PrimaryGleasonBERT / trainer_state.json
jkefeli's picture
Upload trainer_state.json
dd92e3a
{
"best_metric": 0.9426883191871465,
"best_model_checkpoint": "model_output/prad_primary_output/prad_primary_rs0_clinicalbert_16bsize_512max_tokens_default_lr_roc_optim_20e_01-17-2023_02h-22m/checkpoint-190",
"epoch": 9.5,
"global_step": 190,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.1,
"eval_accuracy": 0.5373134328358209,
"eval_f1": 0.3574758574758575,
"eval_f1_weighted": 0.489443392428467,
"eval_loss": 0.9268761873245239,
"eval_roc_auc": 0.6107379598441218,
"eval_runtime": 1.2207,
"eval_samples_per_second": 54.886,
"eval_steps_per_second": 4.096,
"step": 2
},
{
"epoch": 0.2,
"learning_rate": 4.9500000000000004e-05,
"loss": 0.9899,
"step": 4
},
{
"epoch": 0.2,
"eval_accuracy": 0.4925373134328358,
"eval_f1": 0.22,
"eval_f1_weighted": 0.3250746268656717,
"eval_loss": 0.9124901294708252,
"eval_roc_auc": 0.5582552156247731,
"eval_runtime": 1.2158,
"eval_samples_per_second": 55.108,
"eval_steps_per_second": 4.113,
"step": 4
},
{
"epoch": 0.3,
"eval_accuracy": 0.4925373134328358,
"eval_f1": 0.22,
"eval_f1_weighted": 0.3250746268656717,
"eval_loss": 0.981939971446991,
"eval_roc_auc": 0.4301269582562523,
"eval_runtime": 1.2194,
"eval_samples_per_second": 54.945,
"eval_steps_per_second": 4.1,
"step": 6
},
{
"epoch": 0.4,
"learning_rate": 4.9e-05,
"loss": 0.9933,
"step": 8
},
{
"epoch": 0.4,
"eval_accuracy": 0.4925373134328358,
"eval_f1": 0.22,
"eval_f1_weighted": 0.3250746268656717,
"eval_loss": 0.9560447931289673,
"eval_roc_auc": 0.5123977839606143,
"eval_runtime": 1.2219,
"eval_samples_per_second": 54.832,
"eval_steps_per_second": 4.092,
"step": 8
},
{
"epoch": 0.5,
"eval_accuracy": 0.4925373134328358,
"eval_f1": 0.22,
"eval_f1_weighted": 0.3250746268656717,
"eval_loss": 0.9014336466789246,
"eval_roc_auc": 0.5345296882078375,
"eval_runtime": 1.2192,
"eval_samples_per_second": 54.954,
"eval_steps_per_second": 4.101,
"step": 10
},
{
"epoch": 0.6,
"learning_rate": 4.85e-05,
"loss": 1.0187,
"step": 12
},
{
"epoch": 0.6,
"eval_accuracy": 0.43283582089552236,
"eval_f1": 0.20138888888888887,
"eval_f1_weighted": 0.2615049751243781,
"eval_loss": 0.9236512184143066,
"eval_roc_auc": 0.4107449483209271,
"eval_runtime": 1.2182,
"eval_samples_per_second": 54.998,
"eval_steps_per_second": 4.104,
"step": 12
},
{
"epoch": 0.7,
"eval_accuracy": 0.43283582089552236,
"eval_f1": 0.20138888888888887,
"eval_f1_weighted": 0.2615049751243781,
"eval_loss": 0.9475049376487732,
"eval_roc_auc": 0.622050195769533,
"eval_runtime": 1.2161,
"eval_samples_per_second": 55.094,
"eval_steps_per_second": 4.112,
"step": 14
},
{
"epoch": 0.8,
"learning_rate": 4.8e-05,
"loss": 0.9181,
"step": 16
},
{
"epoch": 0.8,
"eval_accuracy": 0.43283582089552236,
"eval_f1": 0.20138888888888887,
"eval_f1_weighted": 0.2615049751243781,
"eval_loss": 0.9717263579368591,
"eval_roc_auc": 0.6353464355620174,
"eval_runtime": 1.2189,
"eval_samples_per_second": 54.968,
"eval_steps_per_second": 4.102,
"step": 16
},
{
"epoch": 0.9,
"eval_accuracy": 0.43283582089552236,
"eval_f1": 0.20138888888888887,
"eval_f1_weighted": 0.2615049751243781,
"eval_loss": 0.9334698915481567,
"eval_roc_auc": 0.6309403000759409,
"eval_runtime": 1.2236,
"eval_samples_per_second": 54.756,
"eval_steps_per_second": 4.086,
"step": 18
},
{
"epoch": 1.0,
"learning_rate": 4.75e-05,
"loss": 0.8413,
"step": 20
},
{
"epoch": 1.0,
"eval_accuracy": 0.4925373134328358,
"eval_f1": 0.282216999608304,
"eval_f1_weighted": 0.3793693109072733,
"eval_loss": 0.9048787355422974,
"eval_roc_auc": 0.6341481396372107,
"eval_runtime": 1.2194,
"eval_samples_per_second": 54.947,
"eval_steps_per_second": 4.101,
"step": 20
},
{
"epoch": 1.1,
"eval_accuracy": 0.5522388059701493,
"eval_f1": 0.33274578926752846,
"eval_f1_weighted": 0.47230650507743316,
"eval_loss": 0.8940262198448181,
"eval_roc_auc": 0.6367384434369768,
"eval_runtime": 1.2185,
"eval_samples_per_second": 54.986,
"eval_steps_per_second": 4.103,
"step": 22
},
{
"epoch": 1.2,
"learning_rate": 4.7e-05,
"loss": 0.8928,
"step": 24
},
