BERTweet-WNUT17 / trainer_state.json
emilys's picture
Upload
ee5640f
raw
history blame
36.1 kB
{
"best_metric": 0.6788413098236775,
"best_model_checkpoint": "trained_models/wnut_17/checkpoint-9000",
"epoch": 21.176470588235293,
"global_step": 9000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.24,
"eval_accuracy_score": 0.9205491641772071,
"eval_f1": 0.0,
"eval_loss": 0.5049907565116882,
"eval_precision": 0.0,
"eval_recall": 0.0,
"eval_runtime": 6.761,
"eval_samples_per_second": 149.239,
"eval_steps_per_second": 18.784,
"step": 100
},
{
"epoch": 0.47,
"eval_accuracy_score": 0.9205491641772071,
"eval_f1": 0.0,
"eval_loss": 0.48477333784103394,
"eval_precision": 0.0,
"eval_recall": 0.0,
"eval_runtime": 6.7798,
"eval_samples_per_second": 148.825,
"eval_steps_per_second": 18.732,
"step": 200
},
{
"epoch": 0.71,
"eval_accuracy_score": 0.9205491641772071,
"eval_f1": 0.0,
"eval_loss": 0.3834511935710907,
"eval_precision": 0.0,
"eval_recall": 0.0,
"eval_runtime": 6.7494,
"eval_samples_per_second": 149.494,
"eval_steps_per_second": 18.816,
"step": 300
},
{
"epoch": 0.94,
"eval_accuracy_score": 0.9218203775503718,
"eval_f1": 0.016241299303944315,
"eval_loss": 0.3428375720977783,
"eval_precision": 0.2692307692307692,
"eval_recall": 0.008373205741626795,
"eval_runtime": 6.7397,
"eval_samples_per_second": 149.711,
"eval_steps_per_second": 18.844,
"step": 400
},
{
"epoch": 1.18,
"learning_rate": 4.803921568627452e-06,
"loss": 0.4388,
"step": 500
},
{
"epoch": 1.18,
"eval_accuracy_score": 0.9387275154134621,
"eval_f1": 0.3550877192982456,
"eval_loss": 0.31836700439453125,
"eval_precision": 0.4295415959252971,
"eval_recall": 0.3026315789473684,
"eval_runtime": 6.749,
"eval_samples_per_second": 149.504,
"eval_steps_per_second": 18.818,
"step": 500
},
{
"epoch": 1.41,
"eval_accuracy_score": 0.9413335028284497,
"eval_f1": 0.40223463687150846,
"eval_loss": 0.3077085316181183,
"eval_precision": 0.48322147651006714,
"eval_recall": 0.3444976076555024,
"eval_runtime": 7.2416,
"eval_samples_per_second": 139.333,
"eval_steps_per_second": 17.537,
"step": 600
},
{
"epoch": 1.65,
"eval_accuracy_score": 0.9449564609419691,
"eval_f1": 0.46935933147632314,
"eval_loss": 0.29629671573638916,
"eval_precision": 0.5616666666666666,
"eval_recall": 0.40311004784689,
"eval_runtime": 6.7314,
"eval_samples_per_second": 149.895,
"eval_steps_per_second": 18.867,
"step": 700
},
{
"epoch": 1.88,
"eval_accuracy_score": 0.9471810843450074,
"eval_f1": 0.5139186295503213,
"eval_loss": 0.2875679135322571,
"eval_precision": 0.6371681415929203,
"eval_recall": 0.430622009569378,
"eval_runtime": 6.7537,
"eval_samples_per_second": 149.401,
"eval_steps_per_second": 18.805,
"step": 800
},
{
"epoch": 2.12,
"eval_accuracy_score": 0.949278586410729,
"eval_f1": 0.5329849771391247,
"eval_loss": 0.27145618200302124,
"eval_precision": 0.5870503597122302,
"eval_recall": 0.4880382775119617,
"eval_runtime": 6.7363,
"eval_samples_per_second": 149.786,
"eval_steps_per_second": 18.853,
"step": 900
},
{
"epoch": 2.35,
"learning_rate": 4.607843137254902e-06,
"loss": 0.1974,
"step": 1000
},
{
"epoch": 2.35,
"eval_accuracy_score": 0.9492150257420708,
"eval_f1": 0.5209424083769634,
"eval_loss": 0.2602783441543579,
"eval_precision": 0.5751445086705202,
"eval_recall": 0.47607655502392343,
"eval_runtime": 6.736,
"eval_samples_per_second": 149.792,
"eval_steps_per_second": 18.854,
"step": 1000
},
{
"epoch": 2.59,
"eval_accuracy_score": 0.9504862391152354,
"eval_f1": 0.5558510638297872,
"eval_loss": 0.24513839185237885,
"eval_precision": 0.625748502994012,
"eval_recall": 0.5,
"eval_runtime": 7.332,
"eval_samples_per_second": 137.616,
"eval_steps_per_second": 17.321,
"step": 1100
},
