structroberta_sx_final / finetune /qqp /trainer_state.json
Omar
update
88cccb3
{
"best_metric": 0.8048906048906048,
"best_model_checkpoint": "final_models/glue_models/structroberta_s2_50ep//finetune/qqp/checkpoint-10600",
"epoch": 3.3114323258869907,
"global_step": 12600,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.05,
"eval_accuracy": 0.6895384788513184,
"eval_f1": 0.6217489805165383,
"eval_loss": 0.5790485143661499,
"eval_mcc": 0.3590350743412485,
"eval_runtime": 56.9694,
"eval_samples_per_second": 471.99,
"eval_steps_per_second": 59.014,
"step": 200
},
{
"epoch": 0.11,
"eval_accuracy": 0.7031499743461609,
"eval_f1": 0.659238387978142,
"eval_loss": 0.5603744983673096,
"eval_mcc": 0.397609114853472,
"eval_runtime": 57.0635,
"eval_samples_per_second": 471.212,
"eval_steps_per_second": 58.917,
"step": 400
},
{
"epoch": 0.13,
"learning_rate": 4.934296977660973e-05,
"loss": 0.5879,
"step": 500
},
{
"epoch": 0.16,
"eval_accuracy": 0.7312655448913574,
"eval_f1": 0.6582482027998486,
"eval_loss": 0.5294312834739685,
"eval_mcc": 0.4414063833899318,
"eval_runtime": 57.2359,
"eval_samples_per_second": 469.792,
"eval_steps_per_second": 58.739,
"step": 600
},
{
"epoch": 0.21,
"eval_accuracy": 0.7384432554244995,
"eval_f1": 0.6671399498319845,
"eval_loss": 0.5185254216194153,
"eval_mcc": 0.4564598592164868,
"eval_runtime": 57.1234,
"eval_samples_per_second": 470.718,
"eval_steps_per_second": 58.855,
"step": 800
},
{
"epoch": 0.26,
"learning_rate": 4.868593955321945e-05,
"loss": 0.5309,
"step": 1000
},
{
"epoch": 0.26,
"eval_accuracy": 0.7485216856002808,
"eval_f1": 0.6855761182925696,
"eval_loss": 0.5134466886520386,
"eval_mcc": 0.4786827802251824,
"eval_runtime": 57.2501,
"eval_samples_per_second": 469.676,
"eval_steps_per_second": 58.725,
"step": 1000
},
{
"epoch": 0.32,
"eval_accuracy": 0.752352237701416,
"eval_f1": 0.7012427654897034,
"eval_loss": 0.4966827630996704,
"eval_mcc": 0.48998691711625764,
"eval_runtime": 57.1252,
"eval_samples_per_second": 470.703,
"eval_steps_per_second": 58.853,
"step": 1200
},
{
"epoch": 0.37,
"eval_accuracy": 0.7582654356956482,
"eval_f1": 0.6868677136525676,
"eval_loss": 0.4926142990589142,
"eval_mcc": 0.4977363897273405,
"eval_runtime": 56.8921,
"eval_samples_per_second": 472.631,
"eval_steps_per_second": 59.094,
"step": 1400
},
{
"epoch": 0.39,
"learning_rate": 4.8028909329829176e-05,
"loss": 0.5081,
"step": 1500
},
{
"epoch": 0.42,
"eval_accuracy": 0.7646249532699585,
"eval_f1": 0.7162774017124669,
"eval_loss": 0.4827042818069458,
"eval_mcc": 0.5153659006018214,
"eval_runtime": 57.181,
"eval_samples_per_second": 470.244,
"eval_steps_per_second": 58.796,
"step": 1600
},
{
"epoch": 0.47,
"eval_accuracy": 0.7632861137390137,
"eval_f1": 0.6841347823929335,
"eval_loss": 0.48687198758125305,
"eval_mcc": 0.5088622782048654,
"eval_runtime": 57.1217,
"eval_samples_per_second": 470.732,
"eval_steps_per_second": 58.857,
"step": 1800
},
{
"epoch": 0.53,
"learning_rate": 4.73718791064389e-05,
"loss": 0.4924,
"step": 2000
},
{
"epoch": 0.53,
"eval_accuracy": 0.7514225244522095,
"eval_f1": 0.6440136344269279,
"eval_loss": 0.49411216378211975,
"eval_mcc": 0.4884785053093981,
"eval_runtime": 57.12,
"eval_samples_per_second": 470.746,
"eval_steps_per_second": 58.859,
"step": 2000
},
{
"epoch": 0.58,
"eval_accuracy": 0.7751868963241577,
"eval_f1": 0.7179056418871623,
"eval_loss": 0.4720868468284607,
"eval_mcc": 0.5341846799936681,
"eval_runtime": 57.0874,
"eval_samples_per_second": 471.015,
"eval_steps_per_second": 58.892,
"step": 2200
},
{
"epoch": 0.63,
"eval_accuracy": 0.7680836319923401,
"eval_f1": 0.689070602313522,
