
"reward_std": 0.09516609460115433,
"reward_std_all": 0.09418881498277187,
"rewards-max/forensics_tempsamp_iou_reward": 0.9845910370349884,
"rewards/forensics_tempsamp_iou_reward": 0.6905488073825836,
"step": 781
},
{
"completion_length": 19.640625,
"epoch": 3.2719665271966525,
"grad_norm": 6.074277877807617,
"kl": 0.22314453125,
"learning_rate": 1.830543933054393e-07,
"loss": 0.0089,
"reward": 0.7536022067070007,
"reward_std": 0.1414812058210373,
"reward_std_all": 0.12642348557710648,
"rewards-max/forensics_tempsamp_iou_reward": 0.9612291157245636,
"rewards/forensics_tempsamp_iou_reward": 0.5223522335290909,
"step": 782
},
{
"completion_length": 18.875,
"epoch": 3.276150627615063,
"grad_norm": 5.411194801330566,
"kl": 0.22900390625,
"learning_rate": 1.8200836820083681e-07,
"loss": 0.0092,
"reward": 0.8095413744449615,
"reward_std": 0.09146954864263535,
"reward_std_all": 0.08262849226593971,
"rewards-max/forensics_tempsamp_iou_reward": 0.9666666686534882,
"rewards/forensics_tempsamp_iou_reward": 0.5595413446426392,
"step": 783
},
{
"completion_length": 21.265625,
"epoch": 3.280334728033473,
"grad_norm": 7.856323719024658,
"kl": 0.4072265625,
"learning_rate": 1.8096234309623427e-07,
"loss": 0.0163,
"reward": 0.7711780965328217,
"reward_std": 0.12817538529634476,
"reward_std_all": 0.11991313844919205,
"rewards-max/forensics_tempsamp_iou_reward": 0.8472331464290619,
"rewards/forensics_tempsamp_iou_reward": 0.5086781084537506,
"step": 784
},
{
"completion_length": 20.953125,
"epoch": 3.284518828451883,
"grad_norm": 8.922173500061035,
"kl": 0.3017578125,
"learning_rate": 1.799163179916318e-07,
"loss": 0.0121,
"reward": 0.7475005984306335,
"reward_std": 0.12746990472078323,
"reward_std_all": 0.12264284119009972,
"rewards-max/forensics_tempsamp_iou_reward": 0.9451344013214111,
"rewards/forensics_tempsamp_iou_reward": 0.5162505507469177,
"step": 785
},
{
"completion_length": 21.0625,
"epoch": 3.288702928870293,
"grad_norm": 6.8719401359558105,
"kl": 0.2802734375,
"learning_rate": 1.788702928870293e-07,
"loss": 0.0112,
"reward": 0.7200064361095428,
"reward_std": 0.1657305434346199,
"reward_std_all": 0.14830468595027924,
"rewards-max/forensics_tempsamp_iou_reward": 0.8989205658435822,
"rewards/forensics_tempsamp_iou_reward": 0.46375641226768494,
"step": 786
},
{
"completion_length": 21.734375,
"epoch": 3.292887029288703,
"grad_norm": 4.821695804595947,
"kl": 0.20458984375,
"learning_rate": 1.7782426778242676e-07,
"loss": 0.0082,
"reward": 0.8252646327018738,
"reward_std": 0.1199355199933052,
"reward_std_all": 0.11364282295107841,
"rewards-max/forensics_tempsamp_iou_reward": 0.9880660772323608,
"rewards/forensics_tempsamp_iou_reward": 0.6065146028995514,
"step": 787
},
{
"completion_length": 20.34375,
"epoch": 3.297071129707113,
"grad_norm": 6.379543781280518,
"kl": 0.18115234375,
"learning_rate": 1.7677824267782427e-07,
"loss": 0.0072,
"reward": 0.8359558582305908,
"reward_std": 0.04706565849483013,
"reward_std_all": 0.053799163550138474,
"rewards-max/forensics_tempsamp_iou_reward": 0.9505093395709991,
"rewards/forensics_tempsamp_iou_reward": 0.5734557807445526,
"step": 788
},
{
"completion_length": 20.171875,
"epoch": 3.301255230125523,
"grad_norm": 5.928640365600586,
"kl": 0.2998046875,
"learning_rate": 1.7573221757322173e-07,
"loss": 0.012,
"reward": 0.830297440290451,
"reward_std": 0.14801708981394768,
"reward_std_all": 0.17267251014709473,
"rewards-max/forensics_tempsamp_iou_reward": 0.8889763951301575,
"rewards/forensics_tempsamp_iou_reward": 0.555297464132309,
"step": 789
},
{
"completion_length": 19.71875,
"epoch": 3.305439330543933,
"grad_norm": 6.751445293426514,
"kl": 0.2421875,
"learning_rate": 1.7468619246861925e-07,
"loss": 0.0097,
"reward": 0.8900960981845856,
"reward_std": 0.09513960406184196,
"reward_std_all": 0.12660599872469902,
"rewards-max/forensics_tempsamp_iou_reward": 0.9262245297431946,
"rewards/forensics_tempsamp_iou_reward": 0.6150960326194763,
"step": 790
},
{
"completion_length": 22.921875,
"epoch": 3.309623430962343,
"grad_norm": 9.745909690856934,
"kl": 0.19287109375,
"learning_rate": 1.7364016736401673e-07,
"loss": 0.0077,
"reward": 0.8933237195014954,
"reward_std": 0.1003534346818924,
"reward_std_all": 0.0919060967862606,
"rewards-max/forensics_tempsamp_iou_reward": 0.9531557559967041,
"rewards/forensics_tempsamp_iou_reward": 0.6058237254619598,
"step": 791
},
{
"completion_length": 21.5,
"epoch": 3.3138075313807533,
"grad_norm": 6.555712699890137,
"kl": 0.23974609375,
"learning_rate": 1.7259414225941422e-07,
"loss": 0.0096,
"reward": 0.8364173769950867,
"reward_std": 0.08648781478404999,
"reward_std_all": 0.10595997422933578,
"rewards-max/forensics_tempsamp_iou_reward": 0.96880704164505,
"rewards/forensics_tempsamp_iou_reward": 0.5551673471927643,
"step": 792
},
{
"completion_length": 19.1875,
"epoch": 3.3179916317991633,
"grad_norm": 6.225279331207275,
"kl": 0.291015625,
"learning_rate": 1.715481171548117e-07,
"loss": 0.0116,
"reward": 0.8402729630470276,
"reward_std": 0.10121035948395729,
"reward_std_all": 0.0971168577671051,
"rewards-max/forensics_tempsamp_iou_reward": 0.9552186131477356,
"rewards/forensics_tempsamp_iou_reward": 0.5652729570865631,
"step": 793
},
{
"completion_length": 21.921875,
"epoch": 3.3221757322175733,
"grad_norm": 4.963297367095947,
"kl": 0.19970703125,
"learning_rate": 1.705020920502092e-07,
"loss": 0.008,
"reward": 0.6710978448390961,
"reward_std": 0.08930085226893425,
"reward_std_all": 0.08030064776539803,
"rewards-max/forensics_tempsamp_iou_reward": 0.9258301854133606,
"rewards/forensics_tempsamp_iou_reward": 0.44609785079956055,
"step": 794
},
{
"completion_length": 22.703125,
"epoch": 3.3263598326359833,
"grad_norm": 6.216242790222168,
"kl": 0.1748046875,
"learning_rate": 1.6945606694560668e-07,
"loss": 0.007,
"reward": 0.9185225069522858,
"reward_std": 0.09104876592755318,
"reward_std_all": 0.08312413096427917,
"rewards-max/forensics_tempsamp_iou_reward": 0.9408210813999176,
"rewards/forensics_tempsamp_iou_reward": 0.6372725069522858,
"step": 795
},
{
"completion_length": 19.59375,
"epoch": 3.3305439330543933,
"grad_norm": 8.423406600952148,
"kl": 0.4130859375,
"learning_rate": 1.684100418410042e-07,
"loss": 0.0165,
"reward": 0.8226988911628723,
"reward_std": 0.10409851372241974,
"reward_std_all": 0.11555466055870056,
"rewards-max/forensics_tempsamp_iou_reward": 0.9551932215690613,
"rewards/forensics_tempsamp_iou_reward": 0.5726988315582275,
"step": 796
},
{
"completion_length": 22.703125,
"epoch": 3.3347280334728033,
"grad_norm": 5.264939308166504,
"kl": 0.23974609375,
"learning_rate": 1.6736401673640165e-07,
"loss": 0.0096,
"reward": 0.7799777984619141,
"reward_std": 0.09941427782177925,
"reward_std_all": 0.11123489588499069,
"rewards-max/forensics_tempsamp_iou_reward": 0.9423867166042328,
"rewards/forensics_tempsamp_iou_reward": 0.4799777716398239,
"step": 797
},
{
"completion_length": 20.46875,
"epoch": 3.3389121338912133,
"grad_norm": 5.100539684295654,
"kl": 0.17138671875,
"learning_rate": 1.6631799163179916e-07,
"loss": 0.0069,
"reward": 1.0450254082679749,
"reward_std": 0.05066349171102047,
"reward_std_all": 0.04641056805849075,
"rewards-max/forensics_tempsamp_iou_reward": 0.9812967479228973,
"rewards/forensics_tempsamp_iou_reward": 0.7450254559516907,
"step": 798
},
{
"completion_length": 20.6875,
"epoch": 3.3430962343096233,
"grad_norm": 4.66460657119751,
"kl": 0.2646484375,
"learning_rate": 1.6527196652719662e-07,
"loss": 0.0106,
"reward": 0.7959193885326385,
"reward_std": 0.09975581616163254,
"reward_std_all": 0.08780565485358238,
"rewards-max/forensics_tempsamp_iou_reward": 0.9133106768131256,
"rewards/forensics_tempsamp_iou_reward": 0.5521694123744965,
"step": 799
},
{
"completion_length": 21.0625,
"epoch": 3.3472803347280333,
"grad_norm": 8.570697784423828,
"kl": 0.2607421875,
"learning_rate": 1.6422594142259414e-07,
"loss": 0.0104,
"reward": 0.9172329604625702,
"reward_std": 0.08995519950985909,
"reward_std_all": 0.08601542562246323,
"rewards-max/forensics_tempsamp_iou_reward": 0.9575392007827759,
"rewards/forensics_tempsamp_iou_reward": 0.6359829306602478,
"step": 800
},
{
"completion_length": 19.5625,
"epoch": 3.3514644351464433,
"grad_norm": 5.142190456390381,
"kl": 0.21533203125,
"learning_rate": 1.6317991631799162e-07,
"loss": 0.0086,
"reward": 0.8083488345146179,
"reward_std": 0.04134682007133961,
"reward_std_all": 0.05283203348517418,
"rewards-max/forensics_tempsamp_iou_reward": 0.9170800745487213,
"rewards/forensics_tempsamp_iou_reward": 0.5458487868309021,
"step": 801
},
{
"completion_length": 20.765625,
"epoch": 3.3556485355648533,
"grad_norm": 7.857107162475586,
"kl": 0.16455078125,
"learning_rate": 1.621338912133891e-07,
"loss": 0.0066,
"reward": 0.7156846225261688,
"reward_std": 0.05759577266871929,
"reward_std_all": 0.05435681901872158,
"rewards-max/forensics_tempsamp_iou_reward": 0.9333920180797577,
"rewards/forensics_tempsamp_iou_reward": 0.5094346404075623,
"step": 802
},
{
"completion_length": 19.03125,
"epoch": 3.3598326359832638,
"grad_norm": 4.674337387084961,
"kl": 0.3134765625,
"learning_rate": 1.610878661087866e-07,
"loss": 0.0126,