{
"epoch": 2.82,
"eval_accuracy_score": 0.9520116951630331,
"eval_f1": 0.5626566416040101,
"eval_loss": 0.23386946320533752,
"eval_precision": 0.5907894736842105,
"eval_recall": 0.5370813397129187,
"eval_runtime": 6.737,
"eval_samples_per_second": 149.771,
"eval_steps_per_second": 18.851,
"step": 1200
},
{
"epoch": 3.06,
"eval_accuracy_score": 0.9529651051929067,
"eval_f1": 0.5635782747603835,
"eval_loss": 0.23247480392456055,
"eval_precision": 0.6049382716049383,
"eval_recall": 0.527511961722488,
"eval_runtime": 6.7326,
"eval_samples_per_second": 149.868,
"eval_steps_per_second": 18.864,
"step": 1300
},
{
"epoch": 3.29,
"eval_accuracy_score": 0.9557617746138689,
"eval_f1": 0.5892526250772081,
"eval_loss": 0.2185893952846527,
"eval_precision": 0.6091954022988506,
"eval_recall": 0.5705741626794258,
"eval_runtime": 6.7427,
"eval_samples_per_second": 149.643,
"eval_steps_per_second": 18.835,
"step": 1400
},
{
"epoch": 3.53,
"learning_rate": 4.411764705882353e-06,
"loss": 0.1395,
"step": 1500
},
{
"epoch": 3.53,
"eval_accuracy_score": 0.9572236699930083,
"eval_f1": 0.6000000000000001,
"eval_loss": 0.2226746827363968,
"eval_precision": 0.6464088397790055,
"eval_recall": 0.5598086124401914,
"eval_runtime": 6.7181,
"eval_samples_per_second": 150.191,
"eval_steps_per_second": 18.904,
"step": 1500
},
{
"epoch": 3.76,
"eval_accuracy_score": 0.9590033687154389,
"eval_f1": 0.598405885959534,
"eval_loss": 0.21203213930130005,
"eval_precision": 0.6138364779874214,
"eval_recall": 0.583732057416268,
"eval_runtime": 6.7595,
"eval_samples_per_second": 149.27,
"eval_steps_per_second": 18.788,
"step": 1600
},
{
"epoch": 4.0,
"eval_accuracy_score": 0.9586855653721477,
"eval_f1": 0.618421052631579,
"eval_loss": 0.22468985617160797,
"eval_precision": 0.6871345029239766,
"eval_recall": 0.562200956937799,
"eval_runtime": 6.7215,
"eval_samples_per_second": 150.115,
"eval_steps_per_second": 18.895,
"step": 1700
},
{
"epoch": 4.24,
"eval_accuracy_score": 0.9599567787453124,
"eval_f1": 0.6330749354005167,
"eval_loss": 0.2214759886264801,
"eval_precision": 0.6882022471910112,
"eval_recall": 0.5861244019138756,
"eval_runtime": 6.7281,
"eval_samples_per_second": 149.969,
"eval_steps_per_second": 18.876,
"step": 1800
},
{
"epoch": 4.47,
"eval_accuracy_score": 0.96148223479311,
"eval_f1": 0.6507115135834413,
"eval_loss": 0.21751971542835236,
"eval_precision": 0.7084507042253522,
"eval_recall": 0.6016746411483254,
"eval_runtime": 6.7505,
"eval_samples_per_second": 149.471,
"eval_steps_per_second": 18.813,
"step": 1900
},
{
"epoch": 4.71,
"learning_rate": 4.215686274509805e-06,
"loss": 0.1057,
"step": 2000
},
{
"epoch": 4.71,
"eval_accuracy_score": 0.9605923854318947,
"eval_f1": 0.6401062416998671,
"eval_loss": 0.22828274965286255,
"eval_precision": 0.7194029850746269,
"eval_recall": 0.5765550239234449,
"eval_runtime": 6.7552,
"eval_samples_per_second": 149.366,
"eval_steps_per_second": 18.8,
"step": 2000
},
{
"epoch": 4.94,
"eval_accuracy_score": 0.9621814021483506,
"eval_f1": 0.6484424666242848,
"eval_loss": 0.2120533287525177,
"eval_precision": 0.6919945725915875,
"eval_recall": 0.6100478468899522,
"eval_runtime": 6.7607,
"eval_samples_per_second": 149.245,
"eval_steps_per_second": 18.785,
"step": 2100
},
{
"epoch": 5.18,
"eval_accuracy_score": 0.9617364774677429,
"eval_f1": 0.6427196921103272,
"eval_loss": 0.21272720396518707,
"eval_precision": 0.6929460580912863,
"eval_recall": 0.5992822966507177,
"eval_runtime": 6.823,
"eval_samples_per_second": 147.882,
"eval_steps_per_second": 18.613,
"step": 2200
},
{
"epoch": 5.41,
"eval_accuracy_score": 0.9614186741244518,
"eval_f1": 0.6440460947503202,
"eval_loss": 0.21506403386592865,