{
"epoch": 1.2,
"eval_accuracy": 0.4925373134328358,
"eval_f1": 0.239192451174018,
"eval_f1_weighted": 0.34958191268804106,
"eval_loss": 0.8931290507316589,
"eval_roc_auc": 0.6255647978074341,
"eval_runtime": 1.2132,
"eval_samples_per_second": 55.227,
"eval_steps_per_second": 4.121,
"step": 24
},
{
"epoch": 1.3,
"eval_accuracy": 0.5522388059701493,
"eval_f1": 0.3825320512820512,
"eval_f1_weighted": 0.52978616532721,
"eval_loss": 0.8918970823287964,
"eval_roc_auc": 0.6465011472546505,
"eval_runtime": 1.2216,
"eval_samples_per_second": 54.848,
"eval_steps_per_second": 4.093,
"step": 26
},
{
"epoch": 1.4,
"learning_rate": 4.6500000000000005e-05,
"loss": 0.9183,
"step": 28
},
{
"epoch": 1.4,
"eval_accuracy": 0.44776119402985076,
"eval_f1": 0.22311661506707947,
"eval_f1_weighted": 0.2932304422161638,
"eval_loss": 0.9188483357429504,
"eval_roc_auc": 0.6651756826192129,
"eval_runtime": 1.2181,
"eval_samples_per_second": 55.005,
"eval_steps_per_second": 4.105,
"step": 28
},
{
"epoch": 1.5,
"eval_accuracy": 0.5074626865671642,
"eval_f1": 0.3401191658391261,
"eval_f1_weighted": 0.46642902666409763,
"eval_loss": 0.8897700309753418,
"eval_roc_auc": 0.6640330788365928,
"eval_runtime": 1.2183,
"eval_samples_per_second": 54.996,
"eval_steps_per_second": 4.104,
"step": 30
},
{
"epoch": 1.6,
"learning_rate": 4.600000000000001e-05,
"loss": 1.0359,
"step": 32
},
{
"epoch": 1.6,
"eval_accuracy": 0.5223880597014925,
"eval_f1": 0.3450960041515309,
"eval_f1_weighted": 0.4845130858421953,
"eval_loss": 0.8961546421051025,
"eval_roc_auc": 0.5583101007352929,
"eval_runtime": 1.2172,
"eval_samples_per_second": 55.043,
"eval_steps_per_second": 4.108,
"step": 32
},
{
"epoch": 1.7,
"eval_accuracy": 0.4925373134328358,
"eval_f1": 0.22,
"eval_f1_weighted": 0.3250746268656717,
"eval_loss": 0.9399242997169495,
"eval_roc_auc": 0.4854997107034975,
"eval_runtime": 1.2176,
"eval_samples_per_second": 55.026,
"eval_steps_per_second": 4.106,
"step": 34
},
{
"epoch": 1.8,
"learning_rate": 4.55e-05,
"loss": 0.8073,
"step": 36
},
{
"epoch": 1.8,
"eval_accuracy": 0.4925373134328358,
"eval_f1": 0.22,
"eval_f1_weighted": 0.3250746268656717,
"eval_loss": 0.9720465540885925,
"eval_roc_auc": 0.5332854471353733,
"eval_runtime": 1.2149,
"eval_samples_per_second": 55.149,
"eval_steps_per_second": 4.116,
"step": 36
},
{
"epoch": 1.9,
"eval_accuracy": 0.47761194029850745,
"eval_f1": 0.23389175257731962,
"eval_f1_weighted": 0.3418699030620096,
"eval_loss": 0.9526171684265137,
"eval_roc_auc": 0.5305002138807847,
"eval_runtime": 1.2175,
"eval_samples_per_second": 55.033,
"eval_steps_per_second": 4.107,
"step": 38
},
{
"epoch": 2.0,
"learning_rate": 4.5e-05,
"loss": 0.9176,
"step": 40
},
{
"epoch": 2.0,
"eval_accuracy": 0.5671641791044776,
"eval_f1": 0.3499132446500868,
"eval_f1_weighted": 0.4950407016392877,
"eval_loss": 0.9059419631958008,
"eval_roc_auc": 0.5067589980465311,
"eval_runtime": 1.215,
"eval_samples_per_second": 55.143,
"eval_steps_per_second": 4.115,
"step": 40
},
{
"epoch": 2.1,
"eval_accuracy": 0.5373134328358209,
"eval_f1": 0.37104798587706633,
"eval_f1_weighted": 0.5126222522438363,
"eval_loss": 0.8836531043052673,
"eval_roc_auc": 0.658280560593416,
"eval_runtime": 1.2194,
"eval_samples_per_second": 54.946,
"eval_steps_per_second": 4.1,
"step": 42
},
{
"epoch": 2.2,
"learning_rate": 4.4500000000000004e-05,
"loss": 0.7806,
"step": 44
},
{
"epoch": 2.2,
"eval_accuracy": 0.5074626865671642,
"eval_f1": 0.3503450489488043,
"eval_f1_weighted": 0.48381347954498094,
"eval_loss": 0.8884266018867493,
"eval_roc_auc": 0.6245352619442165,
"eval_runtime": 1.2159,
"eval_samples_per_second": 55.104,
"eval_steps_per_second": 4.112,
"step": 44
},
{
"epoch": 2.3,
"eval_accuracy": 0.5223880597014925,
"eval_f1": 0.36190017653667145,
"eval_f1_weighted": 0.5020084938811,
"eval_loss": 0.8871902823448181,
"eval_roc_auc": 0.5753152423966572,
"eval_runtime": 1.2182,
"eval_samples_per_second": 54.998,
"eval_steps_per_second": 4.104,
"step": 46
},
{
"epoch": 2.4,
"learning_rate": 4.4000000000000006e-05,
"loss": 0.8738,
"step": 48
},
{