"reward": 0.79454705119133,
"reward_std": 0.09533223882317543,
"reward_std_all": 0.11437316611409187,
"rewards-max/forensics_tempsamp_iou_reward": 0.9550455212593079,
"rewards/forensics_tempsamp_iou_reward": 0.5570470988750458,
"step": 803
},
{
"completion_length": 20.8125,
"epoch": 3.3640167364016738,
"grad_norm": 5.619439601898193,
"kl": 0.20703125,
"learning_rate": 1.6004184100418408e-07,
"loss": 0.0083,
"reward": 0.9088247418403625,
"reward_std": 0.09866487979888916,
"reward_std_all": 0.10643739253282547,
"rewards-max/forensics_tempsamp_iou_reward": 0.9203117787837982,
"rewards/forensics_tempsamp_iou_reward": 0.6275746822357178,
"step": 804
},
{
"completion_length": 21.21875,
"epoch": 3.3682008368200838,
"grad_norm": 10.346659660339355,
"kl": 0.40966796875,
"learning_rate": 1.5899581589958157e-07,
"loss": 0.0164,
"reward": 0.7925573885440826,
"reward_std": 0.12542904540896416,
"reward_std_all": 0.11283312365412712,
"rewards-max/forensics_tempsamp_iou_reward": 0.9375704824924469,
"rewards/forensics_tempsamp_iou_reward": 0.5363074094057083,
"step": 805
},
{
"completion_length": 21.15625,
"epoch": 3.3723849372384938,
"grad_norm": 6.313467025756836,
"kl": 0.2021484375,
"learning_rate": 1.5794979079497908e-07,
"loss": 0.0081,
"reward": 0.937359631061554,
"reward_std": 0.09260537847876549,
"reward_std_all": 0.09928587079048157,
"rewards-max/forensics_tempsamp_iou_reward": 0.9266416430473328,
"rewards/forensics_tempsamp_iou_reward": 0.668609619140625,
"step": 806
},
{
"completion_length": 21.171875,
"epoch": 3.3765690376569037,
"grad_norm": 5.5224409103393555,
"kl": 0.20361328125,
"learning_rate": 1.5690376569037657e-07,
"loss": 0.0081,
"reward": 0.8534778356552124,
"reward_std": 0.09768132492899895,
"reward_std_all": 0.09949023649096489,
"rewards-max/forensics_tempsamp_iou_reward": 0.9557310342788696,
"rewards/forensics_tempsamp_iou_reward": 0.5909777879714966,
"step": 807
},
{
"completion_length": 20.0625,
"epoch": 3.3807531380753137,
"grad_norm": 7.895532608032227,
"kl": 0.384765625,
"learning_rate": 1.5585774058577405e-07,
"loss": 0.0154,
"reward": 0.8375965058803558,
"reward_std": 0.09065173752605915,
"reward_std_all": 0.09094279818236828,
"rewards-max/forensics_tempsamp_iou_reward": 0.9587129652500153,
"rewards/forensics_tempsamp_iou_reward": 0.5625964999198914,
"step": 808
},
{
"completion_length": 21.015625,
"epoch": 3.3849372384937237,
"grad_norm": 6.47365665435791,
"kl": 0.3232421875,
"learning_rate": 1.5481171548117154e-07,
"loss": 0.0129,
"reward": 0.9467189610004425,
"reward_std": 0.08931037783622742,
"reward_std_all": 0.0983632504940033,
"rewards-max/forensics_tempsamp_iou_reward": 0.9979381561279297,
"rewards/forensics_tempsamp_iou_reward": 0.6654689311981201,
"step": 809
},
{
"completion_length": 20.125,
"epoch": 3.3891213389121337,
"grad_norm": 5.836177349090576,
"kl": 0.2138671875,
"learning_rate": 1.5376569037656903e-07,
"loss": 0.0086,
"reward": 0.7829257547855377,
"reward_std": 0.1281553991138935,
"reward_std_all": 0.1323682777583599,
"rewards-max/forensics_tempsamp_iou_reward": 0.9413366913795471,
"rewards/forensics_tempsamp_iou_reward": 0.5141757726669312,
"step": 810
},
{
"completion_length": 19.515625,
"epoch": 3.393305439330544,
"grad_norm": 4.9963579177856445,
"kl": 0.1669921875,
"learning_rate": 1.5271966527196654e-07,
"loss": 0.0067,
"reward": 0.8683551847934723,
"reward_std": 0.08011734485626221,
"reward_std_all": 0.0759599432349205,
"rewards-max/forensics_tempsamp_iou_reward": 0.9422879219055176,
"rewards/forensics_tempsamp_iou_reward": 0.5933551788330078,
"step": 811
},
{
"completion_length": 22.09375,
"epoch": 3.397489539748954,
"grad_norm": 4.649418354034424,
"kl": 0.18798828125,
"learning_rate": 1.51673640167364e-07,
"loss": 0.0075,
"reward": 0.8215027153491974,
"reward_std": 0.07159652002155781,
"reward_std_all": 0.06869477592408657,
"rewards-max/forensics_tempsamp_iou_reward": 0.9754155576229095,
"rewards/forensics_tempsamp_iou_reward": 0.5590027272701263,
"step": 812
},
{
"completion_length": 20.828125,
"epoch": 3.401673640167364,
"grad_norm": 10.073711395263672,
"kl": 0.3115234375,
"learning_rate": 1.506276150627615e-07,
"loss": 0.0125,
"reward": 0.61151123046875,
"reward_std": 0.09958121180534363,
"reward_std_all": 0.11781410127878189,
"rewards-max/forensics_tempsamp_iou_reward": 0.9429070949554443,
"rewards/forensics_tempsamp_iou_reward": 0.41151121258735657,
"step": 813
},
{
"completion_length": 20.6875,
"epoch": 3.405857740585774,
"grad_norm": 6.753358840942383,
"kl": 0.25390625,
"learning_rate": 1.4958158995815897e-07,
"loss": 0.0102,
"reward": 0.8634264469146729,
"reward_std": 0.08456418663263321,
"reward_std_all": 0.08632449805736542,
"rewards-max/forensics_tempsamp_iou_reward": 0.954511284828186,
"rewards/forensics_tempsamp_iou_reward": 0.6384264826774597,
"step": 814
},
{
"completion_length": 23.578125,
"epoch": 3.410041841004184,
"grad_norm": 4.468890190124512,
"kl": 0.38671875,
"learning_rate": 1.4853556485355648e-07,
"loss": 0.0155,
"reward": 0.7988404035568237,
"reward_std": 0.12965120375156403,
"reward_std_all": 0.12622442096471786,
"rewards-max/forensics_tempsamp_iou_reward": 0.9527513086795807,
"rewards/forensics_tempsamp_iou_reward": 0.5425904393196106,
"step": 815
},
{
"completion_length": 21.109375,
"epoch": 3.414225941422594,
"grad_norm": 4.342041492462158,
"kl": 0.22900390625,
"learning_rate": 1.4748953974895397e-07,
"loss": 0.0092,
"reward": 1.012256145477295,
"reward_std": 0.10422732308506966,
"reward_std_all": 0.11652003228664398,
"rewards-max/forensics_tempsamp_iou_reward": 0.9730221331119537,
"rewards/forensics_tempsamp_iou_reward": 0.7185061275959015,
"step": 816
},
{
"completion_length": 21.1875,
"epoch": 3.418410041841004,
"grad_norm": 5.495375156402588,
"kl": 0.23828125,
"learning_rate": 1.4644351464435146e-07,
"loss": 0.0095,
"reward": 0.7235157787799835,
"reward_std": 0.14411965012550354,
"reward_std_all": 0.15162212774157524,
"rewards-max/forensics_tempsamp_iou_reward": 0.917025089263916,
"rewards/forensics_tempsamp_iou_reward": 0.4860157072544098,
"step": 817
},
{
"completion_length": 19.5625,
"epoch": 3.422594142259414,
"grad_norm": 4.783437728881836,
"kl": 0.267578125,
"learning_rate": 1.4539748953974894e-07,
"loss": 0.0107,
"reward": 0.6973970830440521,
"reward_std": 0.09564090520143509,
"reward_std_all": 0.09361863136291504,
"rewards-max/forensics_tempsamp_iou_reward": 0.9145216345787048,
"rewards/forensics_tempsamp_iou_reward": 0.48489704728126526,
"step": 818
},
{
"completion_length": 19.796875,
"epoch": 3.426778242677824,
"grad_norm": 5.605738639831543,
"kl": 0.3798828125,
"learning_rate": 1.4435146443514643e-07,
"loss": 0.0152,
"reward": 0.7314243614673615,
"reward_std": 0.12717458978295326,
"reward_std_all": 0.12287924438714981,
"rewards-max/forensics_tempsamp_iou_reward": 0.9550304412841797,
"rewards/forensics_tempsamp_iou_reward": 0.4876743406057358,
"step": 819
},
{
"completion_length": 19.4375,
"epoch": 3.430962343096234,
"grad_norm": 6.134143352508545,
"kl": 0.27392578125,
"learning_rate": 1.4330543933054392e-07,
"loss": 0.011,
"reward": 0.8458690047264099,
"reward_std": 0.04842314310371876,
"reward_std_all": 0.0940762311220169,
"rewards-max/forensics_tempsamp_iou_reward": 0.9681596755981445,
"rewards/forensics_tempsamp_iou_reward": 0.5833689272403717,
"step": 820
},
{
"completion_length": 20.75,
"epoch": 3.435146443514644,
"grad_norm": 5.443774223327637,
"kl": 0.16259765625,
"learning_rate": 1.4225941422594143e-07,
"loss": 0.0065,
"reward": 0.8961964249610901,
"reward_std": 0.10253185778856277,
"reward_std_all": 0.10204831510782242,
"rewards-max/forensics_tempsamp_iou_reward": 0.9390356838703156,
"rewards/forensics_tempsamp_iou_reward": 0.6149463653564453,
"step": 821
},
{
"completion_length": 19.796875,
"epoch": 3.439330543933054,
"grad_norm": 6.646647930145264,
"kl": 0.32958984375,
"learning_rate": 1.412133891213389e-07,
"loss": 0.0132,
"reward": 0.9359737038612366,
"reward_std": 0.12835264205932617,
"reward_std_all": 0.13063885271549225,
"rewards-max/forensics_tempsamp_iou_reward": 0.9810335338115692,
"rewards/forensics_tempsamp_iou_reward": 0.6609737277030945,
"step": 822
},
{
"completion_length": 20.578125,
"epoch": 3.4435146443514646,
"grad_norm": 5.525972366333008,
"kl": 0.18603515625,
"learning_rate": 1.401673640167364e-07,
"loss": 0.0074,
"reward": 0.897067129611969,
"reward_std": 0.07021529972553253,
"reward_std_all": 0.07146213576197624,
"rewards-max/forensics_tempsamp_iou_reward": 0.9546042382717133,
"rewards/forensics_tempsamp_iou_reward": 0.6095671355724335,
"step": 823
},
{
"completion_length": 21.1875,
"epoch": 3.4476987447698746,
"grad_norm": 4.636652946472168,
"kl": 0.2197265625,
"learning_rate": 1.3912133891213386e-07,
"loss": 0.0088,
"reward": 0.6288564205169678,
"reward_std": 0.10212391801178455,
"reward_std_all": 0.12880387902259827,
"rewards-max/forensics_tempsamp_iou_reward": 0.8917259871959686,
"rewards/forensics_tempsamp_iou_reward": 0.3913563936948776,
"step": 824
},
{
"completion_length": 20.765625,
"epoch": 3.4518828451882846,
"grad_norm": 5.649405479431152,
"kl": 0.267578125,
"learning_rate": 1.3807531380753137e-07,
"loss": 0.0107,