"eval_precision": 0.6928374655647382,
"eval_recall": 0.6016746411483254,
"eval_runtime": 6.7509,
"eval_samples_per_second": 149.462,
"eval_steps_per_second": 18.812,
"step": 2300
},
{
"epoch": 5.65,
"eval_accuracy_score": 0.9625627661603,
"eval_f1": 0.6498740554156172,
"eval_loss": 0.20958468317985535,
"eval_precision": 0.6861702127659575,
"eval_recall": 0.6172248803827751,
"eval_runtime": 6.7421,
"eval_samples_per_second": 149.656,
"eval_steps_per_second": 18.837,
"step": 2400
},
{
"epoch": 5.88,
"learning_rate": 4.019607843137255e-06,
"loss": 0.0811,
"step": 2500
},
{
"epoch": 5.88,
"eval_accuracy_score": 0.9624356448229835,
"eval_f1": 0.6591337099811676,
"eval_loss": 0.2008962631225586,
"eval_precision": 0.6935270805812418,
"eval_recall": 0.6279904306220095,
"eval_runtime": 6.7089,
"eval_samples_per_second": 150.397,
"eval_steps_per_second": 18.93,
"step": 2500
},
{
"epoch": 6.12,
"eval_accuracy_score": 0.9621178414796924,
"eval_f1": 0.6448717948717948,
"eval_loss": 0.21091684699058533,
"eval_precision": 0.6947513812154696,
"eval_recall": 0.6016746411483254,
"eval_runtime": 6.7609,
"eval_samples_per_second": 149.241,
"eval_steps_per_second": 18.785,
"step": 2600
},
{
"epoch": 6.35,
"eval_accuracy_score": 0.9621178414796924,
"eval_f1": 0.6499057196731616,
"eval_loss": 0.21349507570266724,
"eval_precision": 0.6847682119205298,
"eval_recall": 0.618421052631579,
"eval_runtime": 6.7376,
"eval_samples_per_second": 149.757,
"eval_steps_per_second": 18.849,
"step": 2700
},
{
"epoch": 6.59,
"eval_accuracy_score": 0.9620542808110342,
"eval_f1": 0.6467218332272437,
"eval_loss": 0.2102096974849701,
"eval_precision": 0.691156462585034,
"eval_recall": 0.6076555023923444,
"eval_runtime": 6.7394,
"eval_samples_per_second": 149.717,
"eval_steps_per_second": 18.844,
"step": 2800
},
{
"epoch": 6.82,
"eval_accuracy_score": 0.9630076908409076,
"eval_f1": 0.6658385093167701,
"eval_loss": 0.2015107125043869,
"eval_precision": 0.6925064599483204,
"eval_recall": 0.6411483253588517,
"eval_runtime": 6.7773,
"eval_samples_per_second": 148.879,
"eval_steps_per_second": 18.739,
"step": 2900
},
{
"epoch": 7.06,
"learning_rate": 3.8235294117647055e-06,
"loss": 0.0674,
"step": 3000
},
{
"epoch": 7.06,
"eval_accuracy_score": 0.96364329752749,
"eval_f1": 0.6624921728240452,
"eval_loss": 0.20179474353790283,
"eval_precision": 0.695137976346912,
"eval_recall": 0.6327751196172249,
"eval_runtime": 6.8013,
"eval_samples_per_second": 148.354,
"eval_steps_per_second": 18.673,
"step": 3000
},
{
"epoch": 7.29,
"eval_accuracy_score": 0.9629441301722494,
"eval_f1": 0.6454431960049937,
"eval_loss": 0.20865464210510254,
"eval_precision": 0.674934725848564,
"eval_recall": 0.618421052631579,
"eval_runtime": 6.7901,
"eval_samples_per_second": 148.599,
"eval_steps_per_second": 18.704,
"step": 3100
},
{
"epoch": 7.53,
"eval_accuracy_score": 0.9631983728468824,
"eval_f1": 0.6507237256135934,
"eval_loss": 0.2077988088130951,
"eval_precision": 0.6865869853917662,
"eval_recall": 0.618421052631579,
"eval_runtime": 6.7566,
"eval_samples_per_second": 149.335,
"eval_steps_per_second": 18.796,
"step": 3200
},
{
"epoch": 7.76,
"eval_accuracy_score": 0.9626898874976165,
"eval_f1": 0.6499678869621066,
"eval_loss": 0.2155311405658722,
"eval_precision": 0.7018030513176144,
"eval_recall": 0.6052631578947368,
"eval_runtime": 6.7572,
"eval_samples_per_second": 149.321,
"eval_steps_per_second": 18.795,
"step": 3300
},
{
"epoch": 8.0,
"eval_accuracy_score": 0.9624356448229835,
"eval_f1": 0.6524271844660194,
"eval_loss": 0.21822784841060638,
"eval_precision": 0.7108603667136812,
"eval_recall": 0.6028708133971292,
"eval_runtime": 6.7807,
"eval_samples_per_second": 148.804,
"eval_steps_per_second": 18.73,