"eval_loss": 0.48496729135513306,
"eval_mcc": 0.5194278885822263,
"eval_runtime": 56.9884,
"eval_samples_per_second": 471.833,
"eval_steps_per_second": 58.994,
"step": 2400
},
{
"epoch": 0.66,
"learning_rate": 4.6714848883048626e-05,
"loss": 0.475,
"step": 2500
},
{
"epoch": 0.68,
"eval_accuracy": 0.7773066759109497,
"eval_f1": 0.7529091359247337,
"eval_loss": 0.4599149525165558,
"eval_mcc": 0.5560435147097561,
"eval_runtime": 57.3093,
"eval_samples_per_second": 469.191,
"eval_steps_per_second": 58.664,
"step": 2600
},
{
"epoch": 0.74,
"eval_accuracy": 0.7876827120780945,
"eval_f1": 0.7471320370288347,
"eval_loss": 0.45423364639282227,
"eval_mcc": 0.5641587819264187,
"eval_runtime": 57.0781,
"eval_samples_per_second": 471.091,
"eval_steps_per_second": 58.902,
"step": 2800
},
{
"epoch": 0.79,
"learning_rate": 4.605781865965835e-05,
"loss": 0.4584,
"step": 3000
},
{
"epoch": 0.79,
"eval_accuracy": 0.7906206846237183,
"eval_f1": 0.7403136531365313,
"eval_loss": 0.44478940963745117,
"eval_mcc": 0.5668900307355025,
"eval_runtime": 57.1687,
"eval_samples_per_second": 470.345,
"eval_steps_per_second": 58.808,
"step": 3000
},
{
"epoch": 0.84,
"eval_accuracy": 0.789616584777832,
"eval_f1": 0.7573248680880271,
"eval_loss": 0.4422828257083893,
"eval_mcc": 0.5727700186824436,
"eval_runtime": 57.2015,
"eval_samples_per_second": 470.075,
"eval_steps_per_second": 58.775,
"step": 3200
},
{
"epoch": 0.89,
"eval_accuracy": 0.7930380702018738,
"eval_f1": 0.7571035747021082,
"eval_loss": 0.4391527771949768,
"eval_mcc": 0.5770275603454963,
"eval_runtime": 57.1341,
"eval_samples_per_second": 470.63,
"eval_steps_per_second": 58.844,
"step": 3400
},
{
"epoch": 0.92,
"learning_rate": 4.540078843626807e-05,
"loss": 0.4479,
"step": 3500
},
{
"epoch": 0.95,
"eval_accuracy": 0.7989140748977661,
"eval_f1": 0.7556599936734602,
"eval_loss": 0.43771788477897644,
"eval_mcc": 0.585340244040819,
"eval_runtime": 57.2391,
"eval_samples_per_second": 469.766,
"eval_steps_per_second": 58.736,
"step": 3600
},
{
"epoch": 1.0,
"eval_accuracy": 0.7988768815994263,
"eval_f1": 0.7547614728822782,
"eval_loss": 0.42634981870651245,
"eval_mcc": 0.5850092755942151,
"eval_runtime": 57.0922,
"eval_samples_per_second": 470.975,
"eval_steps_per_second": 58.887,
"step": 3800
},
{
"epoch": 1.05,
"learning_rate": 4.474375821287779e-05,
"loss": 0.4171,
"step": 4000
},
{
"epoch": 1.05,
"eval_accuracy": 0.8009223341941833,
"eval_f1": 0.7616545705507815,
"eval_loss": 0.4461493492126465,
"eval_mcc": 0.5907844940997748,
"eval_runtime": 57.2068,
"eval_samples_per_second": 470.031,
"eval_steps_per_second": 58.769,
"step": 4000
},
{
"epoch": 1.1,
"eval_accuracy": 0.7885008454322815,
"eval_f1": 0.7156642167891606,
"eval_loss": 0.46828022599220276,
"eval_mcc": 0.5637818475370291,
"eval_runtime": 57.3281,
"eval_samples_per_second": 469.037,
"eval_steps_per_second": 58.645,
"step": 4200
},
{
"epoch": 1.16,
"eval_accuracy": 0.8006619811058044,
"eval_f1": 0.768546506606788,
"eval_loss": 0.45249348878860474,
"eval_mcc": 0.5941976311906684,
"eval_runtime": 57.3016,
"eval_samples_per_second": 469.254,
"eval_steps_per_second": 58.672,
"step": 4400
},
{
"epoch": 1.18,
"learning_rate": 4.408672798948752e-05,
"loss": 0.3771,
"step": 4500
},
{
"epoch": 1.21,
"eval_accuracy": 0.8068727254867554,
"eval_f1": 0.7759223300970873,
"eval_loss": 0.4169043004512787,
"eval_mcc": 0.606982696864278,
"eval_runtime": 57.1406,
"eval_samples_per_second": 470.576,
"eval_steps_per_second": 58.837,
"step": 4600
},
{
"epoch": 1.26,
"eval_accuracy": 0.7954925894737244,
"eval_f1": 0.7291533270945181,
"eval_loss": 0.4451744556427002,
"eval_mcc": 0.5777518921950682,