"reward": 0.6608505845069885,
"reward_std": 0.08037957549095154,
"reward_std_all": 0.09471864253282547,
"rewards-max/forensics_tempsamp_iou_reward": 0.894340842962265,
"rewards/forensics_tempsamp_iou_reward": 0.44210055470466614,
"step": 825
},
{
"completion_length": 22.109375,
"epoch": 3.4560669456066946,
"grad_norm": 8.833309173583984,
"kl": 0.40234375,
"learning_rate": 1.370292887029289e-07,
"loss": 0.0161,
"reward": 0.7818557322025299,
"reward_std": 0.08102788031101227,
"reward_std_all": 0.11771801859140396,
"rewards-max/forensics_tempsamp_iou_reward": 0.9548750221729279,
"rewards/forensics_tempsamp_iou_reward": 0.5568557381629944,
"step": 826
},
{
"completion_length": 21.59375,
"epoch": 3.4602510460251046,
"grad_norm": 8.117084503173828,
"kl": 0.23388671875,
"learning_rate": 1.3598326359832635e-07,
"loss": 0.0093,
"reward": 0.9007691740989685,
"reward_std": 0.0666496641933918,
"reward_std_all": 0.0701899528503418,
"rewards-max/forensics_tempsamp_iou_reward": 0.9755902886390686,
"rewards/forensics_tempsamp_iou_reward": 0.6257691979408264,
"step": 827
},
{
"completion_length": 20.5625,
"epoch": 3.4644351464435146,
"grad_norm": 5.6521759033203125,
"kl": 0.2138671875,
"learning_rate": 1.3493723849372386e-07,
"loss": 0.0085,
"reward": 0.8311149775981903,
"reward_std": 0.05541391111910343,
"reward_std_all": 0.088836420327425,
"rewards-max/forensics_tempsamp_iou_reward": 0.8837298154830933,
"rewards/forensics_tempsamp_iou_reward": 0.5498649775981903,
"step": 828
},
{
"completion_length": 20.484375,
"epoch": 3.4686192468619246,
"grad_norm": 5.4272260665893555,
"kl": 0.2734375,
"learning_rate": 1.3389121338912132e-07,
"loss": 0.0109,
"reward": 0.7570550441741943,
"reward_std": 0.08201073110103607,
"reward_std_all": 0.08096684515476227,
"rewards-max/forensics_tempsamp_iou_reward": 0.9424999952316284,
"rewards/forensics_tempsamp_iou_reward": 0.5195550173521042,
"step": 829
},
{
"completion_length": 21.890625,
"epoch": 3.4728033472803346,
"grad_norm": 6.118084907531738,
"kl": 0.20703125,
"learning_rate": 1.3284518828451883e-07,
"loss": 0.0083,
"reward": 0.9161318242549896,
"reward_std": 0.06905283406376839,
"reward_std_all": 0.07246368937194347,
"rewards-max/forensics_tempsamp_iou_reward": 0.9844437837600708,
"rewards/forensics_tempsamp_iou_reward": 0.6723818182945251,
"step": 830
},
{
"completion_length": 22.515625,
"epoch": 3.476987447698745,
"grad_norm": 5.964905261993408,
"kl": 0.169921875,
"learning_rate": 1.3179916317991632e-07,
"loss": 0.0068,
"reward": 0.9093535542488098,
"reward_std": 0.08345672488212585,
"reward_std_all": 0.12216313928365707,
"rewards-max/forensics_tempsamp_iou_reward": 0.9297304153442383,
"rewards/forensics_tempsamp_iou_reward": 0.6343535482883453,
"step": 831
},
{
"completion_length": 19.828125,
"epoch": 3.481171548117155,
"grad_norm": 5.7027997970581055,
"kl": 0.22119140625,
"learning_rate": 1.307531380753138e-07,
"loss": 0.0088,
"reward": 0.8317398726940155,
"reward_std": 0.08292530104517937,
"reward_std_all": 0.07720471918582916,
"rewards-max/forensics_tempsamp_iou_reward": 0.9124975800514221,
"rewards/forensics_tempsamp_iou_reward": 0.5504898279905319,
"step": 832
},
{
"completion_length": 20.15625,
"epoch": 3.485355648535565,
"grad_norm": 4.667632102966309,
"kl": 0.1455078125,
"learning_rate": 1.297071129707113e-07,
"loss": 0.0058,
"reward": 0.8114595115184784,
"reward_std": 0.053958479315042496,
"reward_std_all": 0.06338765099644661,
"rewards-max/forensics_tempsamp_iou_reward": 0.9579326808452606,
"rewards/forensics_tempsamp_iou_reward": 0.548959493637085,
"step": 833
},
{
"completion_length": 21.125,
"epoch": 3.489539748953975,
"grad_norm": 5.837747097015381,
"kl": 0.361328125,
"learning_rate": 1.2866108786610878e-07,
"loss": 0.0144,
"reward": 0.8194628357887268,
"reward_std": 0.14464281499385834,
"reward_std_all": 0.13419357687234879,
"rewards-max/forensics_tempsamp_iou_reward": 0.962036520242691,
"rewards/forensics_tempsamp_iou_reward": 0.569462850689888,
"step": 834
},
{
"completion_length": 19.5,
"epoch": 3.493723849372385,
"grad_norm": 4.60792350769043,
"kl": 0.16162109375,
"learning_rate": 1.2761506276150627e-07,
"loss": 0.0065,
"reward": 0.9211877286434174,
"reward_std": 0.06761127524077892,
"reward_std_all": 0.06433133408427238,
"rewards-max/forensics_tempsamp_iou_reward": 0.9899696707725525,
"rewards/forensics_tempsamp_iou_reward": 0.6586876809597015,
"step": 835
},
{
"completion_length": 19.296875,
"epoch": 3.497907949790795,
"grad_norm": 9.310798645019531,
"kl": 0.1904296875,
"learning_rate": 1.2656903765690378e-07,
"loss": 0.0076,
"reward": 0.8444644212722778,
"reward_std": 0.053272414952516556,
"reward_std_all": 0.06317177973687649,
"rewards-max/forensics_tempsamp_iou_reward": 0.972398191690445,
"rewards/forensics_tempsamp_iou_reward": 0.581964373588562,
"step": 836
},
{
"completion_length": 20.84375,
"epoch": 3.502092050209205,
"grad_norm": 7.485479831695557,
"kl": 0.35791015625,
"learning_rate": 1.2552301255230124e-07,
"loss": 0.0144,
"reward": 0.8603480160236359,
"reward_std": 0.126125056296587,
"reward_std_all": 0.11007135361433029,
"rewards-max/forensics_tempsamp_iou_reward": 0.9308039247989655,
"rewards/forensics_tempsamp_iou_reward": 0.5915979743003845,
"step": 837
},
{
"completion_length": 19.984375,
"epoch": 3.506276150627615,
"grad_norm": 6.2481770515441895,
"kl": 0.185546875,
"learning_rate": 1.2447698744769875e-07,
"loss": 0.0074,
"reward": 0.7704323828220367,
"reward_std": 0.07153974287211895,
"reward_std_all": 0.07005337066948414,
"rewards-max/forensics_tempsamp_iou_reward": 0.9179902970790863,
"rewards/forensics_tempsamp_iou_reward": 0.5141823291778564,
"step": 838
},
{
"completion_length": 19.703125,
"epoch": 3.510460251046025,
"grad_norm": 5.078647613525391,
"kl": 0.21435546875,
"learning_rate": 1.2343096234309624e-07,
"loss": 0.0086,
"reward": 0.9125867486000061,
"reward_std": 0.09719084203243256,
"reward_std_all": 0.0895625688135624,
"rewards-max/forensics_tempsamp_iou_reward": 0.9431822597980499,
"rewards/forensics_tempsamp_iou_reward": 0.6563367247581482,
"step": 839
},
{
"completion_length": 19.796875,
"epoch": 3.514644351464435,
"grad_norm": 4.325468063354492,
"kl": 0.22705078125,
"learning_rate": 1.2238493723849372e-07,
"loss": 0.0091,
"reward": 0.8566451668739319,
"reward_std": 0.10597805306315422,
"reward_std_all": 0.1157272607088089,
"rewards-max/forensics_tempsamp_iou_reward": 0.8919721245765686,
"rewards/forensics_tempsamp_iou_reward": 0.5753951072692871,
"step": 840
},
{
"completion_length": 21.015625,
"epoch": 3.518828451882845,
"grad_norm": 5.8359808921813965,
"kl": 0.2275390625,
"learning_rate": 1.213389121338912e-07,
"loss": 0.0091,
"reward": 0.8423251807689667,
"reward_std": 0.14674432761967182,
"reward_std_all": 0.13896996155381203,
"rewards-max/forensics_tempsamp_iou_reward": 0.9097931683063507,
"rewards/forensics_tempsamp_iou_reward": 0.5923252105712891,
"step": 841
},
{
"completion_length": 19.0625,
"epoch": 3.523012552301255,
"grad_norm": 3.7932450771331787,
"kl": 0.18408203125,
"learning_rate": 1.202928870292887e-07,
"loss": 0.0074,
"reward": 0.822777271270752,
"reward_std": 0.12781166657805443,
"reward_std_all": 0.11433867737650871,
"rewards-max/forensics_tempsamp_iou_reward": 0.9893617033958435,
"rewards/forensics_tempsamp_iou_reward": 0.5915272831916809,
"step": 842
},
{
"completion_length": 20.203125,
"epoch": 3.5271966527196654,
"grad_norm": 7.167655944824219,
"kl": 0.22802734375,
"learning_rate": 1.1924686192468618e-07,
"loss": 0.0091,
"reward": 0.8332101404666901,
"reward_std": 0.08904916606843472,
"reward_std_all": 0.10978654026985168,
"rewards-max/forensics_tempsamp_iou_reward": 0.9737073481082916,
"rewards/forensics_tempsamp_iou_reward": 0.5769601464271545,
"step": 843
},
{
"completion_length": 20.0625,
"epoch": 3.5313807531380754,
"grad_norm": 11.87281608581543,
"kl": 0.16552734375,
"learning_rate": 1.1820083682008368e-07,
"loss": 0.0066,
"reward": 0.77862149477005,
"reward_std": 0.0711123775690794,
"reward_std_all": 0.07320344261825085,
"rewards-max/forensics_tempsamp_iou_reward": 0.972427636384964,
"rewards/forensics_tempsamp_iou_reward": 0.5411215275526047,
"step": 844
},
{
"completion_length": 19.140625,
"epoch": 3.5355648535564854,
"grad_norm": 5.882035732269287,
"kl": 0.1962890625,
"learning_rate": 1.1715481171548117e-07,
"loss": 0.0079,
"reward": 0.879595935344696,
"reward_std": 0.06537919491529465,
"reward_std_all": 0.06598380580544472,
"rewards-max/forensics_tempsamp_iou_reward": 0.93675497174263,
"rewards/forensics_tempsamp_iou_reward": 0.598345935344696,
"step": 845
},
{
"completion_length": 21.0,
"epoch": 3.5397489539748954,
"grad_norm": 5.341065883636475,
"kl": 0.20263671875,
"learning_rate": 1.1610878661087866e-07,
"loss": 0.0081,
"reward": 0.6839787662029266,
"reward_std": 0.13166215270757675,
"reward_std_all": 0.12205830216407776,
"rewards-max/forensics_tempsamp_iou_reward": 0.8944413363933563,
"rewards/forensics_tempsamp_iou_reward": 0.43397870659828186,
"step": 846
},
{
"completion_length": 21.03125,
"epoch": 3.5439330543933054,
"grad_norm": 5.8039469718933105,
"kl": 0.2001953125,
"learning_rate": 1.1506276150627614e-07,
"loss": 0.008,