"epoch": 2.4,
"eval_accuracy": 0.5373134328358209,
"eval_f1": 0.3720668595306975,
"eval_f1_weighted": 0.516494192489817,
"eval_loss": 0.8924217224121094,
"eval_roc_auc": 0.5522660805668758,
"eval_runtime": 1.216,
"eval_samples_per_second": 55.099,
"eval_steps_per_second": 4.112,
"step": 48
},
{
"epoch": 2.5,
"eval_accuracy": 0.5074626865671642,
"eval_f1": 0.35161135161135154,
"eval_f1_weighted": 0.4873252783700545,
"eval_loss": 0.9068598747253418,
"eval_roc_auc": 0.545635856249989,
"eval_runtime": 1.2169,
"eval_samples_per_second": 55.059,
"eval_steps_per_second": 4.109,
"step": 50
},
{
"epoch": 2.6,
"learning_rate": 4.35e-05,
"loss": 0.8057,
"step": 52
},
{
"epoch": 2.6,
"eval_accuracy": 0.5373134328358209,
"eval_f1": 0.3625316455696203,
"eval_f1_weighted": 0.507024371811827,
"eval_loss": 0.9044484496116638,
"eval_roc_auc": 0.5214351158888778,
"eval_runtime": 1.218,
"eval_samples_per_second": 55.009,
"eval_steps_per_second": 4.105,
"step": 52
},
{
"epoch": 2.7,
"eval_accuracy": 0.5223880597014925,
"eval_f1": 0.3617433414043583,
"eval_f1_weighted": 0.502135810053847,
"eval_loss": 0.9086884260177612,
"eval_roc_auc": 0.5436828813590747,
"eval_runtime": 1.2157,
"eval_samples_per_second": 55.111,
"eval_steps_per_second": 4.113,
"step": 54
},
{
"epoch": 2.8,
"learning_rate": 4.3e-05,
"loss": 0.8282,
"step": 56
},
{
"epoch": 2.8,
"eval_accuracy": 0.5522388059701493,
"eval_f1": 0.3738271604938272,
"eval_f1_weighted": 0.5141625207296849,
"eval_loss": 0.8964908123016357,
"eval_roc_auc": 0.6378891801507918,
"eval_runtime": 1.22,
"eval_samples_per_second": 54.917,
"eval_steps_per_second": 4.098,
"step": 56
},
{
"epoch": 2.9,
"eval_accuracy": 0.5223880597014925,
"eval_f1": 0.3606237816764133,
"eval_f1_weighted": 0.501440167583137,
"eval_loss": 0.8872122168540955,
"eval_roc_auc": 0.5484830326635979,
"eval_runtime": 1.2197,
"eval_samples_per_second": 54.931,
"eval_steps_per_second": 4.099,
"step": 58
},
{
"epoch": 3.0,
"learning_rate": 4.25e-05,
"loss": 0.8939,
"step": 60
},
{
"epoch": 3.0,
"eval_accuracy": 0.5074626865671642,
"eval_f1": 0.2615979381443299,
"eval_f1_weighted": 0.37907755039236807,
"eval_loss": 1.0024551153182983,
"eval_roc_auc": 0.5364163299657502,
"eval_runtime": 1.2167,
"eval_samples_per_second": 55.066,
"eval_steps_per_second": 4.109,
"step": 60
},
{
"epoch": 3.1,
"eval_accuracy": 0.4925373134328358,
"eval_f1": 0.2556818181818182,
"eval_f1_weighted": 0.37056196291270915,
"eval_loss": 1.0271281003952026,
"eval_roc_auc": 0.5599080164786127,
"eval_runtime": 1.219,
"eval_samples_per_second": 54.964,
"eval_steps_per_second": 4.102,
"step": 62
},
{
"epoch": 3.2,
"learning_rate": 4.2e-05,
"loss": 0.8565,
"step": 64
},
{
"epoch": 3.2,
"eval_accuracy": 0.5671641791044776,
"eval_f1": 0.3650793650793651,
"eval_f1_weighted": 0.513859275053305,
"eval_loss": 0.9313604235649109,
"eval_roc_auc": 0.559152227843963,
"eval_runtime": 1.2182,
"eval_samples_per_second": 55.001,
"eval_steps_per_second": 4.105,
"step": 64
},
{
"epoch": 3.3,
"eval_accuracy": 0.5522388059701493,
"eval_f1": 0.3823717948717949,
"eval_f1_weighted": 0.5314150401836969,
"eval_loss": 0.8814096450805664,
"eval_roc_auc": 0.6012026058913381,
"eval_runtime": 1.2165,
"eval_samples_per_second": 55.076,
"eval_steps_per_second": 4.11,
"step": 66
},
{
"epoch": 3.4,
"learning_rate": 4.15e-05,
"loss": 0.6669,
"step": 68
},
{
"epoch": 3.4,
"eval_accuracy": 0.5671641791044776,
"eval_f1": 0.38861709067188527,
"eval_f1_weighted": 0.5366051347956889,
"eval_loss": 0.9118794798851013,
"eval_roc_auc": 0.5886753942271424,
"eval_runtime": 1.215,
"eval_samples_per_second": 55.144,
"eval_steps_per_second": 4.115,
"step": 68
},
{
"epoch": 3.5,
"eval_accuracy": 0.5671641791044776,
"eval_f1": 0.3925533622211523,
"eval_f1_weighted": 0.5453761524586983,
"eval_loss": 0.895846962928772,
"eval_roc_auc": 0.6272478032128417,
"eval_runtime": 1.218,
"eval_samples_per_second": 55.007,
"eval_steps_per_second": 4.105,
"step": 70
},
{
"epoch": 3.6,
"learning_rate": 4.1e-05,
"loss": 0.7601,
"step": 72
},
{
"epoch": 3.6,
"eval_accuracy": 0.5671641791044776,