"step": 3400
},
{
"epoch": 8.24,
"learning_rate": 3.6274509803921573e-06,
"loss": 0.0559,
"step": 3500
},
{
"epoch": 8.24,
"eval_accuracy_score": 0.9631348121782242,
"eval_f1": 0.659062103929024,
"eval_loss": 0.21273677051067352,
"eval_precision": 0.7008086253369272,
"eval_recall": 0.6220095693779905,
"eval_runtime": 6.7444,
"eval_samples_per_second": 149.606,
"eval_steps_per_second": 18.831,
"step": 3500
},
{
"epoch": 8.47,
"eval_accuracy_score": 0.9634526155215153,
"eval_f1": 0.6608478802992518,
"eval_loss": 0.20193149149417877,
"eval_precision": 0.6901041666666666,
"eval_recall": 0.6339712918660287,
"eval_runtime": 8.1777,
"eval_samples_per_second": 123.384,
"eval_steps_per_second": 15.53,
"step": 3600
},
{
"epoch": 8.71,
"eval_accuracy_score": 0.9634526155215153,
"eval_f1": 0.6587942821628341,
"eval_loss": 0.20160645246505737,
"eval_precision": 0.685640362225097,
"eval_recall": 0.6339712918660287,
"eval_runtime": 6.7668,
"eval_samples_per_second": 149.11,
"eval_steps_per_second": 18.768,
"step": 3700
},
{
"epoch": 8.94,
"eval_accuracy_score": 0.9634526155215153,
"eval_f1": 0.662553979025293,
"eval_loss": 0.20665320754051208,
"eval_precision": 0.6840764331210191,
"eval_recall": 0.6423444976076556,
"eval_runtime": 6.7328,
"eval_samples_per_second": 149.863,
"eval_steps_per_second": 18.863,
"step": 3800
},
{
"epoch": 9.18,
"eval_accuracy_score": 0.9629441301722494,
"eval_f1": 0.6566775244299674,
"eval_loss": 0.22287048399448395,
"eval_precision": 0.721030042918455,
"eval_recall": 0.6028708133971292,
"eval_runtime": 6.7851,
"eval_samples_per_second": 148.709,
"eval_steps_per_second": 18.718,
"step": 3900
},
{
"epoch": 9.41,
"learning_rate": 3.431372549019608e-06,
"loss": 0.0459,
"step": 4000
},
{
"epoch": 9.41,
"eval_accuracy_score": 0.9635797368588318,
"eval_f1": 0.6574827802128992,
"eval_loss": 0.20987075567245483,
"eval_precision": 0.6898817345597897,
"eval_recall": 0.6279904306220095,
"eval_runtime": 6.7469,
"eval_samples_per_second": 149.549,
"eval_steps_per_second": 18.823,
"step": 4000
},
{
"epoch": 9.65,
"eval_accuracy_score": 0.9628805695035911,
"eval_f1": 0.6568690095846645,
"eval_loss": 0.22190582752227783,
"eval_precision": 0.7050754458161865,
"eval_recall": 0.6148325358851675,
"eval_runtime": 6.7586,
"eval_samples_per_second": 149.292,
"eval_steps_per_second": 18.791,
"step": 4100
},
{
"epoch": 9.88,
"eval_accuracy_score": 0.9632619335155406,
"eval_f1": 0.6525265127885215,
"eval_loss": 0.21028834581375122,
"eval_precision": 0.681877444589309,
"eval_recall": 0.6255980861244019,
"eval_runtime": 6.7583,
"eval_samples_per_second": 149.298,
"eval_steps_per_second": 18.792,
"step": 4200
},
{
"epoch": 10.12,
"eval_accuracy_score": 0.9640882222080976,
"eval_f1": 0.6692111959287532,
"eval_loss": 0.2170829027891159,
"eval_precision": 0.7146739130434783,
"eval_recall": 0.6291866028708134,
"eval_runtime": 6.7544,
"eval_samples_per_second": 149.383,
"eval_steps_per_second": 18.802,
"step": 4300
},
{
"epoch": 10.35,
"eval_accuracy_score": 0.9630712515095659,
"eval_f1": 0.6495619524405507,
"eval_loss": 0.2193835973739624,
"eval_precision": 0.6811023622047244,
"eval_recall": 0.6208133971291866,
"eval_runtime": 6.7362,
"eval_samples_per_second": 149.787,
"eval_steps_per_second": 18.853,
"step": 4400
},
{
"epoch": 10.59,
"learning_rate": 3.2352941176470594e-06,
"loss": 0.0399,
"step": 4500
},
{
"epoch": 10.59,
"eval_accuracy_score": 0.9634526155215153,
"eval_f1": 0.6556543837357053,
"eval_loss": 0.21863853931427002,
"eval_precision": 0.6991869918699187,
"eval_recall": 0.6172248803827751,
"eval_runtime": 6.769,
"eval_samples_per_second": 149.062,
"eval_steps_per_second": 18.762,
"step": 4500
},
{