"reward": 0.865489661693573,
"reward_std": 0.04872780293226242,
"reward_std_all": 0.06766165047883987,
"rewards-max/forensics_tempsamp_iou_reward": 0.9164511263370514,
"rewards/forensics_tempsamp_iou_reward": 0.6029896438121796,
"step": 847
},
{
"completion_length": 20.40625,
"epoch": 3.5481171548117154,
"grad_norm": 7.542144775390625,
"kl": 0.34765625,
"learning_rate": 1.1401673640167364e-07,
"loss": 0.0139,
"reward": 0.8310225903987885,
"reward_std": 0.13980301842093468,
"reward_std_all": 0.12939241528511047,
"rewards-max/forensics_tempsamp_iou_reward": 0.9329364895820618,
"rewards/forensics_tempsamp_iou_reward": 0.5497725903987885,
"step": 848
},
{
"completion_length": 19.03125,
"epoch": 3.5523012552301254,
"grad_norm": 5.07431697845459,
"kl": 0.3291015625,
"learning_rate": 1.1297071129707113e-07,
"loss": 0.0132,
"reward": 0.8713460266590118,
"reward_std": 0.11434970423579216,
"reward_std_all": 0.10784624889492989,
"rewards-max/forensics_tempsamp_iou_reward": 0.9324942529201508,
"rewards/forensics_tempsamp_iou_reward": 0.6150960326194763,
"step": 849
},
{
"completion_length": 22.140625,
"epoch": 3.5564853556485354,
"grad_norm": 4.856814861297607,
"kl": 0.4169921875,
"learning_rate": 1.1192468619246861e-07,
"loss": 0.0167,
"reward": 0.9434925317764282,
"reward_std": 0.11241542547941208,
"reward_std_all": 0.10844680294394493,
"rewards-max/forensics_tempsamp_iou_reward": 0.9745840430259705,
"rewards/forensics_tempsamp_iou_reward": 0.64974245429039,
"step": 850
},
{
"completion_length": 20.140625,
"epoch": 3.560669456066946,
"grad_norm": 5.4602766036987305,
"kl": 0.297119140625,
"learning_rate": 1.108786610878661e-07,
"loss": 0.0119,
"reward": 0.6943217217922211,
"reward_std": 0.08486372418701649,
"reward_std_all": 0.08339666947722435,
"rewards-max/forensics_tempsamp_iou_reward": 0.9372734129428864,
"rewards/forensics_tempsamp_iou_reward": 0.49432173371315,
"step": 851
},
{
"completion_length": 20.765625,
"epoch": 3.564853556485356,
"grad_norm": 4.039261817932129,
"kl": 0.224609375,
"learning_rate": 1.0983263598326359e-07,
"loss": 0.009,
"reward": 0.9335907101631165,
"reward_std": 0.07791449502110481,
"reward_std_all": 0.07555897906422615,
"rewards-max/forensics_tempsamp_iou_reward": 0.9421969056129456,
"rewards/forensics_tempsamp_iou_reward": 0.6523407101631165,
"step": 852
},
{
"completion_length": 21.515625,
"epoch": 3.569037656903766,
"grad_norm": 4.264404296875,
"kl": 0.28125,
"learning_rate": 1.0878661087866109e-07,
"loss": 0.0112,
"reward": 0.7946205139160156,
"reward_std": 0.056819627061486244,
"reward_std_all": 0.10250256955623627,
"rewards-max/forensics_tempsamp_iou_reward": 0.9410712420940399,
"rewards/forensics_tempsamp_iou_reward": 0.5696205496788025,
"step": 853
},
{
"completion_length": 21.390625,
"epoch": 3.573221757322176,
"grad_norm": 7.638987064361572,
"kl": 0.31640625,
"learning_rate": 1.0774058577405857e-07,
"loss": 0.0127,
"reward": 0.7584808170795441,
"reward_std": 0.04950340557843447,
"reward_std_all": 0.0869273878633976,
"rewards-max/forensics_tempsamp_iou_reward": 0.9353296458721161,
"rewards/forensics_tempsamp_iou_reward": 0.5084808170795441,
"step": 854
},
{
"completion_length": 19.453125,
"epoch": 3.577405857740586,
"grad_norm": 6.349947929382324,
"kl": 0.20068359375,
"learning_rate": 1.0669456066945606e-07,
"loss": 0.008,
"reward": 0.8987928032875061,
"reward_std": 0.1002221517264843,
"reward_std_all": 0.094815693795681,
"rewards-max/forensics_tempsamp_iou_reward": 0.9688194692134857,
"rewards/forensics_tempsamp_iou_reward": 0.6237927377223969,
"step": 855
},
{
"completion_length": 22.96875,
"epoch": 3.581589958158996,
"grad_norm": 4.460015773773193,
"kl": 0.263671875,
"learning_rate": 1.0564853556485355e-07,
"loss": 0.0105,
"reward": 0.507900595664978,
"reward_std": 0.09826074540615082,
"reward_std_all": 0.08919547125697136,
"rewards-max/forensics_tempsamp_iou_reward": 0.8746365904808044,
"rewards/forensics_tempsamp_iou_reward": 0.3329005539417267,
"step": 856
},
{
"completion_length": 21.578125,
"epoch": 3.585774058577406,
"grad_norm": 4.511330604553223,
"kl": 0.31103515625,
"learning_rate": 1.0460251046025103e-07,
"loss": 0.0125,
"reward": 0.9930069148540497,
"reward_std": 0.054822586476802826,
"reward_std_all": 0.07162464782595634,
"rewards-max/forensics_tempsamp_iou_reward": 0.9624105989933014,
"rewards/forensics_tempsamp_iou_reward": 0.730506956577301,
"step": 857
},
{
"completion_length": 19.984375,
"epoch": 3.589958158995816,
"grad_norm": 7.1105804443359375,
"kl": 0.2802734375,
"learning_rate": 1.0355648535564853e-07,
"loss": 0.0112,
"reward": 0.7720573544502258,
"reward_std": 0.1013832800090313,
"reward_std_all": 0.10952021554112434,
"rewards-max/forensics_tempsamp_iou_reward": 0.8680019080638885,
"rewards/forensics_tempsamp_iou_reward": 0.5220573544502258,
"step": 858
},
{
"completion_length": 21.6875,
"epoch": 3.594142259414226,
"grad_norm": 4.659698009490967,
"kl": 0.24462890625,
"learning_rate": 1.0251046025104603e-07,
"loss": 0.0098,
"reward": 0.8286031186580658,
"reward_std": 0.10594042018055916,
"reward_std_all": 0.10080787912011147,
"rewards-max/forensics_tempsamp_iou_reward": 0.9757278561592102,
"rewards/forensics_tempsamp_iou_reward": 0.5598531365394592,
"step": 859
},
{
"completion_length": 21.484375,
"epoch": 3.598326359832636,
"grad_norm": 7.870604991912842,
"kl": 0.2421875,
"learning_rate": 1.0146443514644352e-07,
"loss": 0.0097,
"reward": 0.7864042520523071,
"reward_std": 0.07137547433376312,
"reward_std_all": 0.06728427298367023,
"rewards-max/forensics_tempsamp_iou_reward": 0.9835668206214905,
"rewards/forensics_tempsamp_iou_reward": 0.5426542609930038,
"step": 860
},
{
"completion_length": 20.28125,
"epoch": 3.602510460251046,
"grad_norm": 4.479707717895508,
"kl": 0.2021484375,
"learning_rate": 1.00418410041841e-07,
"loss": 0.0081,
"reward": 0.8627891838550568,
"reward_std": 0.09358846955001354,
"reward_std_all": 0.0919106975197792,
"rewards-max/forensics_tempsamp_iou_reward": 0.9749959409236908,
"rewards/forensics_tempsamp_iou_reward": 0.6002891063690186,
"step": 861
},
{
"completion_length": 21.21875,
"epoch": 3.606694560669456,
"grad_norm": 5.517812728881836,
"kl": 0.4384765625,
"learning_rate": 9.937238493723849e-08,
"loss": 0.0175,
"reward": 0.941764235496521,
"reward_std": 0.09718809649348259,
"reward_std_all": 0.0994893740862608,
"rewards-max/forensics_tempsamp_iou_reward": 0.9483720362186432,
"rewards/forensics_tempsamp_iou_reward": 0.6667641699314117,
"step": 862
},
{
"completion_length": 20.75,
"epoch": 3.6108786610878663,
"grad_norm": 5.556605339050293,
"kl": 0.2900390625,
"learning_rate": 9.832635983263599e-08,
"loss": 0.0116,
"reward": 0.9730939269065857,
"reward_std": 0.10693299025297165,
"reward_std_all": 0.10274919494986534,
"rewards-max/forensics_tempsamp_iou_reward": 0.965780645608902,
"rewards/forensics_tempsamp_iou_reward": 0.6918438971042633,
"step": 863
},
{
"completion_length": 20.515625,
"epoch": 3.6150627615062763,
"grad_norm": 6.236261367797852,
"kl": 0.20263671875,
"learning_rate": 9.728033472803348e-08,
"loss": 0.0081,
"reward": 0.9976099133491516,
"reward_std": 0.0863216370344162,
"reward_std_all": 0.08897199481725693,
"rewards-max/forensics_tempsamp_iou_reward": 0.9784913659095764,
"rewards/forensics_tempsamp_iou_reward": 0.7163598835468292,
"step": 864
},
{
"completion_length": 21.109375,
"epoch": 3.6192468619246863,
"grad_norm": 6.183834075927734,
"kl": 0.16064453125,
"learning_rate": 9.623430962343096e-08,
"loss": 0.0064,
"reward": 0.7416985332965851,
"reward_std": 0.09775840304791927,
"reward_std_all": 0.11129650380462408,
"rewards-max/forensics_tempsamp_iou_reward": 0.9909240901470184,
"rewards/forensics_tempsamp_iou_reward": 0.5291985422372818,
"step": 865
},
{
"completion_length": 21.125,
"epoch": 3.6234309623430963,
"grad_norm": 4.755244255065918,
"kl": 0.2451171875,
"learning_rate": 9.518828451882845e-08,
"loss": 0.0098,
"reward": 0.9284606873989105,
"reward_std": 0.1252463199198246,
"reward_std_all": 0.13231854140758514,
"rewards-max/forensics_tempsamp_iou_reward": 0.8987004160881042,
"rewards/forensics_tempsamp_iou_reward": 0.6409606337547302,
"step": 866
},
{
"completion_length": 21.140625,
"epoch": 3.6276150627615062,
"grad_norm": 6.727728366851807,
"kl": 0.498046875,
"learning_rate": 9.414225941422594e-08,
"loss": 0.0199,
"reward": 0.7578774690628052,
"reward_std": 0.08615271002054214,
"reward_std_all": 0.13937264308333397,
"rewards-max/forensics_tempsamp_iou_reward": 0.9733990430831909,
"rewards/forensics_tempsamp_iou_reward": 0.5328774154186249,
"step": 867
},
{
"completion_length": 21.234375,
"epoch": 3.6317991631799162,
"grad_norm": 10.083358764648438,
"kl": 0.22119140625,
"learning_rate": 9.309623430962342e-08,
"loss": 0.0088,
"reward": 0.701654464006424,
"reward_std": 0.07268352434039116,
"reward_std_all": 0.07274341583251953,
"rewards-max/forensics_tempsamp_iou_reward": 0.9133575856685638,
"rewards/forensics_tempsamp_iou_reward": 0.4766544997692108,
"step": 868
},
{
"completion_length": 21.53125,
"epoch": 3.6359832635983262,
"grad_norm": 4.826000690460205,
"kl": 0.22119140625,
"learning_rate": 9.205020920502092e-08,
"loss": 0.0088,