"eval_f1": 0.3917874396135265,
"eval_f1_weighted": 0.545078953060783,
"eval_loss": 0.9002715945243835,
"eval_roc_auc": 0.6277288949652808,
"eval_runtime": 1.2172,
"eval_samples_per_second": 55.044,
"eval_steps_per_second": 4.108,
"step": 72
},
{
"epoch": 3.7,
"eval_accuracy": 0.582089552238806,
"eval_f1": 0.4026036644165863,
"eval_f1_weighted": 0.5596583140229422,
"eval_loss": 0.8845404386520386,
"eval_roc_auc": 0.6422917674804015,
"eval_runtime": 1.2198,
"eval_samples_per_second": 54.928,
"eval_steps_per_second": 4.099,
"step": 74
},
{
"epoch": 3.8,
"learning_rate": 4.05e-05,
"loss": 0.792,
"step": 76
},
{
"epoch": 3.8,
"eval_accuracy": 0.6119402985074627,
"eval_f1": 0.4237179487179487,
"eval_f1_weighted": 0.5875143513203215,
"eval_loss": 0.8394505381584167,
"eval_roc_auc": 0.6545318671933383,
"eval_runtime": 1.22,
"eval_samples_per_second": 54.917,
"eval_steps_per_second": 4.098,
"step": 76
},
{
"epoch": 3.9,
"eval_accuracy": 0.5671641791044776,
"eval_f1": 0.3828202581926514,
"eval_f1_weighted": 0.5263824867717025,
"eval_loss": 0.8603011965751648,
"eval_roc_auc": 0.7116721198700833,
"eval_runtime": 1.2172,
"eval_samples_per_second": 55.046,
"eval_steps_per_second": 4.108,
"step": 78
},
{
"epoch": 4.0,
"learning_rate": 4e-05,
"loss": 0.5152,
"step": 80
},
{
"epoch": 4.0,
"eval_accuracy": 0.6567164179104478,
"eval_f1": 0.4573852051219312,
"eval_f1_weighted": 0.633066621869853,
"eval_loss": 0.7896729111671448,
"eval_roc_auc": 0.7033958257608891,
"eval_runtime": 1.2172,
"eval_samples_per_second": 55.045,
"eval_steps_per_second": 4.108,
"step": 80
},
{
"epoch": 4.1,
"eval_accuracy": 0.7014925373134329,
"eval_f1": 0.4861111111111111,
"eval_f1_weighted": 0.6759950248756219,
"eval_loss": 0.7801089286804199,
"eval_roc_auc": 0.6904540328850794,
"eval_runtime": 1.2191,
"eval_samples_per_second": 54.96,
"eval_steps_per_second": 4.101,
"step": 82
},
{
"epoch": 4.2,
"learning_rate": 3.9500000000000005e-05,
"loss": 0.403,
"step": 84
},
{
"epoch": 4.2,
"eval_accuracy": 0.6268656716417911,
"eval_f1": 0.4164096813381928,
"eval_f1_weighted": 0.5832569699860664,
"eval_loss": 0.9066091775894165,
"eval_roc_auc": 0.666676041452625,
"eval_runtime": 1.2185,
"eval_samples_per_second": 54.988,
"eval_steps_per_second": 4.104,
"step": 84
},
{
"epoch": 4.3,
"eval_accuracy": 0.6417910447761194,
"eval_f1": 0.4350649350649351,
"eval_f1_weighted": 0.6075789881760031,
"eval_loss": 0.8394883275032043,
"eval_roc_auc": 0.679015649561526,
"eval_runtime": 1.2195,
"eval_samples_per_second": 54.939,
"eval_steps_per_second": 4.1,
"step": 86
},
{
"epoch": 4.4,
"learning_rate": 3.9000000000000006e-05,
"loss": 0.7288,
"step": 88
},
{
"epoch": 4.4,
"eval_accuracy": 0.7164179104477612,
"eval_f1": 0.5004887585532747,
"eval_f1_weighted": 0.693307655272027,
"eval_loss": 0.7547105550765991,
"eval_roc_auc": 0.7260807241077717,
"eval_runtime": 1.2195,
"eval_samples_per_second": 54.941,
"eval_steps_per_second": 4.1,
"step": 88
},
{
"epoch": 4.5,
"eval_accuracy": 0.6716417910447762,
"eval_f1": 0.4660098522167488,
"eval_f1_weighted": 0.6442320417616352,
"eval_loss": 0.8313462138175964,
"eval_roc_auc": 0.7827001735083298,
"eval_runtime": 1.219,
"eval_samples_per_second": 54.964,
"eval_steps_per_second": 4.102,
"step": 90
},
{
"epoch": 4.6,
"learning_rate": 3.85e-05,
"loss": 0.5865,
"step": 92
},
{
"epoch": 4.6,
"eval_accuracy": 0.7611940298507462,
"eval_f1": 0.53125,
"eval_f1_weighted": 0.7364738805970149,
"eval_loss": 0.692564845085144,
"eval_roc_auc": 0.8177031301314561,
"eval_runtime": 1.2207,
"eval_samples_per_second": 54.885,
"eval_steps_per_second": 4.096,
"step": 92
},
{
"epoch": 4.7,
"eval_accuracy": 0.7761194029850746,
"eval_f1": 0.5424836601307189,
"eval_f1_weighted": 0.7518290898448932,
"eval_loss": 0.6342676877975464,
"eval_roc_auc": 0.8405736805475765,
"eval_runtime": 1.2164,
"eval_samples_per_second": 55.083,
"eval_steps_per_second": 4.111,
"step": 94
},
{
"epoch": 4.8,
"learning_rate": 3.8e-05,
"loss": 0.5263,
"step": 96
},
{
"epoch": 4.8,
"eval_accuracy": 0.7910447761194029,
"eval_f1": 0.553002223869533,