"epoch": 10.82,
"eval_accuracy_score": 0.9637068581961482,
"eval_f1": 0.6624685138539044,
"eval_loss": 0.21445031464099884,
"eval_precision": 0.699468085106383,
"eval_recall": 0.6291866028708134,
"eval_runtime": 6.7298,
"eval_samples_per_second": 149.93,
"eval_steps_per_second": 18.871,
"step": 4600
},
{
"epoch": 11.06,
"eval_accuracy_score": 0.9644695862200471,
"eval_f1": 0.6759142496847416,
"eval_loss": 0.20991285145282745,
"eval_precision": 0.7146666666666667,
"eval_recall": 0.6411483253588517,
"eval_runtime": 6.734,
"eval_samples_per_second": 149.837,
"eval_steps_per_second": 18.86,
"step": 4700
},
{
"epoch": 11.29,
"eval_accuracy_score": 0.96364329752749,
"eval_f1": 0.6649778340721977,
"eval_loss": 0.22495105862617493,
"eval_precision": 0.7065948855989233,
"eval_recall": 0.6279904306220095,
"eval_runtime": 6.7408,
"eval_samples_per_second": 149.687,
"eval_steps_per_second": 18.841,
"step": 4800
},
{
"epoch": 11.53,
"eval_accuracy_score": 0.9639611008707811,
"eval_f1": 0.6670894102726697,
"eval_loss": 0.22267203032970428,
"eval_precision": 0.7098515519568152,
"eval_recall": 0.6291866028708134,
"eval_runtime": 6.7689,
"eval_samples_per_second": 149.064,
"eval_steps_per_second": 18.762,
"step": 4900
},
{
"epoch": 11.76,
"learning_rate": 3.03921568627451e-06,
"loss": 0.0368,
"step": 5000
},
{
"epoch": 11.76,
"eval_accuracy_score": 0.9637068581961482,
"eval_f1": 0.6588235294117647,
"eval_loss": 0.21161217987537384,
"eval_precision": 0.6829268292682927,
"eval_recall": 0.6363636363636364,
"eval_runtime": 6.7536,
"eval_samples_per_second": 149.402,
"eval_steps_per_second": 18.805,
"step": 5000
},
{
"epoch": 12.0,
"eval_accuracy_score": 0.9625627661603,
"eval_f1": 0.6593830334190232,
"eval_loss": 0.227540984749794,
"eval_precision": 0.7125,
"eval_recall": 0.6136363636363636,
"eval_runtime": 6.7584,
"eval_samples_per_second": 149.295,
"eval_steps_per_second": 18.791,
"step": 5100
},
{
"epoch": 12.24,
"eval_accuracy_score": 0.9628805695035911,
"eval_f1": 0.6640471512770137,
"eval_loss": 0.23519518971443176,
"eval_precision": 0.7337192474674384,
"eval_recall": 0.6064593301435407,
"eval_runtime": 6.7729,
"eval_samples_per_second": 148.976,
"eval_steps_per_second": 18.751,
"step": 5200
},
{
"epoch": 12.47,
"eval_accuracy_score": 0.9628170088349329,
"eval_f1": 0.6679462571976967,
"eval_loss": 0.2271033227443695,
"eval_precision": 0.7180192572214581,
"eval_recall": 0.6244019138755981,
"eval_runtime": 6.7726,
"eval_samples_per_second": 148.983,
"eval_steps_per_second": 18.752,
"step": 5300
},
{
"epoch": 12.71,
"eval_accuracy_score": 0.9630076908409076,
"eval_f1": 0.6628643852978454,
"eval_loss": 0.22853700816631317,
"eval_precision": 0.7048517520215634,
"eval_recall": 0.6255980861244019,
"eval_runtime": 6.7535,
"eval_samples_per_second": 149.405,
"eval_steps_per_second": 18.805,
"step": 5400
},
{
"epoch": 12.94,
"learning_rate": 2.843137254901961e-06,
"loss": 0.0321,
"step": 5500
},
{
"epoch": 12.94,
"eval_accuracy_score": 0.9631348121782242,
"eval_f1": 0.664548919949174,
"eval_loss": 0.22333784401416779,
"eval_precision": 0.7086720867208672,
"eval_recall": 0.6255980861244019,
"eval_runtime": 6.7401,
"eval_samples_per_second": 149.702,
"eval_steps_per_second": 18.843,
"step": 5500
},
{
"epoch": 13.18,
"eval_accuracy_score": 0.9637068581961482,
"eval_f1": 0.6705054382597568,
"eval_loss": 0.222885861992836,
"eval_precision": 0.7207702888583218,
"eval_recall": 0.6267942583732058,
"eval_runtime": 6.7486,
"eval_samples_per_second": 149.513,
"eval_steps_per_second": 18.819,
"step": 5600
},
{
"epoch": 13.41,
"eval_accuracy_score": 0.9633254941841988,
"eval_f1": 0.6619718309859155,
"eval_loss": 0.2285127490758896,
"eval_precision": 0.7121212121212122,