"reward": 0.8120574653148651,
"reward_std": 0.047132786363363266,
"reward_std_all": 0.11629298701882362,
"rewards-max/forensics_tempsamp_iou_reward": 0.9701263010501862,
"rewards/forensics_tempsamp_iou_reward": 0.5495574474334717,
"step": 869
},
{
"completion_length": 20.453125,
"epoch": 3.6401673640167362,
"grad_norm": 5.041384696960449,
"kl": 0.21533203125,
"learning_rate": 9.100418410041841e-08,
"loss": 0.0086,
"reward": 0.9464716911315918,
"reward_std": 0.07732780277729034,
"reward_std_all": 0.07287763059139252,
"rewards-max/forensics_tempsamp_iou_reward": 0.9848497211933136,
"rewards/forensics_tempsamp_iou_reward": 0.6464717388153076,
"step": 870
},
{
"completion_length": 21.5625,
"epoch": 3.6443514644351467,
"grad_norm": 6.401203632354736,
"kl": 0.23193359375,
"learning_rate": 8.99581589958159e-08,
"loss": 0.0093,
"reward": 0.8033560812473297,
"reward_std": 0.0630107931792736,
"reward_std_all": 0.07304154522716999,
"rewards-max/forensics_tempsamp_iou_reward": 0.9200221598148346,
"rewards/forensics_tempsamp_iou_reward": 0.5408560633659363,
"step": 871
},
{
"completion_length": 19.125,
"epoch": 3.6485355648535567,
"grad_norm": 5.696328163146973,
"kl": 0.27197265625,
"learning_rate": 8.891213389121338e-08,
"loss": 0.0109,
"reward": 0.8301882743835449,
"reward_std": 0.07073265127837658,
"reward_std_all": 0.06554544344544411,
"rewards-max/forensics_tempsamp_iou_reward": 0.8984529376029968,
"rewards/forensics_tempsamp_iou_reward": 0.5864382386207581,
"step": 872
},
{
"completion_length": 20.515625,
"epoch": 3.6527196652719667,
"grad_norm": 10.83694076538086,
"kl": 0.2216796875,
"learning_rate": 8.786610878661087e-08,
"loss": 0.0089,
"reward": 0.8230220675468445,
"reward_std": 0.07761647552251816,
"reward_std_all": 0.09884260594844818,
"rewards-max/forensics_tempsamp_iou_reward": 0.9189378023147583,
"rewards/forensics_tempsamp_iou_reward": 0.5417720973491669,
"step": 873
},
{
"completion_length": 21.375,
"epoch": 3.6569037656903767,
"grad_norm": 4.872481822967529,
"kl": 0.23291015625,
"learning_rate": 8.682008368200837e-08,
"loss": 0.0093,
"reward": 0.8568178117275238,
"reward_std": 0.08766801282763481,
"reward_std_all": 0.12178882211446762,
"rewards-max/forensics_tempsamp_iou_reward": 0.9381128549575806,
"rewards/forensics_tempsamp_iou_reward": 0.6130677461624146,
"step": 874
},
{
"completion_length": 20.71875,
"epoch": 3.6610878661087867,
"grad_norm": 8.254778861999512,
"kl": 0.30078125,
"learning_rate": 8.577405857740585e-08,
"loss": 0.012,
"reward": 0.7883977293968201,
"reward_std": 0.12408871948719025,
"reward_std_all": 0.12580358982086182,
"rewards-max/forensics_tempsamp_iou_reward": 0.9519725441932678,
"rewards/forensics_tempsamp_iou_reward": 0.5446477234363556,
"step": 875
},
{
"completion_length": 20.53125,
"epoch": 3.6652719665271967,
"grad_norm": 5.247279644012451,
"kl": 0.3681640625,
"learning_rate": 8.472803347280334e-08,
"loss": 0.0147,
"reward": 0.8806731104850769,
"reward_std": 0.07802805304527283,
"reward_std_all": 0.07773753814399242,
"rewards-max/forensics_tempsamp_iou_reward": 0.9793152511119843,
"rewards/forensics_tempsamp_iou_reward": 0.6056731045246124,
"step": 876
},
{
"completion_length": 19.46875,
"epoch": 3.6694560669456067,
"grad_norm": 4.679651737213135,
"kl": 0.1943359375,
"learning_rate": 8.368200836820083e-08,
"loss": 0.0078,
"reward": 0.8643108606338501,
"reward_std": 0.06636031717061996,
"reward_std_all": 0.08178159222006798,
"rewards-max/forensics_tempsamp_iou_reward": 0.9369485378265381,
"rewards/forensics_tempsamp_iou_reward": 0.6143108308315277,
"step": 877
},
{
"completion_length": 21.90625,
"epoch": 3.6736401673640167,
"grad_norm": 5.766600608825684,
"kl": 0.2783203125,
"learning_rate": 8.263598326359831e-08,
"loss": 0.0111,
"reward": 0.8702087104320526,
"reward_std": 0.16049497574567795,
"reward_std_all": 0.15129640325903893,
"rewards-max/forensics_tempsamp_iou_reward": 0.978288859128952,
"rewards/forensics_tempsamp_iou_reward": 0.6202087104320526,
"step": 878
},
{
"completion_length": 20.703125,
"epoch": 3.6778242677824267,
"grad_norm": 5.8520827293396,
"kl": 0.27197265625,
"learning_rate": 8.158995815899581e-08,
"loss": 0.0109,
"reward": 0.8140555024147034,
"reward_std": 0.14278029650449753,
"reward_std_all": 0.13346076756715775,
"rewards-max/forensics_tempsamp_iou_reward": 0.9784075617790222,
"rewards/forensics_tempsamp_iou_reward": 0.5515554994344711,
"step": 879
},
{
"completion_length": 19.296875,
"epoch": 3.6820083682008367,
"grad_norm": 6.021464824676514,
"kl": 0.26806640625,
"learning_rate": 8.05439330543933e-08,
"loss": 0.0107,
"reward": 0.8564324676990509,
"reward_std": 0.14360371232032776,
"reward_std_all": 0.16398827731609344,
"rewards-max/forensics_tempsamp_iou_reward": 0.9280373156070709,
"rewards/forensics_tempsamp_iou_reward": 0.5814324021339417,
"step": 880
},
{
"completion_length": 21.1875,
"epoch": 3.6861924686192467,
"grad_norm": 5.69766902923584,
"kl": 0.1962890625,
"learning_rate": 7.949790794979078e-08,
"loss": 0.0078,
"reward": 0.8836152851581573,
"reward_std": 0.09374985471367836,
"reward_std_all": 0.09938547015190125,
"rewards-max/forensics_tempsamp_iou_reward": 0.9724435210227966,
"rewards/forensics_tempsamp_iou_reward": 0.6273653209209442,
"step": 881
},
{
"completion_length": 21.84375,
"epoch": 3.6903765690376567,
"grad_norm": 5.505153656005859,
"kl": 0.328125,
"learning_rate": 7.845188284518828e-08,
"loss": 0.0131,
"reward": 0.6831399202346802,
"reward_std": 0.1194673515856266,
"reward_std_all": 0.12507855892181396,
"rewards-max/forensics_tempsamp_iou_reward": 0.9581449329853058,
"rewards/forensics_tempsamp_iou_reward": 0.47063983976840973,
"step": 882
},
{
"completion_length": 20.609375,
"epoch": 3.694560669456067,
"grad_norm": 4.35683012008667,
"kl": 0.15576171875,
"learning_rate": 7.740585774058577e-08,
"loss": 0.0062,
"reward": 0.9202350080013275,
"reward_std": 0.051329001784324646,
"reward_std_all": 0.053204240277409554,
"rewards-max/forensics_tempsamp_iou_reward": 0.9345050156116486,
"rewards/forensics_tempsamp_iou_reward": 0.6202349960803986,
"step": 883
},
{
"completion_length": 23.6875,
"epoch": 3.698744769874477,
"grad_norm": 8.368014335632324,
"kl": 0.27001953125,
"learning_rate": 7.635983263598327e-08,
"loss": 0.0108,
"reward": 0.6054083108901978,
"reward_std": 0.10236990451812744,
"reward_std_all": 0.09766153246164322,
"rewards-max/forensics_tempsamp_iou_reward": 0.8713101446628571,
"rewards/forensics_tempsamp_iou_reward": 0.36790831387043,
"step": 884
},
{
"completion_length": 20.296875,
"epoch": 3.702928870292887,
"grad_norm": 7.405529499053955,
"kl": 0.2998046875,
"learning_rate": 7.531380753138076e-08,
"loss": 0.012,
"reward": 0.829344779253006,
"reward_std": 0.09185052663087845,
"reward_std_all": 0.09074384532868862,
"rewards-max/forensics_tempsamp_iou_reward": 0.9269350469112396,
"rewards/forensics_tempsamp_iou_reward": 0.5730946958065033,
"step": 885
},
{
"completion_length": 20.328125,
"epoch": 3.707112970711297,
"grad_norm": 7.0806427001953125,
"kl": 0.30029296875,
"learning_rate": 7.426778242677824e-08,
"loss": 0.0121,
"reward": 0.8974334597587585,
"reward_std": 0.15645495802164078,
"reward_std_all": 0.15777134895324707,
"rewards-max/forensics_tempsamp_iou_reward": 0.9718482196331024,
"rewards/forensics_tempsamp_iou_reward": 0.609933465719223,
"step": 886
},
{
"completion_length": 20.65625,
"epoch": 3.711297071129707,
"grad_norm": 5.084476947784424,
"kl": 0.21337890625,
"learning_rate": 7.322175732217573e-08,
"loss": 0.0085,
"reward": 0.7560706734657288,
"reward_std": 0.1344953551888466,
"reward_std_all": 0.12363251298666,
"rewards-max/forensics_tempsamp_iou_reward": 0.8592738211154938,
"rewards/forensics_tempsamp_iou_reward": 0.5060706734657288,
"step": 887
},
{
"completion_length": 19.578125,
"epoch": 3.715481171548117,
"grad_norm": 5.088654041290283,
"kl": 0.18994140625,
"learning_rate": 7.217573221757322e-08,
"loss": 0.0076,
"reward": 0.9029326438903809,
"reward_std": 0.08139988780021667,
"reward_std_all": 0.09639563784003258,
"rewards-max/forensics_tempsamp_iou_reward": 0.9261822402477264,
"rewards/forensics_tempsamp_iou_reward": 0.6279326677322388,
"step": 888
},
{
"completion_length": 20.609375,
"epoch": 3.719665271966527,
"grad_norm": 6.43975305557251,
"kl": 0.22509765625,
"learning_rate": 7.112970711297071e-08,
"loss": 0.009,
"reward": 0.8866855502128601,
"reward_std": 0.08140639960765839,
"reward_std_all": 0.08574716374278069,
"rewards-max/forensics_tempsamp_iou_reward": 0.9348717927932739,
"rewards/forensics_tempsamp_iou_reward": 0.6241856217384338,
"step": 889
},
{
"completion_length": 22.203125,
"epoch": 3.723849372384937,
"grad_norm": 4.704716682434082,
"kl": 0.220703125,
"learning_rate": 7.00836820083682e-08,
"loss": 0.0088,
"reward": 0.9086271524429321,
"reward_std": 0.11805829778313637,
"reward_std_all": 0.10859228670597076,
"rewards-max/forensics_tempsamp_iou_reward": 0.9591346085071564,
"rewards/forensics_tempsamp_iou_reward": 0.6336271166801453,
"step": 890
},
{
"completion_length": 20.625,
"epoch": 3.7280334728033475,
"grad_norm": 4.911661624908447,
"kl": 0.3974609375,
"learning_rate": 6.903765690376569e-08,
"loss": 0.0159,