"eval_f1_weighted": 0.7668477479171968,
"eval_loss": 0.6138404011726379,
"eval_roc_auc": 0.8259139155322393,
"eval_runtime": 1.2175,
"eval_samples_per_second": 55.032,
"eval_steps_per_second": 4.107,
"step": 96
},
{
"epoch": 4.9,
"eval_accuracy": 0.8059701492537313,
"eval_f1": 0.5584795321637427,
"eval_f1_weighted": 0.7749410840534171,
"eval_loss": 0.5968928337097168,
"eval_roc_auc": 0.8748494390242875,
"eval_runtime": 1.2175,
"eval_samples_per_second": 55.029,
"eval_steps_per_second": 4.107,
"step": 98
},
{
"epoch": 5.0,
"learning_rate": 3.7500000000000003e-05,
"loss": 0.5847,
"step": 100
},
{
"epoch": 5.0,
"eval_accuracy": 0.8059701492537313,
"eval_f1": 0.5584795321637427,
"eval_f1_weighted": 0.7749410840534171,
"eval_loss": 0.5527657866477966,
"eval_roc_auc": 0.9323773138806697,
"eval_runtime": 1.2195,
"eval_samples_per_second": 54.942,
"eval_steps_per_second": 4.1,
"step": 100
},
{
"epoch": 5.1,
"eval_accuracy": 0.8059701492537313,
"eval_f1": 0.5584795321637427,
"eval_f1_weighted": 0.7749410840534171,
"eval_loss": 0.5388075709342957,
"eval_roc_auc": 0.9383424432250612,
"eval_runtime": 1.2201,
"eval_samples_per_second": 54.913,
"eval_steps_per_second": 4.098,
"step": 102
},
{
"epoch": 5.2,
"learning_rate": 3.7e-05,
"loss": 0.3932,
"step": 104
},
{
"epoch": 5.2,
"eval_accuracy": 0.8059701492537313,
"eval_f1": 0.5584795321637427,
"eval_f1_weighted": 0.7749410840534171,
"eval_loss": 0.5338171124458313,
"eval_roc_auc": 0.9357987315897205,
"eval_runtime": 1.2223,
"eval_samples_per_second": 54.817,
"eval_steps_per_second": 4.091,
"step": 104
},
{
"epoch": 5.3,
"eval_accuracy": 0.7910447761194029,
"eval_f1": 0.5490581490581491,
"eval_f1_weighted": 0.7613627195716748,
"eval_loss": 0.5593022704124451,
"eval_roc_auc": 0.9120323481946476,
"eval_runtime": 1.2192,
"eval_samples_per_second": 54.954,
"eval_steps_per_second": 4.101,
"step": 106
},
{
"epoch": 5.4,
"learning_rate": 3.65e-05,
"loss": 0.2978,
"step": 108
},
{
"epoch": 5.4,
"eval_accuracy": 0.7761194029850746,
"eval_f1": 0.5382716049382715,
"eval_f1_weighted": 0.7467108899944721,
"eval_loss": 0.5857189297676086,
"eval_roc_auc": 0.9198113925251405,
"eval_runtime": 1.2191,
"eval_samples_per_second": 54.96,
"eval_steps_per_second": 4.101,
"step": 108
},
{
"epoch": 5.5,
"eval_accuracy": 0.7761194029850746,
"eval_f1": 0.6409472880061116,
"eval_f1_weighted": 0.7609431832434467,
"eval_loss": 0.6259031295776367,
"eval_roc_auc": 0.9239536321347209,
"eval_runtime": 1.2186,
"eval_samples_per_second": 54.981,
"eval_steps_per_second": 4.103,
"step": 110
},
{
"epoch": 5.6,
"learning_rate": 3.6e-05,
"loss": 0.2672,
"step": 112
},
{
"epoch": 5.6,
"eval_accuracy": 0.8059701492537313,
"eval_f1": 0.6626771920889568,
"eval_f1_weighted": 0.7907103139763368,
"eval_loss": 0.6301503777503967,
"eval_roc_auc": 0.9190969468704965,
"eval_runtime": 1.2171,
"eval_samples_per_second": 55.05,
"eval_steps_per_second": 4.108,
"step": 112
},
{
"epoch": 5.7,
"eval_accuracy": 0.7910447761194029,
"eval_f1": 0.6517503805175039,
"eval_f1_weighted": 0.7759694677298439,
"eval_loss": 0.5792377591133118,
"eval_roc_auc": 0.876367107295521,
"eval_runtime": 1.2176,
"eval_samples_per_second": 55.025,
"eval_steps_per_second": 4.106,
"step": 114
},
{
"epoch": 5.8,
"learning_rate": 3.55e-05,
"loss": 0.2806,
"step": 116
},
{
"epoch": 5.8,
"eval_accuracy": 0.7910447761194029,
"eval_f1": 0.6416725894337836,
"eval_f1_weighted": 0.778988851053899,
"eval_loss": 0.5505416393280029,
"eval_roc_auc": 0.8766408197382413,
"eval_runtime": 1.2235,
"eval_samples_per_second": 54.762,
"eval_steps_per_second": 4.087,
"step": 116
},
{
"epoch": 5.9,
"eval_accuracy": 0.7014925373134329,
"eval_f1": 0.6693895476504172,
"eval_f1_weighted": 0.6933592905040018,
"eval_loss": 0.8218042254447937,
"eval_roc_auc": 0.8531633397655846,
"eval_runtime": 1.2201,
"eval_samples_per_second": 54.913,
"eval_steps_per_second": 4.098,
"step": 118
},
{
"epoch": 6.0,
"learning_rate": 3.5e-05,
"loss": 0.5082,
"step": 120
},
{
"epoch": 6.0,
"eval_accuracy": 0.746268656716418,
"eval_f1": 0.7028011204481793,