"eval_recall": 0.618421052631579,
"eval_runtime": 6.8129,
"eval_samples_per_second": 148.102,
"eval_steps_per_second": 18.641,
"step": 5700
},
{
"epoch": 13.65,
"eval_accuracy_score": 0.9644695862200471,
"eval_f1": 0.6734177215189874,
"eval_loss": 0.22083869576454163,
"eval_precision": 0.7150537634408602,
"eval_recall": 0.6363636363636364,
"eval_runtime": 6.7283,
"eval_samples_per_second": 149.964,
"eval_steps_per_second": 18.876,
"step": 5800
},
{
"epoch": 13.88,
"eval_accuracy_score": 0.9635161761901735,
"eval_f1": 0.6684005201560469,
"eval_loss": 0.2334665209054947,
"eval_precision": 0.7321937321937322,
"eval_recall": 0.6148325358851675,
"eval_runtime": 6.7215,
"eval_samples_per_second": 150.116,
"eval_steps_per_second": 18.895,
"step": 5900
},
{
"epoch": 14.12,
"learning_rate": 2.647058823529412e-06,
"loss": 0.0291,
"step": 6000
},
{
"epoch": 14.12,
"eval_accuracy_score": 0.9641517828767558,
"eval_f1": 0.663265306122449,
"eval_loss": 0.2206926792860031,
"eval_precision": 0.7103825136612022,
"eval_recall": 0.6220095693779905,
"eval_runtime": 6.7314,
"eval_samples_per_second": 149.894,
"eval_steps_per_second": 18.867,
"step": 6000
},
{
"epoch": 14.35,
"eval_accuracy_score": 0.9650416322379711,
"eval_f1": 0.6737633061991233,
"eval_loss": 0.21664156019687653,
"eval_precision": 0.7069645203679369,
"eval_recall": 0.6435406698564593,
"eval_runtime": 6.8128,
"eval_samples_per_second": 148.104,
"eval_steps_per_second": 18.641,
"step": 6100
},
{
"epoch": 14.59,
"eval_accuracy_score": 0.9646602682260217,
"eval_f1": 0.6654275092936804,
"eval_loss": 0.20899562537670135,
"eval_precision": 0.6902313624678663,
"eval_recall": 0.6423444976076556,
"eval_runtime": 6.7812,
"eval_samples_per_second": 148.794,
"eval_steps_per_second": 18.728,
"step": 6200
},
{
"epoch": 14.82,
"eval_accuracy_score": 0.9647873895633382,
"eval_f1": 0.6766917293233083,
"eval_loss": 0.21245865523815155,
"eval_precision": 0.7105263157894737,
"eval_recall": 0.645933014354067,
"eval_runtime": 6.7236,
"eval_samples_per_second": 150.069,
"eval_steps_per_second": 18.889,
"step": 6300
},
{
"epoch": 15.06,
"eval_accuracy_score": 0.9637704188648064,
"eval_f1": 0.6666666666666667,
"eval_loss": 0.22959989309310913,
"eval_precision": 0.713506139154161,
"eval_recall": 0.6255980861244019,
"eval_runtime": 6.7691,
"eval_samples_per_second": 149.061,
"eval_steps_per_second": 18.762,
"step": 6400
},
{
"epoch": 15.29,
"learning_rate": 2.450980392156863e-06,
"loss": 0.0251,
"step": 6500
},
{
"epoch": 15.29,
"eval_accuracy_score": 0.9645331468887053,
"eval_f1": 0.6733542319749217,
"eval_loss": 0.22114437818527222,
"eval_precision": 0.7075098814229249,
"eval_recall": 0.6423444976076556,
"eval_runtime": 6.8495,
"eval_samples_per_second": 147.311,
"eval_steps_per_second": 18.542,
"step": 6500
},
{
"epoch": 15.53,
"eval_accuracy_score": 0.9640882222080976,
"eval_f1": 0.6713197969543148,
"eval_loss": 0.22384566068649292,
"eval_precision": 0.7148648648648649,
"eval_recall": 0.6327751196172249,
"eval_runtime": 6.7143,
"eval_samples_per_second": 150.277,
"eval_steps_per_second": 18.915,
"step": 6600
},
{
"epoch": 15.76,
"eval_accuracy_score": 0.9637068581961482,
"eval_f1": 0.6641172721478649,
"eval_loss": 0.23098178207874298,
"eval_precision": 0.7107776261937244,
"eval_recall": 0.6232057416267942,
"eval_runtime": 6.7457,
"eval_samples_per_second": 149.576,
"eval_steps_per_second": 18.827,
"step": 6700
},
{
"epoch": 16.0,
"eval_accuracy_score": 0.9650416322379711,
"eval_f1": 0.6762860727728984,
"eval_loss": 0.21991874277591705,
"eval_precision": 0.7110817941952506,
"eval_recall": 0.6447368421052632,
"eval_runtime": 6.7686,
"eval_samples_per_second": 149.07,