"reward": 0.8663450181484222,
"reward_std": 0.12173736840486526,
"reward_std_all": 0.11691295355558395,
"rewards-max/forensics_tempsamp_iou_reward": 0.9827542304992676,
"rewards/forensics_tempsamp_iou_reward": 0.5913449823856354,
"step": 891
},
{
"completion_length": 22.140625,
"epoch": 3.7322175732217575,
"grad_norm": 4.616601943969727,
"kl": 0.203125,
"learning_rate": 6.799163179916317e-08,
"loss": 0.0081,
"reward": 0.7287180721759796,
"reward_std": 0.10737812519073486,
"reward_std_all": 0.12414642423391342,
"rewards-max/forensics_tempsamp_iou_reward": 0.8780888319015503,
"rewards/forensics_tempsamp_iou_reward": 0.49746811389923096,
"step": 892
},
{
"completion_length": 21.71875,
"epoch": 3.7364016736401675,
"grad_norm": 5.18781042098999,
"kl": 0.2265625,
"learning_rate": 6.694560669456066e-08,
"loss": 0.0091,
"reward": 0.6676499843597412,
"reward_std": 0.1181462612003088,
"reward_std_all": 0.11917007714509964,
"rewards-max/forensics_tempsamp_iou_reward": 0.8457275032997131,
"rewards/forensics_tempsamp_iou_reward": 0.43015000224113464,
"step": 893
},
{
"completion_length": 21.046875,
"epoch": 3.7405857740585775,
"grad_norm": 5.85235071182251,
"kl": 0.2353515625,
"learning_rate": 6.589958158995816e-08,
"loss": 0.0094,
"reward": 0.8422322571277618,
"reward_std": 0.10420811921358109,
"reward_std_all": 0.10019094869494438,
"rewards-max/forensics_tempsamp_iou_reward": 0.9735502004623413,
"rewards/forensics_tempsamp_iou_reward": 0.6234821677207947,
"step": 894
},
{
"completion_length": 19.6875,
"epoch": 3.7447698744769875,
"grad_norm": 7.507809638977051,
"kl": 0.1904296875,
"learning_rate": 6.485355648535565e-08,
"loss": 0.0076,
"reward": 0.9719577729701996,
"reward_std": 0.11907558888196945,
"reward_std_all": 0.11702779680490494,
"rewards-max/forensics_tempsamp_iou_reward": 0.9585810005664825,
"rewards/forensics_tempsamp_iou_reward": 0.7032077610492706,
"step": 895
},
{
"completion_length": 21.78125,
"epoch": 3.7489539748953975,
"grad_norm": 6.468235969543457,
"kl": 0.19775390625,
"learning_rate": 6.380753138075313e-08,
"loss": 0.0079,
"reward": 0.9164972007274628,
"reward_std": 0.06375706940889359,
"reward_std_all": 0.071169912815094,
"rewards-max/forensics_tempsamp_iou_reward": 0.943981260061264,
"rewards/forensics_tempsamp_iou_reward": 0.6352471709251404,
"step": 896
},
{
"completion_length": 20.390625,
"epoch": 3.7531380753138075,
"grad_norm": 6.330550193786621,
"kl": 0.1640625,
"learning_rate": 6.276150627615062e-08,
"loss": 0.0066,
"reward": 0.7894903719425201,
"reward_std": 0.07481728866696358,
"reward_std_all": 0.09342667832970619,
"rewards-max/forensics_tempsamp_iou_reward": 0.9647893607616425,
"rewards/forensics_tempsamp_iou_reward": 0.5644903779029846,
"step": 897
},
{
"completion_length": 20.34375,
"epoch": 3.7573221757322175,
"grad_norm": 6.335285663604736,
"kl": 0.177734375,
"learning_rate": 6.171548117154812e-08,
"loss": 0.0071,
"reward": 0.8895220458507538,
"reward_std": 0.11000756919384003,
"reward_std_all": 0.10817784070968628,
"rewards-max/forensics_tempsamp_iou_reward": 0.9751251637935638,
"rewards/forensics_tempsamp_iou_reward": 0.6207720637321472,
"step": 898
},
{
"completion_length": 19.515625,
"epoch": 3.7615062761506275,
"grad_norm": 5.707841873168945,
"kl": 0.212890625,
"learning_rate": 6.06694560669456e-08,
"loss": 0.0085,
"reward": 0.7893189787864685,
"reward_std": 0.09815448895096779,
"reward_std_all": 0.10869376920163631,
"rewards-max/forensics_tempsamp_iou_reward": 0.94936203956604,
"rewards/forensics_tempsamp_iou_reward": 0.5518189072608948,
"step": 899
},
{
"completion_length": 19.515625,
"epoch": 3.7656903765690375,
"grad_norm": 5.8518524169921875,
"kl": 0.17041015625,
"learning_rate": 5.962343096234309e-08,
"loss": 0.0068,
"reward": 0.8761235177516937,
"reward_std": 0.08787978813052177,
"reward_std_all": 0.08684396371245384,
"rewards-max/forensics_tempsamp_iou_reward": 0.9599614441394806,
"rewards/forensics_tempsamp_iou_reward": 0.6073734760284424,
"step": 900
},
{
"completion_length": 19.921875,
"epoch": 3.7698744769874475,
"grad_norm": 3.9296717643737793,
"kl": 0.2685546875,
"learning_rate": 5.8577405857740584e-08,
"loss": 0.0108,
"reward": 0.9111180901527405,
"reward_std": 0.07769006863236427,
"reward_std_all": 0.06961308605968952,
"rewards-max/forensics_tempsamp_iou_reward": 0.9877252280712128,
"rewards/forensics_tempsamp_iou_reward": 0.6298680901527405,
"step": 901
},
{
"completion_length": 21.734375,
"epoch": 3.7740585774058575,
"grad_norm": 5.604909420013428,
"kl": 0.22216796875,
"learning_rate": 5.753138075313807e-08,
"loss": 0.0089,
"reward": 0.8071779608726501,
"reward_std": 0.09441426023840904,
"reward_std_all": 0.08602534793317318,
"rewards-max/forensics_tempsamp_iou_reward": 0.931553989648819,
"rewards/forensics_tempsamp_iou_reward": 0.5509279519319534,
"step": 902
},
{
"completion_length": 20.765625,
"epoch": 3.778242677824268,
"grad_norm": 10.903571128845215,
"kl": 0.16748046875,
"learning_rate": 5.6485355648535564e-08,
"loss": 0.0067,
"reward": 0.872653603553772,
"reward_std": 0.08389346674084663,
"reward_std_all": 0.08920396864414215,
"rewards-max/forensics_tempsamp_iou_reward": 0.9729682505130768,
"rewards/forensics_tempsamp_iou_reward": 0.6164036095142365,
"step": 903
},
{
"completion_length": 22.125,
"epoch": 3.782426778242678,
"grad_norm": 4.751976013183594,
"kl": 0.3330078125,
"learning_rate": 5.543933054393305e-08,
"loss": 0.0133,
"reward": 0.9351911544799805,
"reward_std": 0.14752981066703796,
"reward_std_all": 0.14608148485422134,
"rewards-max/forensics_tempsamp_iou_reward": 0.9249931573867798,
"rewards/forensics_tempsamp_iou_reward": 0.6476910412311554,
"step": 904
},
{
"completion_length": 19.265625,
"epoch": 3.786610878661088,
"grad_norm": 6.897910118103027,
"kl": 0.35400390625,
"learning_rate": 5.439330543933054e-08,
"loss": 0.0142,
"reward": 0.687904566526413,
"reward_std": 0.09078638628125191,
"reward_std_all": 0.08730916678905487,
"rewards-max/forensics_tempsamp_iou_reward": 0.93233922123909,
"rewards/forensics_tempsamp_iou_reward": 0.4504045099020004,
"step": 905
},
{
"completion_length": 20.375,
"epoch": 3.790794979079498,
"grad_norm": 5.609810829162598,
"kl": 0.2080078125,
"learning_rate": 5.334728033472803e-08,
"loss": 0.0083,
"reward": 0.8629111647605896,
"reward_std": 0.0883413776755333,
"reward_std_all": 0.08856727182865143,
"rewards-max/forensics_tempsamp_iou_reward": 0.9076113402843475,
"rewards/forensics_tempsamp_iou_reward": 0.6004111170768738,
"step": 906
},
{
"completion_length": 21.65625,
"epoch": 3.794979079497908,
"grad_norm": 4.309364318847656,
"kl": 0.267578125,
"learning_rate": 5.2301255230125516e-08,
"loss": 0.0107,
"reward": 0.7696221172809601,
"reward_std": 0.09047429636120796,
"reward_std_all": 0.08770891651511192,
"rewards-max/forensics_tempsamp_iou_reward": 0.948811948299408,
"rewards/forensics_tempsamp_iou_reward": 0.5258721262216568,
"step": 907
},
{
"completion_length": 20.375,
"epoch": 3.799163179916318,
"grad_norm": 5.440506935119629,
"kl": 0.2353515625,
"learning_rate": 5.1255230125523015e-08,
"loss": 0.0094,
"reward": 0.8435210287570953,
"reward_std": 0.11753758788108826,
"reward_std_all": 0.10611177235841751,
"rewards-max/forensics_tempsamp_iou_reward": 0.9601820409297943,
"rewards/forensics_tempsamp_iou_reward": 0.5810210406780243,
"step": 908
},
{
"completion_length": 20.265625,
"epoch": 3.803347280334728,
"grad_norm": 7.488703727722168,
"kl": 0.22216796875,
"learning_rate": 5.02092050209205e-08,
"loss": 0.0089,
"reward": 1.031741738319397,
"reward_std": 0.058227572590112686,
"reward_std_all": 0.07007449492812157,
"rewards-max/forensics_tempsamp_iou_reward": 0.9858506619930267,
"rewards/forensics_tempsamp_iou_reward": 0.731741726398468,
"step": 909
},
{
"completion_length": 19.953125,
"epoch": 3.8075313807531384,
"grad_norm": 6.324355602264404,
"kl": 0.22119140625,
"learning_rate": 4.9163179916317995e-08,
"loss": 0.0088,
"reward": 0.9839460849761963,
"reward_std": 0.11609040945768356,
"reward_std_all": 0.10228943079710007,
"rewards-max/forensics_tempsamp_iou_reward": 0.9672704935073853,
"rewards/forensics_tempsamp_iou_reward": 0.6964461505413055,
"step": 910
},
{
"completion_length": 19.9375,
"epoch": 3.8117154811715483,
"grad_norm": 6.671986103057861,
"kl": 0.3837890625,
"learning_rate": 4.811715481171548e-08,
"loss": 0.0154,
"reward": 0.7480193972587585,
"reward_std": 0.07986175827682018,
"reward_std_all": 0.07494200021028519,
"rewards-max/forensics_tempsamp_iou_reward": 0.853715717792511,
"rewards/forensics_tempsamp_iou_reward": 0.491769403219223,
"step": 911
},
{
"completion_length": 20.71875,
"epoch": 3.8158995815899583,
"grad_norm": 4.804161071777344,
"kl": 0.21044921875,
"learning_rate": 4.707112970711297e-08,
"loss": 0.0084,
"reward": 0.8746697902679443,
"reward_std": 0.12718023359775543,
"reward_std_all": 0.12115528807044029,
"rewards-max/forensics_tempsamp_iou_reward": 0.9371907114982605,
"rewards/forensics_tempsamp_iou_reward": 0.5996697545051575,
"step": 912
},
{
"completion_length": 19.890625,
"epoch": 3.8200836820083683,
"grad_norm": 5.3621110916137695,
"kl": 0.1796875,
"learning_rate": 4.602510460251046e-08,
"loss": 0.0072,