"eval_f1_weighted": 0.7403110497930515,
"eval_loss": 0.7941116094589233,
"eval_roc_auc": 0.882679465493187,
"eval_runtime": 1.2184,
"eval_samples_per_second": 54.989,
"eval_steps_per_second": 4.104,
"step": 120
},
{
"epoch": 6.1,
"eval_accuracy": 0.8507462686567164,
"eval_f1": 0.7489281210592686,
"eval_f1_weighted": 0.8448194086314957,
"eval_loss": 0.47939425706863403,
"eval_roc_auc": 0.9171689795255024,
"eval_runtime": 1.2187,
"eval_samples_per_second": 54.976,
"eval_steps_per_second": 4.103,
"step": 122
},
{
"epoch": 6.2,
"learning_rate": 3.45e-05,
"loss": 0.2646,
"step": 124
},
{
"epoch": 6.2,
"eval_accuracy": 0.835820895522388,
"eval_f1": 0.6841817186644773,
"eval_f1_weighted": 0.820184790334044,
"eval_loss": 0.5200427770614624,
"eval_roc_auc": 0.8998004509846013,
"eval_runtime": 1.2187,
"eval_samples_per_second": 54.975,
"eval_steps_per_second": 4.103,
"step": 124
},
{
"epoch": 6.3,
"eval_accuracy": 0.7910447761194029,
"eval_f1": 0.6517503805175039,
"eval_f1_weighted": 0.7759694677298439,
"eval_loss": 0.633881151676178,
"eval_roc_auc": 0.9095164374087981,
"eval_runtime": 1.2183,
"eval_samples_per_second": 54.996,
"eval_steps_per_second": 4.104,
"step": 126
},
{
"epoch": 6.4,
"learning_rate": 3.4000000000000007e-05,
"loss": 0.2178,
"step": 128
},
{
"epoch": 6.4,
"eval_accuracy": 0.7910447761194029,
"eval_f1": 0.5480140774258421,
"eval_f1_weighted": 0.7605862092253663,
"eval_loss": 0.7460798025131226,
"eval_roc_auc": 0.9236619354087335,
"eval_runtime": 1.2193,
"eval_samples_per_second": 54.952,
"eval_steps_per_second": 4.101,
"step": 128
},
{
"epoch": 6.5,
"eval_accuracy": 0.7313432835820896,
"eval_f1": 0.5012973533990659,
"eval_f1_weighted": 0.6975346412721033,
"eval_loss": 0.8184891939163208,
"eval_roc_auc": 0.9135357751398279,
"eval_runtime": 1.2209,
"eval_samples_per_second": 54.878,
"eval_steps_per_second": 4.095,
"step": 130
},
{
"epoch": 6.6,
"learning_rate": 3.35e-05,
"loss": 0.481,
"step": 132
},
{
"epoch": 6.6,
"eval_accuracy": 0.7761194029850746,
"eval_f1": 0.5367088607594936,
"eval_f1_weighted": 0.7452862270923862,
"eval_loss": 0.7749145030975342,
"eval_roc_auc": 0.9241951537539688,
"eval_runtime": 1.2188,
"eval_samples_per_second": 54.972,
"eval_steps_per_second": 4.102,
"step": 132
},
{
"epoch": 6.7,
"eval_accuracy": 0.7910447761194029,
"eval_f1": 0.5480140774258421,
"eval_f1_weighted": 0.7605862092253663,
"eval_loss": 0.70408034324646,
"eval_roc_auc": 0.9353670079123542,
"eval_runtime": 1.2137,
"eval_samples_per_second": 55.205,
"eval_steps_per_second": 4.12,
"step": 134
},
{
"epoch": 6.8,
"learning_rate": 3.3e-05,
"loss": 0.357,
"step": 136
},
{
"epoch": 6.8,
"eval_accuracy": 0.7761194029850746,
"eval_f1": 0.5418752085418752,
"eval_f1_weighted": 0.7520356177072596,
"eval_loss": 0.5984435677528381,
"eval_roc_auc": 0.9422144107122601,
"eval_runtime": 1.2172,
"eval_samples_per_second": 55.046,
"eval_steps_per_second": 4.108,
"step": 136
},
{
"epoch": 6.9,
"eval_accuracy": 0.7910447761194029,
"eval_f1": 0.5568922305764411,
"eval_f1_weighted": 0.7725956682751655,
"eval_loss": 0.6042739152908325,
"eval_roc_auc": 0.8886977232627885,
"eval_runtime": 1.2199,
"eval_samples_per_second": 54.923,
"eval_steps_per_second": 4.099,
"step": 138
},
{
"epoch": 7.0,
"learning_rate": 3.2500000000000004e-05,
"loss": 0.2442,
"step": 140
},
{
"epoch": 7.0,
"eval_accuracy": 0.7761194029850746,
"eval_f1": 0.6232240437158469,
"eval_f1_weighted": 0.7674700269146073,
"eval_loss": 0.6625877618789673,
"eval_roc_auc": 0.885796771407202,
"eval_runtime": 1.2178,
"eval_samples_per_second": 55.016,
"eval_steps_per_second": 4.106,
"step": 140
},
{
"epoch": 7.1,
"eval_accuracy": 0.746268656716418,
"eval_f1": 0.601851851851852,
"eval_f1_weighted": 0.7374437337123906,
"eval_loss": 0.7249980568885803,
"eval_roc_auc": 0.8836999918218473,
"eval_runtime": 1.2196,
"eval_samples_per_second": 54.937,
"eval_steps_per_second": 4.1,
"step": 142
},
{
"epoch": 7.2,
"learning_rate": 3.2000000000000005e-05,
"loss": 0.2895,
"step": 144
},
{
"epoch": 7.2,
"eval_accuracy": 0.7611940298507462,
"eval_f1": 0.6125672043010751,