"eval_steps_per_second": 18.763,
"step": 6800
},
{
"epoch": 16.24,
"eval_accuracy_score": 0.96364329752749,
"eval_f1": 0.6636771300448431,
"eval_loss": 0.23369288444519043,
"eval_precision": 0.7144827586206897,
"eval_recall": 0.6196172248803827,
"eval_runtime": 6.8164,
"eval_samples_per_second": 148.025,
"eval_steps_per_second": 18.631,
"step": 6900
},
{
"epoch": 16.47,
"learning_rate": 2.254901960784314e-06,
"loss": 0.0227,
"step": 7000
},
{
"epoch": 16.47,
"eval_accuracy_score": 0.9646602682260217,
"eval_f1": 0.6712243629583592,
"eval_loss": 0.22149834036827087,
"eval_precision": 0.6985769728331177,
"eval_recall": 0.645933014354067,
"eval_runtime": 6.7644,
"eval_samples_per_second": 149.164,
"eval_steps_per_second": 18.775,
"step": 7000
},
{
"epoch": 16.71,
"eval_accuracy_score": 0.9645331468887053,
"eval_f1": 0.6687422166874221,
"eval_loss": 0.2212422788143158,
"eval_precision": 0.6974025974025974,
"eval_recall": 0.6423444976076556,
"eval_runtime": 6.7009,
"eval_samples_per_second": 150.576,
"eval_steps_per_second": 18.953,
"step": 7100
},
{
"epoch": 16.94,
"eval_accuracy_score": 0.9635161761901735,
"eval_f1": 0.6653821451509313,
"eval_loss": 0.24178394675254822,
"eval_precision": 0.7184466019417476,
"eval_recall": 0.6196172248803827,
"eval_runtime": 6.7952,
"eval_samples_per_second": 148.486,
"eval_steps_per_second": 18.69,
"step": 7200
},
{
"epoch": 17.18,
"eval_accuracy_score": 0.9633890548528571,
"eval_f1": 0.6675307841866494,
"eval_loss": 0.24713467061519623,
"eval_precision": 0.7284299858557284,
"eval_recall": 0.6160287081339713,
"eval_runtime": 6.807,
"eval_samples_per_second": 148.229,
"eval_steps_per_second": 18.657,
"step": 7300
},
{
"epoch": 17.41,
"eval_accuracy_score": 0.9643424648827306,
"eval_f1": 0.670031545741325,
"eval_loss": 0.22886203229427338,
"eval_precision": 0.7089452603471295,
"eval_recall": 0.6351674641148325,
"eval_runtime": 6.7775,
"eval_samples_per_second": 148.874,
"eval_steps_per_second": 18.738,
"step": 7400
},
{
"epoch": 17.65,
"learning_rate": 2.058823529411765e-06,
"loss": 0.0197,
"step": 7500
},
{
"epoch": 17.65,
"eval_accuracy_score": 0.9646602682260217,
"eval_f1": 0.6753731343283582,
"eval_loss": 0.22524945437908173,
"eval_precision": 0.7033678756476683,
"eval_recall": 0.6495215311004785,
"eval_runtime": 6.7338,
"eval_samples_per_second": 149.84,
"eval_steps_per_second": 18.86,
"step": 7500
},
{
"epoch": 17.88,
"eval_accuracy_score": 0.9633890548528571,
"eval_f1": 0.664528543938422,
"eval_loss": 0.24701638519763947,
"eval_precision": 0.7164591977869986,
"eval_recall": 0.6196172248803827,
"eval_runtime": 6.7838,
"eval_samples_per_second": 148.736,
"eval_steps_per_second": 18.721,
"step": 7600
},
{
"epoch": 18.12,
"eval_accuracy_score": 0.9654865569185788,
"eval_f1": 0.6825,
"eval_loss": 0.22353364527225494,
"eval_precision": 0.7146596858638743,
"eval_recall": 0.65311004784689,
"eval_runtime": 6.7595,
"eval_samples_per_second": 149.272,
"eval_steps_per_second": 18.788,
"step": 7700
},
{
"epoch": 18.35,
"eval_accuracy_score": 0.9638975402021229,
"eval_f1": 0.6590765338393421,
"eval_loss": 0.23786970973014832,
"eval_precision": 0.6993288590604027,
"eval_recall": 0.6232057416267942,
"eval_runtime": 6.8141,
"eval_samples_per_second": 148.075,
"eval_steps_per_second": 18.638,
"step": 7800
},
{
"epoch": 18.59,
"eval_accuracy_score": 0.9644060255513888,
"eval_f1": 0.6666666666666667,
"eval_loss": 0.23264023661613464,
"eval_precision": 0.7014531043593131,
"eval_recall": 0.6351674641148325,
"eval_runtime": 6.8013,
"eval_samples_per_second": 148.353,
"eval_steps_per_second": 18.673,
"step": 7900
},
{
"epoch": 18.82,
"learning_rate": 1.8627450980392158e-06,
"loss": 0.0183,
"step": 8000