"reward": 0.8733021020889282,
"reward_std": 0.05764736607670784,
"reward_std_all": 0.07428093254566193,
"rewards-max/forensics_tempsamp_iou_reward": 0.9545956254005432,
"rewards/forensics_tempsamp_iou_reward": 0.6170521080493927,
"step": 913
},
{
"completion_length": 20.03125,
"epoch": 3.8242677824267783,
"grad_norm": 5.830326557159424,
"kl": 0.1611328125,
"learning_rate": 4.497907949790795e-08,
"loss": 0.0064,
"reward": 0.7481701076030731,
"reward_std": 0.115584347397089,
"reward_std_all": 0.10359958559274673,
"rewards-max/forensics_tempsamp_iou_reward": 0.9701865613460541,
"rewards/forensics_tempsamp_iou_reward": 0.5106701254844666,
"step": 914
},
{
"completion_length": 21.546875,
"epoch": 3.8284518828451883,
"grad_norm": 4.310945510864258,
"kl": 0.216796875,
"learning_rate": 4.393305439330543e-08,
"loss": 0.0087,
"reward": 0.816105306148529,
"reward_std": 0.1989930048584938,
"reward_std_all": 0.1834089457988739,
"rewards-max/forensics_tempsamp_iou_reward": 0.9778798818588257,
"rewards/forensics_tempsamp_iou_reward": 0.5598552823066711,
"step": 915
},
{
"completion_length": 22.625,
"epoch": 3.8326359832635983,
"grad_norm": 6.208245754241943,
"kl": 0.27685546875,
"learning_rate": 4.2887029288702926e-08,
"loss": 0.0111,
"reward": 0.8294122219085693,
"reward_std": 0.1142343282699585,
"reward_std_all": 0.11382659524679184,
"rewards-max/forensics_tempsamp_iou_reward": 0.8986714780330658,
"rewards/forensics_tempsamp_iou_reward": 0.5669121742248535,
"step": 916
},
{
"completion_length": 22.078125,
"epoch": 3.8368200836820083,
"grad_norm": 4.6699066162109375,
"kl": 0.24365234375,
"learning_rate": 4.184100418410041e-08,
"loss": 0.0097,
"reward": 0.8938531279563904,
"reward_std": 0.1103402953594923,
"reward_std_all": 0.11489731073379517,
"rewards-max/forensics_tempsamp_iou_reward": 0.9585767686367035,
"rewards/forensics_tempsamp_iou_reward": 0.6188532114028931,
"step": 917
},
{
"completion_length": 21.15625,
"epoch": 3.8410041841004183,
"grad_norm": 6.559220314025879,
"kl": 0.236328125,
"learning_rate": 4.0794979079497906e-08,
"loss": 0.0095,
"reward": 0.9397952258586884,
"reward_std": 0.12742630019783974,
"reward_std_all": 0.12212477251887321,
"rewards-max/forensics_tempsamp_iou_reward": 0.977254182100296,
"rewards/forensics_tempsamp_iou_reward": 0.6460452079772949,
"step": 918
},
{
"completion_length": 21.09375,
"epoch": 3.8451882845188283,
"grad_norm": 5.954547882080078,
"kl": 0.2490234375,
"learning_rate": 3.974895397489539e-08,
"loss": 0.01,
"reward": 0.8114929795265198,
"reward_std": 0.10792623087763786,
"reward_std_all": 0.16329004615545273,
"rewards-max/forensics_tempsamp_iou_reward": 0.9568099975585938,
"rewards/forensics_tempsamp_iou_reward": 0.5739929974079132,
"step": 919
},
{
"completion_length": 20.984375,
"epoch": 3.8493723849372383,
"grad_norm": 9.074251174926758,
"kl": 0.228515625,
"learning_rate": 3.8702928870292885e-08,
"loss": 0.0091,
"reward": 0.7699578702449799,
"reward_std": 0.07165630161762238,
"reward_std_all": 0.06731626018881798,
"rewards-max/forensics_tempsamp_iou_reward": 0.8807091414928436,
"rewards/forensics_tempsamp_iou_reward": 0.526207834482193,
"step": 920
},
{
"completion_length": 19.484375,
"epoch": 3.8535564853556483,
"grad_norm": 6.307717323303223,
"kl": 0.234375,
"learning_rate": 3.765690376569038e-08,
"loss": 0.0094,
"reward": 0.7961857914924622,
"reward_std": 0.09013990685343742,
"reward_std_all": 0.0848006159067154,
"rewards-max/forensics_tempsamp_iou_reward": 0.9734150171279907,
"rewards/forensics_tempsamp_iou_reward": 0.5086857676506042,
"step": 921
},
{
"completion_length": 21.46875,
"epoch": 3.8577405857740583,
"grad_norm": 4.636782646179199,
"kl": 0.3701171875,
"learning_rate": 3.6610878661087864e-08,
"loss": 0.0148,
"reward": 0.6855726838111877,
"reward_std": 0.10715112835168839,
"reward_std_all": 0.10614701732993126,
"rewards-max/forensics_tempsamp_iou_reward": 0.934808611869812,
"rewards/forensics_tempsamp_iou_reward": 0.4543227255344391,
"step": 922
},
{
"completion_length": 20.21875,
"epoch": 3.8619246861924688,
"grad_norm": 6.385278701782227,
"kl": 0.19580078125,
"learning_rate": 3.556485355648536e-08,
"loss": 0.0078,
"reward": 0.8873542547225952,
"reward_std": 0.046028824523091316,
"reward_std_all": 0.06619622185826302,
"rewards-max/forensics_tempsamp_iou_reward": 0.9836679995059967,
"rewards/forensics_tempsamp_iou_reward": 0.6061041951179504,
"step": 923
},
{
"completion_length": 20.84375,
"epoch": 3.8661087866108788,
"grad_norm": 5.682069778442383,
"kl": 0.2734375,
"learning_rate": 3.4518828451882844e-08,
"loss": 0.0109,
"reward": 0.9315243065357208,
"reward_std": 0.1117018274962902,
"reward_std_all": 0.10211874544620514,
"rewards-max/forensics_tempsamp_iou_reward": 0.9637078940868378,
"rewards/forensics_tempsamp_iou_reward": 0.656524270772934,
"step": 924
},
{
"completion_length": 21.46875,
"epoch": 3.8702928870292888,
"grad_norm": 4.717803955078125,
"kl": 0.2490234375,
"learning_rate": 3.347280334728033e-08,
"loss": 0.01,
"reward": 0.8491030633449554,
"reward_std": 0.1474885530769825,
"reward_std_all": 0.1337961032986641,
"rewards-max/forensics_tempsamp_iou_reward": 0.9198960661888123,
"rewards/forensics_tempsamp_iou_reward": 0.6116030216217041,
"step": 925
},
{
"completion_length": 20.828125,
"epoch": 3.8744769874476988,
"grad_norm": 5.920896530151367,
"kl": 0.2138671875,
"learning_rate": 3.242677824267782e-08,
"loss": 0.0086,
"reward": 0.9483626782894135,
"reward_std": 0.07277579605579376,
"reward_std_all": 0.07444415614008904,
"rewards-max/forensics_tempsamp_iou_reward": 0.9083167910575867,
"rewards/forensics_tempsamp_iou_reward": 0.6671126484870911,
"step": 926
},
{
"completion_length": 19.6875,
"epoch": 3.8786610878661087,
"grad_norm": 8.171234130859375,
"kl": 0.22216796875,
"learning_rate": 3.138075313807531e-08,
"loss": 0.0089,
"reward": 0.6900125443935394,
"reward_std": 0.06979247182607651,
"reward_std_all": 0.06765744648873806,
"rewards-max/forensics_tempsamp_iou_reward": 0.9450435042381287,
"rewards/forensics_tempsamp_iou_reward": 0.4837625175714493,
"step": 927
},
{
"completion_length": 21.515625,
"epoch": 3.8828451882845187,
"grad_norm": 6.592249870300293,
"kl": 0.23779296875,
"learning_rate": 3.03347280334728e-08,
"loss": 0.0095,
"reward": 0.8830245435237885,
"reward_std": 0.1007138267159462,
"reward_std_all": 0.10570503398776054,
"rewards-max/forensics_tempsamp_iou_reward": 0.940922200679779,
"rewards/forensics_tempsamp_iou_reward": 0.6267745196819305,
"step": 928
},
{
"completion_length": 19.625,
"epoch": 3.8870292887029287,
"grad_norm": 5.940771102905273,
"kl": 0.1953125,
"learning_rate": 2.9288702928870292e-08,
"loss": 0.0078,
"reward": 0.9152162969112396,
"reward_std": 0.041747868061065674,
"reward_std_all": 0.04327717050909996,
"rewards-max/forensics_tempsamp_iou_reward": 0.990105926990509,
"rewards/forensics_tempsamp_iou_reward": 0.652716338634491,
"step": 929
},
{
"completion_length": 19.75,
"epoch": 3.891213389121339,
"grad_norm": 9.43647289276123,
"kl": 0.19482421875,
"learning_rate": 2.8242677824267782e-08,
"loss": 0.0078,
"reward": 0.8601429164409637,
"reward_std": 0.0991157814860344,
"reward_std_all": 0.09268182888627052,
"rewards-max/forensics_tempsamp_iou_reward": 0.9810481667518616,
"rewards/forensics_tempsamp_iou_reward": 0.5913928747177124,
"step": 930
},
{
"completion_length": 21.078125,
"epoch": 3.895397489539749,
"grad_norm": 6.4652862548828125,
"kl": 0.189453125,
"learning_rate": 2.719665271966527e-08,
"loss": 0.0076,
"reward": 0.716121256351471,
"reward_std": 0.12855156511068344,
"reward_std_all": 0.1259080283343792,
"rewards-max/forensics_tempsamp_iou_reward": 0.9204641580581665,
"rewards/forensics_tempsamp_iou_reward": 0.4848712235689163,
"step": 931
},
{
"completion_length": 21.046875,
"epoch": 3.899581589958159,
"grad_norm": 8.086743354797363,
"kl": 0.201171875,
"learning_rate": 2.6150627615062758e-08,
"loss": 0.008,
"reward": 0.9244536757469177,
"reward_std": 0.11047577112913132,
"reward_std_all": 0.12249820306897163,
"rewards-max/forensics_tempsamp_iou_reward": 0.9824941158294678,
"rewards/forensics_tempsamp_iou_reward": 0.6432036459445953,
"step": 932
},
{
"completion_length": 20.828125,
"epoch": 3.903765690376569,
"grad_norm": 5.321623802185059,
"kl": 0.17236328125,
"learning_rate": 2.510460251046025e-08,
"loss": 0.0069,
"reward": 0.9175682663917542,
"reward_std": 0.08342614769935608,
"reward_std_all": 0.08209828101098537,
"rewards-max/forensics_tempsamp_iou_reward": 0.9731698632240295,
"rewards/forensics_tempsamp_iou_reward": 0.6363182663917542,
"step": 933
},
{
"completion_length": 21.453125,
"epoch": 3.907949790794979,
"grad_norm": 16.230131149291992,
"kl": 0.16796875,
"learning_rate": 2.405857740585774e-08,
"loss": 0.0067,
"reward": 0.7294031083583832,
"reward_std": 0.073759064078331,
"reward_std_all": 0.06616142205893993,
"rewards-max/forensics_tempsamp_iou_reward": 0.981149286031723,
"rewards/forensics_tempsamp_iou_reward": 0.49815310537815094,
"step": 934
},
{
"completion_length": 21.578125,
"epoch": 3.912133891213389,
"grad_norm": 4.623726844787598,
"kl": 0.18505859375,
"learning_rate": 2.301255230125523e-08,
"loss": 0.0074,
"reward": 0.893830418586731,