"eval_f1_weighted": 0.7525126384207992,
"eval_loss": 0.7177485227584839,
"eval_roc_auc": 0.8988318738409138,
"eval_runtime": 1.219,
"eval_samples_per_second": 54.963,
"eval_steps_per_second": 4.102,
"step": 144
},
{
"epoch": 7.3,
"eval_accuracy": 0.8059701492537313,
"eval_f1": 0.6442283244342741,
"eval_f1_weighted": 0.7971669114382322,
"eval_loss": 0.6158971786499023,
"eval_roc_auc": 0.9257433466065693,
"eval_runtime": 1.2179,
"eval_samples_per_second": 55.013,
"eval_steps_per_second": 4.105,
"step": 146
},
{
"epoch": 7.4,
"learning_rate": 3.15e-05,
"loss": 0.1247,
"step": 148
},
{
"epoch": 7.4,
"eval_accuracy": 0.7761194029850746,
"eval_f1": 0.6135265700483091,
"eval_f1_weighted": 0.7752109020116807,
"eval_loss": 0.5960801243782043,
"eval_roc_auc": 0.9299025860599263,
"eval_runtime": 1.2186,
"eval_samples_per_second": 54.98,
"eval_steps_per_second": 4.103,
"step": 148
},
{
"epoch": 7.5,
"eval_accuracy": 0.7910447761194029,
"eval_f1": 0.6803418803418803,
"eval_f1_weighted": 0.7898966704936854,
"eval_loss": 0.6297043561935425,
"eval_roc_auc": 0.928043745151505,
"eval_runtime": 1.2207,
"eval_samples_per_second": 54.884,
"eval_steps_per_second": 4.096,
"step": 150
},
{
"epoch": 7.6,
"learning_rate": 3.1e-05,
"loss": 0.1401,
"step": 152
},
{
"epoch": 7.6,
"eval_accuracy": 0.7761194029850746,
"eval_f1": 0.6214568794692857,
"eval_f1_weighted": 0.7664838768164862,
"eval_loss": 0.7288922667503357,
"eval_roc_auc": 0.9169795844990506,
"eval_runtime": 1.2177,
"eval_samples_per_second": 55.02,
"eval_steps_per_second": 4.106,
"step": 152
},
{
"epoch": 7.7,
"eval_accuracy": 0.7611940298507462,
"eval_f1": 0.5344212136664966,
"eval_f1_weighted": 0.7423565497347531,
"eval_loss": 0.7684075832366943,
"eval_roc_auc": 0.9071037779761267,
"eval_runtime": 1.2191,
"eval_samples_per_second": 54.959,
"eval_steps_per_second": 4.101,
"step": 154
},
{
"epoch": 7.8,
"learning_rate": 3.05e-05,
"loss": 0.1425,
"step": 156
},
{
"epoch": 7.8,
"eval_accuracy": 0.7611940298507462,
"eval_f1": 0.5344212136664966,
"eval_f1_weighted": 0.7423565497347531,
"eval_loss": 0.7739881873130798,
"eval_roc_auc": 0.9037702780185827,
"eval_runtime": 1.2219,
"eval_samples_per_second": 54.832,
"eval_steps_per_second": 4.092,
"step": 156
},
{
"epoch": 7.9,
"eval_accuracy": 0.8059701492537313,
"eval_f1": 0.6499480563407505,
"eval_f1_weighted": 0.792322362962314,
"eval_loss": 0.7463679909706116,
"eval_roc_auc": 0.9108798026167514,
"eval_runtime": 1.2161,
"eval_samples_per_second": 55.094,
"eval_steps_per_second": 4.111,
"step": 158
},
{
"epoch": 8.0,
"learning_rate": 3e-05,
"loss": 0.0541,
"step": 160
},
{
"epoch": 8.0,
"eval_accuracy": 0.7910447761194029,
"eval_f1": 0.6396825396825397,
"eval_f1_weighted": 0.7780383795309169,
"eval_loss": 0.6964292526245117,
"eval_roc_auc": 0.924278504123671,
"eval_runtime": 1.2218,
"eval_samples_per_second": 54.838,
"eval_steps_per_second": 4.092,
"step": 160
},
{
"epoch": 8.1,
"eval_accuracy": 0.7910447761194029,
"eval_f1": 0.6396825396825397,
"eval_f1_weighted": 0.7780383795309169,
"eval_loss": 0.6750349402427673,
"eval_roc_auc": 0.9306260238562039,
"eval_runtime": 1.2115,
"eval_samples_per_second": 55.304,
"eval_steps_per_second": 4.127,
"step": 162
},
{
"epoch": 8.2,
"learning_rate": 2.95e-05,
"loss": 0.0735,
"step": 164
},
{
"epoch": 8.2,
"eval_accuracy": 0.7910447761194029,
"eval_f1": 0.6396825396825397,
"eval_f1_weighted": 0.7780383795309169,
"eval_loss": 0.6345735788345337,
"eval_roc_auc": 0.9307606763941116,
"eval_runtime": 1.2207,
"eval_samples_per_second": 54.885,
"eval_steps_per_second": 4.096,
"step": 164
},
{
"epoch": 8.3,
"eval_accuracy": 0.8059701492537313,
"eval_f1": 0.6507377598926894,
"eval_f1_weighted": 0.7930943872188354,
"eval_loss": 0.6417430639266968,
"eval_roc_auc": 0.9207915603132855,
"eval_runtime": 1.2167,
"eval_samples_per_second": 55.069,
"eval_steps_per_second": 4.11,
"step": 166
},
{
"epoch": 8.4,
"learning_rate": 2.9e-05,
"loss": 0.0701,
"step": 168
},
{
"epoch": 8.4,
"eval_accuracy": 0.8208955223880597,
"eval_f1": 0.6735030063421464,