},
{
"epoch": 18.82,
"eval_accuracy_score": 0.9649780715693129,
"eval_f1": 0.6724244293645897,
"eval_loss": 0.22304123640060425,
"eval_precision": 0.6942675159235668,
"eval_recall": 0.6519138755980861,
"eval_runtime": 6.7426,
"eval_samples_per_second": 149.646,
"eval_steps_per_second": 18.835,
"step": 8000
},
{
"epoch": 19.06,
"eval_accuracy_score": 0.9643424648827306,
"eval_f1": 0.6683480453972257,
"eval_loss": 0.2350449413061142,
"eval_precision": 0.7066666666666667,
"eval_recall": 0.6339712918660287,
"eval_runtime": 6.8057,
"eval_samples_per_second": 148.259,
"eval_steps_per_second": 18.661,
"step": 8100
},
{
"epoch": 19.29,
"eval_accuracy_score": 0.9652323142439458,
"eval_f1": 0.6790582403965303,
"eval_loss": 0.226871520280838,
"eval_precision": 0.7043701799485861,
"eval_recall": 0.6555023923444976,
"eval_runtime": 6.7483,
"eval_samples_per_second": 149.519,
"eval_steps_per_second": 18.819,
"step": 8200
},
{
"epoch": 19.53,
"eval_accuracy_score": 0.96472382889468,
"eval_f1": 0.6683354192740926,
"eval_loss": 0.23186342418193817,
"eval_precision": 0.7007874015748031,
"eval_recall": 0.638755980861244,
"eval_runtime": 6.7632,
"eval_samples_per_second": 149.19,
"eval_steps_per_second": 18.778,
"step": 8300
},
{
"epoch": 19.76,
"eval_accuracy_score": 0.9645967075573635,
"eval_f1": 0.6712933753943218,
"eval_loss": 0.2337142825126648,
"eval_precision": 0.7102803738317757,
"eval_recall": 0.6363636363636364,
"eval_runtime": 6.7451,
"eval_samples_per_second": 149.591,
"eval_steps_per_second": 18.829,
"step": 8400
},
{
"epoch": 20.0,
"learning_rate": 1.6666666666666667e-06,
"loss": 0.0162,
"step": 8500
},
{
"epoch": 20.0,
"eval_accuracy_score": 0.9647873895633382,
"eval_f1": 0.6751430387794024,
"eval_loss": 0.2389359027147293,
"eval_precision": 0.7204884667571235,
"eval_recall": 0.6351674641148325,
"eval_runtime": 6.7503,
"eval_samples_per_second": 149.474,
"eval_steps_per_second": 18.814,
"step": 8500
},
{
"epoch": 20.24,
"eval_accuracy_score": 0.9638975402021229,
"eval_f1": 0.6709677419354839,
"eval_loss": 0.25359493494033813,
"eval_precision": 0.7282913165266106,
"eval_recall": 0.6220095693779905,
"eval_runtime": 6.7422,
"eval_samples_per_second": 149.655,
"eval_steps_per_second": 18.837,
"step": 8600
},
{
"epoch": 20.47,
"eval_accuracy_score": 0.964215343545414,
"eval_f1": 0.6709346991037132,
"eval_loss": 0.24693100154399872,
"eval_precision": 0.721763085399449,
"eval_recall": 0.6267942583732058,
"eval_runtime": 6.7296,
"eval_samples_per_second": 149.934,
"eval_steps_per_second": 18.872,
"step": 8700
},
{
"epoch": 20.71,
"eval_accuracy_score": 0.9637704188648064,
"eval_f1": 0.6687938736439056,
"eval_loss": 0.2458009421825409,
"eval_precision": 0.7168262653898769,
"eval_recall": 0.6267942583732058,
"eval_runtime": 6.7627,
"eval_samples_per_second": 149.201,
"eval_steps_per_second": 18.78,
"step": 8800
},
{
"epoch": 20.94,
"eval_accuracy_score": 0.9648509502319964,
"eval_f1": 0.676304211187932,
"eval_loss": 0.23430827260017395,
"eval_precision": 0.7125827814569536,
"eval_recall": 0.6435406698564593,
"eval_runtime": 6.7289,
"eval_samples_per_second": 149.951,
"eval_steps_per_second": 18.874,
"step": 8900
},
{
"epoch": 21.18,
"learning_rate": 1.4705882352941177e-06,
"loss": 0.0149,
"step": 9000
},
{
"epoch": 21.18,
"eval_accuracy_score": 0.9651687535752876,
"eval_f1": 0.6788413098236775,
"eval_loss": 0.23078210651874542,
"eval_precision": 0.7167553191489362,
"eval_recall": 0.6447368421052632,
"eval_runtime": 6.7181,
"eval_samples_per_second": 150.19,
"eval_steps_per_second": 18.904,
"step": 9000
}
],
"max_steps": 12750,
"num_train_epochs": 30,
"total_flos": 4695577703714304.0,
"trial_name": null,
"trial_params": null
}