"reward_std": 0.08444034680724144,
"reward_std_all": 0.10884244740009308,
"rewards-max/forensics_tempsamp_iou_reward": 0.9479551017284393,
"rewards/forensics_tempsamp_iou_reward": 0.6188303828239441,
"step": 935
},
{
"completion_length": 20.703125,
"epoch": 3.916317991631799,
"grad_norm": 6.565701961517334,
"kl": 0.32470703125,
"learning_rate": 2.1966527196652717e-08,
"loss": 0.013,
"reward": 0.8870225548744202,
"reward_std": 0.062730323523283,
"reward_std_all": 0.0630013719201088,
"rewards-max/forensics_tempsamp_iou_reward": 0.9806225895881653,
"rewards/forensics_tempsamp_iou_reward": 0.6120225340127945,
"step": 936
},
{
"completion_length": 20.09375,
"epoch": 3.920502092050209,
"grad_norm": 5.91127347946167,
"kl": 0.208984375,
"learning_rate": 2.0920502092050206e-08,
"loss": 0.0084,
"reward": 0.8595757186412811,
"reward_std": 0.06559843756258488,
"reward_std_all": 0.06446823664009571,
"rewards-max/forensics_tempsamp_iou_reward": 0.9694003760814667,
"rewards/forensics_tempsamp_iou_reward": 0.6158257126808167,
"step": 937
},
{
"completion_length": 21.5,
"epoch": 3.924686192468619,
"grad_norm": 6.339508056640625,
"kl": 0.26708984375,
"learning_rate": 1.9874476987447696e-08,
"loss": 0.0107,
"reward": 0.8767828345298767,
"reward_std": 0.09914219006896019,
"reward_std_all": 0.13176052272319794,
"rewards-max/forensics_tempsamp_iou_reward": 0.9440205991268158,
"rewards/forensics_tempsamp_iou_reward": 0.6017828285694122,
"step": 938
},
{
"completion_length": 20.75,
"epoch": 3.928870292887029,
"grad_norm": 8.125292778015137,
"kl": 0.25341796875,
"learning_rate": 1.882845188284519e-08,
"loss": 0.0101,
"reward": 0.7949934303760529,
"reward_std": 0.06801292113959789,
"reward_std_all": 0.11879760026931763,
"rewards-max/forensics_tempsamp_iou_reward": 0.8938144445419312,
"rewards/forensics_tempsamp_iou_reward": 0.569993406534195,
"step": 939
},
{
"completion_length": 21.21875,
"epoch": 3.933054393305439,
"grad_norm": 7.110605716705322,
"kl": 0.33349609375,
"learning_rate": 1.778242677824268e-08,
"loss": 0.0133,
"reward": 0.7699072659015656,
"reward_std": 0.0631523784250021,
"reward_std_all": 0.0931149497628212,
"rewards-max/forensics_tempsamp_iou_reward": 0.9195725619792938,
"rewards/forensics_tempsamp_iou_reward": 0.5074072778224945,
"step": 940
},
{
"completion_length": 21.5625,
"epoch": 3.937238493723849,
"grad_norm": 7.594330310821533,
"kl": 0.2060546875,
"learning_rate": 1.6736401673640165e-08,
"loss": 0.0082,
"reward": 0.8370694816112518,
"reward_std": 0.08187224622815847,
"reward_std_all": 0.07728641107678413,
"rewards-max/forensics_tempsamp_iou_reward": 0.9714024066925049,
"rewards/forensics_tempsamp_iou_reward": 0.6058194935321808,
"step": 941
},
{
"completion_length": 21.609375,
"epoch": 3.941422594142259,
"grad_norm": 5.133851528167725,
"kl": 0.251953125,
"learning_rate": 1.5690376569037655e-08,
"loss": 0.0101,
"reward": 0.8150250315666199,
"reward_std": 0.07322727143764496,
"reward_std_all": 0.10003375634551048,
"rewards-max/forensics_tempsamp_iou_reward": 0.9300488829612732,
"rewards/forensics_tempsamp_iou_reward": 0.552524983882904,
"step": 942
},
{
"completion_length": 21.015625,
"epoch": 3.9456066945606696,
"grad_norm": 6.3706841468811035,
"kl": 0.22021484375,
"learning_rate": 1.4644351464435146e-08,
"loss": 0.0088,
"reward": 0.8070446848869324,
"reward_std": 0.12882741913199425,
"reward_std_all": 0.15475410223007202,
"rewards-max/forensics_tempsamp_iou_reward": 0.9555293321609497,
"rewards/forensics_tempsamp_iou_reward": 0.5757946372032166,
"step": 943
},
{
"completion_length": 21.359375,
"epoch": 3.9497907949790796,
"grad_norm": 5.397436618804932,
"kl": 0.2080078125,
"learning_rate": 1.3598326359832636e-08,
"loss": 0.0083,
"reward": 0.8697863519191742,
"reward_std": 0.06273455917835236,
"reward_std_all": 0.06475750729441643,
"rewards-max/forensics_tempsamp_iou_reward": 0.9586488306522369,
"rewards/forensics_tempsamp_iou_reward": 0.6072863042354584,
"step": 944
},
{
"completion_length": 21.421875,
"epoch": 3.9539748953974896,
"grad_norm": 5.2257585525512695,
"kl": 0.1796875,
"learning_rate": 1.2552301255230125e-08,
"loss": 0.0072,
"reward": 0.7168184220790863,
"reward_std": 0.06451789103448391,
"reward_std_all": 0.06362028047442436,
"rewards-max/forensics_tempsamp_iou_reward": 0.9477232992649078,
"rewards/forensics_tempsamp_iou_reward": 0.45431841909885406,
"step": 945
},
{
"completion_length": 20.453125,
"epoch": 3.9581589958158996,
"grad_norm": 4.843359470367432,
"kl": 0.18701171875,
"learning_rate": 1.1506276150627615e-08,
"loss": 0.0075,
"reward": 0.7498654723167419,
"reward_std": 0.1015048548579216,
"reward_std_all": 0.09705512970685959,
"rewards-max/forensics_tempsamp_iou_reward": 0.9266602993011475,
"rewards/forensics_tempsamp_iou_reward": 0.5248654782772064,
"step": 946
},
{
"completion_length": 21.953125,
"epoch": 3.9623430962343096,
"grad_norm": 5.650334358215332,
"kl": 0.2109375,
"learning_rate": 1.0460251046025103e-08,
"loss": 0.0084,
"reward": 0.898598313331604,
"reward_std": 0.05045663192868233,
"reward_std_all": 0.0865335687994957,
"rewards-max/forensics_tempsamp_iou_reward": 0.9658755958080292,
"rewards/forensics_tempsamp_iou_reward": 0.617348313331604,
"step": 947
},
{
"completion_length": 20.265625,
"epoch": 3.9665271966527196,
"grad_norm": 6.154312610626221,
"kl": 0.2998046875,
"learning_rate": 9.414225941422594e-09,
"loss": 0.012,
"reward": 0.8529863953590393,
"reward_std": 0.10316770896315575,
"reward_std_all": 0.11940418183803558,
"rewards-max/forensics_tempsamp_iou_reward": 0.8759325444698334,
"rewards/forensics_tempsamp_iou_reward": 0.5779862999916077,
"step": 948
},
{
"completion_length": 19.015625,
"epoch": 3.9707112970711296,
"grad_norm": 7.147533416748047,
"kl": 0.21728515625,
"learning_rate": 8.368200836820083e-09,
"loss": 0.0087,
"reward": 0.9128108322620392,
"reward_std": 0.08935629948973656,
"reward_std_all": 0.08393597416579723,
"rewards-max/forensics_tempsamp_iou_reward": 0.9064815044403076,
"rewards/forensics_tempsamp_iou_reward": 0.6378107964992523,
"step": 949
},
{
"completion_length": 21.515625,
"epoch": 3.97489539748954,
"grad_norm": 5.450655460357666,
"kl": 0.326171875,
"learning_rate": 7.322175732217573e-09,
"loss": 0.0131,
"reward": 0.7504894733428955,
"reward_std": 0.1517994925379753,
"reward_std_all": 0.14700240641832352,
"rewards-max/forensics_tempsamp_iou_reward": 0.8743136823177338,
"rewards/forensics_tempsamp_iou_reward": 0.4942394644021988,
"step": 950
},
{
"completion_length": 19.796875,
"epoch": 3.97907949790795,
"grad_norm": 5.009927272796631,
"kl": 0.22216796875,
"learning_rate": 6.276150627615063e-09,
"loss": 0.0089,
"reward": 0.9036398231983185,
"reward_std": 0.11926144361495972,
"reward_std_all": 0.10731877386569977,
"rewards-max/forensics_tempsamp_iou_reward": 0.9563969671726227,
"rewards/forensics_tempsamp_iou_reward": 0.6286398470401764,
"step": 951
},
{
"completion_length": 19.421875,
"epoch": 3.98326359832636,
"grad_norm": 6.332992076873779,
"kl": 0.21484375,
"learning_rate": 5.2301255230125516e-09,
"loss": 0.0086,
"reward": 0.8351412713527679,
"reward_std": 0.08582228422164917,
"reward_std_all": 0.08325288817286491,
"rewards-max/forensics_tempsamp_iou_reward": 0.9416854381561279,
"rewards/forensics_tempsamp_iou_reward": 0.5601412951946259,
"step": 952
},
{
"completion_length": 20.125,
"epoch": 3.98744769874477,
"grad_norm": 6.541018009185791,
"kl": 0.2109375,
"learning_rate": 4.184100418410041e-09,
"loss": 0.0084,
"reward": 0.8531163930892944,
"reward_std": 0.12394003942608833,
"reward_std_all": 0.12066958099603653,
"rewards-max/forensics_tempsamp_iou_reward": 0.9464318454265594,
"rewards/forensics_tempsamp_iou_reward": 0.5906163454055786,
"step": 953
},
{
"completion_length": 23.234375,
"epoch": 3.99163179916318,
"grad_norm": 6.977333068847656,
"kl": 0.1875,
"learning_rate": 3.1380753138075314e-09,
"loss": 0.0075,
"reward": 0.9596216678619385,
"reward_std": 0.11678776890039444,
"reward_std_all": 0.1257307082414627,
"rewards-max/forensics_tempsamp_iou_reward": 0.9887369275093079,
"rewards/forensics_tempsamp_iou_reward": 0.6971216797828674,
"step": 954
},
{
"completion_length": 22.75,
"epoch": 3.99581589958159,
"grad_norm": 4.763286113739014,
"kl": 0.18603515625,
"learning_rate": 2.0920502092050206e-09,
"loss": 0.0074,
"reward": 0.9176446199417114,
"reward_std": 0.1233234740793705,
"reward_std_all": 0.11935047432780266,
"rewards-max/forensics_tempsamp_iou_reward": 0.9978218376636505,
"rewards/forensics_tempsamp_iou_reward": 0.6613945960998535,
"step": 955
},
{
"completion_length": 19.92857265472412,
"epoch": 4.0,
"grad_norm": 4.9303717613220215,
"kl": 0.25244140625,
"learning_rate": 1.0460251046025103e-09,
"loss": 0.0095,
"reward": 0.905520498752594,
"reward_std": 0.04561471659690142,
"reward_std_all": 0.06424046494066715,
"rewards-max/forensics_tempsamp_iou_reward": 0.8204344511032104,
"rewards/forensics_tempsamp_iou_reward": 0.6055204570293427,
"step": 956
}
],
"logging_steps": 1.0,
"max_steps": 956,
"num_input_tokens_seen": 0,
"num_train_epochs": 4,
"save_steps": 30,
"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": 1,
"trial_name": null,
"trial_params": null
}