"eval_f1_weighted": 0.8055976603269788,
"eval_loss": 0.6598219871520996,
"eval_roc_auc": 0.9178574505876599,
"eval_runtime": 1.2178,
"eval_samples_per_second": 55.017,
"eval_steps_per_second": 4.106,
"step": 168
},
{
"epoch": 8.5,
"eval_accuracy": 0.8208955223880597,
"eval_f1": 0.6735030063421464,
"eval_f1_weighted": 0.8055976603269788,
"eval_loss": 0.674333393573761,
"eval_roc_auc": 0.9147662966739546,
"eval_runtime": 1.2139,
"eval_samples_per_second": 55.194,
"eval_steps_per_second": 4.119,
"step": 170
},
{
"epoch": 8.6,
"learning_rate": 2.8499999999999998e-05,
"loss": 0.0415,
"step": 172
},
{
"epoch": 8.6,
"eval_accuracy": 0.8208955223880597,
"eval_f1": 0.6735030063421464,
"eval_f1_weighted": 0.8055976603269788,
"eval_loss": 0.6726630330085754,
"eval_roc_auc": 0.9180612295621405,
"eval_runtime": 1.2166,
"eval_samples_per_second": 55.074,
"eval_steps_per_second": 4.11,
"step": 172
},
{
"epoch": 8.7,
"eval_accuracy": 0.835820895522388,
"eval_f1": 0.6841817186644773,
"eval_f1_weighted": 0.820184790334044,
"eval_loss": 0.6476911902427673,
"eval_roc_auc": 0.9287753506648254,
"eval_runtime": 1.2178,
"eval_samples_per_second": 55.016,
"eval_steps_per_second": 4.106,
"step": 174
},
{
"epoch": 8.8,
"learning_rate": 2.8000000000000003e-05,
"loss": 0.1207,
"step": 176
},
{
"epoch": 8.8,
"eval_accuracy": 0.8507462686567164,
"eval_f1": 0.6947514943093426,
"eval_f1_weighted": 0.8346513563354928,
"eval_loss": 0.6322030425071716,
"eval_roc_auc": 0.9343940407946855,
"eval_runtime": 1.2175,
"eval_samples_per_second": 55.032,
"eval_steps_per_second": 4.107,
"step": 176
},
{
"epoch": 8.9,
"eval_accuracy": 0.8507462686567164,
"eval_f1": 0.6947514943093426,
"eval_f1_weighted": 0.8346513563354928,
"eval_loss": 0.631219744682312,
"eval_roc_auc": 0.9355877867306113,
"eval_runtime": 1.2171,
"eval_samples_per_second": 55.047,
"eval_steps_per_second": 4.108,
"step": 178
},
{
"epoch": 9.0,
"learning_rate": 2.7500000000000004e-05,
"loss": 0.0677,
"step": 180
},
{
"epoch": 9.0,
"eval_accuracy": 0.835820895522388,
"eval_f1": 0.6843137254901962,
"eval_f1_weighted": 0.8201053555750658,
"eval_loss": 0.6727584600448608,
"eval_roc_auc": 0.9294350538309254,
"eval_runtime": 1.213,
"eval_samples_per_second": 55.237,
"eval_steps_per_second": 4.122,
"step": 180
},
{
"epoch": 9.1,
"eval_accuracy": 0.8059701492537313,
"eval_f1": 0.6633986928104575,
"eval_f1_weighted": 0.7911911033069944,
"eval_loss": 0.7204322218894958,
"eval_roc_auc": 0.9322838617007921,
"eval_runtime": 1.2192,
"eval_samples_per_second": 54.955,
"eval_steps_per_second": 4.101,
"step": 182
},
{
"epoch": 9.2,
"learning_rate": 2.7000000000000002e-05,
"loss": 0.0334,
"step": 184
},
{
"epoch": 9.2,
"eval_accuracy": 0.8059701492537313,
"eval_f1": 0.6633986928104575,
"eval_f1_weighted": 0.7911911033069944,
"eval_loss": 0.6748175621032715,
"eval_roc_auc": 0.9319867731684095,
"eval_runtime": 1.2166,
"eval_samples_per_second": 55.073,
"eval_steps_per_second": 4.11,
"step": 184
},
{
"epoch": 9.3,
"eval_accuracy": 0.8059701492537313,
"eval_f1": 0.6631636562671046,
"eval_f1_weighted": 0.7911869225302061,
"eval_loss": 0.6425347924232483,
"eval_roc_auc": 0.9327048191582917,
"eval_runtime": 1.2192,
"eval_samples_per_second": 54.952,
"eval_steps_per_second": 4.101,
"step": 186
},
{
"epoch": 9.4,
"learning_rate": 2.6500000000000004e-05,
"loss": 0.0184,
"step": 188
},
{
"epoch": 9.4,
"eval_accuracy": 0.8208955223880597,
"eval_f1": 0.7118127444214402,
"eval_f1_weighted": 0.8170045219039379,
"eval_loss": 0.6353668570518494,
"eval_roc_auc": 0.9364646683797305,
"eval_runtime": 1.2147,
"eval_samples_per_second": 55.155,
"eval_steps_per_second": 4.116,
"step": 188
},
{
"epoch": 9.5,
"eval_accuracy": 0.835820895522388,
"eval_f1": 0.7686429512516469,
"eval_f1_weighted": 0.8350775765441567,
"eval_loss": 0.6435410380363464,
"eval_roc_auc": 0.9426883191871465,
"eval_runtime": 1.2183,
"eval_samples_per_second": 54.995,
"eval_steps_per_second": 4.104,
"step": 190
}
],
"max_steps": 400,
"num_train_epochs": 20,
"total_flos": 769080520018944.0,
"trial_name": null,
"trial_params": null
}