"eval_runtime": 57.2245,
"eval_samples_per_second": 469.886,
"eval_steps_per_second": 58.751,
"step": 4800
},
{
"epoch": 1.31,
"learning_rate": 4.342969776609724e-05,
"loss": 0.3773,
"step": 5000
},
{
"epoch": 1.31,
"eval_accuracy": 0.8078768253326416,
"eval_f1": 0.7645396536007292,
"eval_loss": 0.42197084426879883,
"eval_mcc": 0.603319424026119,
"eval_runtime": 57.0783,
"eval_samples_per_second": 471.09,
"eval_steps_per_second": 58.902,
"step": 5000
},
{
"epoch": 1.37,
"eval_accuracy": 0.8099594712257385,
"eval_f1": 0.7777294475859068,
"eval_loss": 0.4131057560443878,
"eval_mcc": 0.6121005762340846,
"eval_runtime": 57.038,
"eval_samples_per_second": 471.423,
"eval_steps_per_second": 58.943,
"step": 5200
},
{
"epoch": 1.42,
"eval_accuracy": 0.8107776641845703,
"eval_f1": 0.7651186409380483,
"eval_loss": 0.4410916864871979,
"eval_mcc": 0.6088060459447059,
"eval_runtime": 57.1661,
"eval_samples_per_second": 470.366,
"eval_steps_per_second": 58.811,
"step": 5400
},
{
"epoch": 1.45,
"learning_rate": 4.2772667542706967e-05,
"loss": 0.3716,
"step": 5500
},
{
"epoch": 1.47,
"eval_accuracy": 0.8099594712257385,
"eval_f1": 0.7631188577786018,
"eval_loss": 0.43267035484313965,
"eval_mcc": 0.6069920811878712,
"eval_runtime": 56.9593,
"eval_samples_per_second": 472.074,
"eval_steps_per_second": 59.025,
"step": 5600
},
{
"epoch": 1.52,
"eval_accuracy": 0.8123767971992493,
"eval_f1": 0.7751481927173864,
"eval_loss": 0.4137977659702301,
"eval_mcc": 0.6142423355676122,
"eval_runtime": 57.2611,
"eval_samples_per_second": 469.586,
"eval_steps_per_second": 58.713,
"step": 5800
},
{
"epoch": 1.58,
"learning_rate": 4.211563731931669e-05,
"loss": 0.3652,
"step": 6000
},
{
"epoch": 1.58,
"eval_accuracy": 0.811670184135437,
"eval_f1": 0.762119503945885,
"eval_loss": 0.4430968165397644,
"eval_mcc": 0.610427625756327,
"eval_runtime": 57.15,
"eval_samples_per_second": 470.499,
"eval_steps_per_second": 58.828,
"step": 6000
},
{
"epoch": 1.63,
"eval_accuracy": 0.811149537563324,
"eval_f1": 0.773667320377964,
"eval_loss": 0.42289894819259644,
"eval_mcc": 0.6117143787328192,
"eval_runtime": 57.3569,
"eval_samples_per_second": 468.801,
"eval_steps_per_second": 58.615,
"step": 6200
},
{
"epoch": 1.68,
"eval_accuracy": 0.813975989818573,
"eval_f1": 0.772325898953118,
"eval_loss": 0.4201619327068329,
"eval_mcc": 0.6160239371091856,
"eval_runtime": 57.1429,
"eval_samples_per_second": 470.557,
"eval_steps_per_second": 58.835,
"step": 6400
},
{
"epoch": 1.71,
"learning_rate": 4.1458607095926416e-05,
"loss": 0.3651,
"step": 6500
},
{
"epoch": 1.73,
"eval_accuracy": 0.8141990900039673,
"eval_f1": 0.787512759441987,
"eval_loss": 0.42058825492858887,
"eval_mcc": 0.6243481982422204,
"eval_runtime": 57.0624,
"eval_samples_per_second": 471.221,
"eval_steps_per_second": 58.918,
"step": 6600
},
{
"epoch": 1.79,
"eval_accuracy": 0.8160214424133301,
"eval_f1": 0.7705366668212812,
"eval_loss": 0.4051779508590698,
"eval_mcc": 0.6196198122308723,
"eval_runtime": 57.3622,
"eval_samples_per_second": 468.758,
"eval_steps_per_second": 58.61,
"step": 6800
},
{
"epoch": 1.84,
"learning_rate": 4.080157687253614e-05,
"loss": 0.3559,
"step": 7000
},
{
"epoch": 1.84,
"eval_accuracy": 0.8187363147735596,
"eval_f1": 0.778816482120167,
"eval_loss": 0.4070768654346466,
"eval_mcc": 0.6260354328376837,
"eval_runtime": 57.2066,
"eval_samples_per_second": 470.033,
"eval_steps_per_second": 58.769,
"step": 7000
},
{
"epoch": 1.89,
"eval_accuracy": 0.8202982544898987,
"eval_f1": 0.7892900750043608,
"eval_loss": 0.40111517906188965,
"eval_mcc": 0.6328935271313376,
"eval_runtime": 57.2191,
"eval_samples_per_second": 469.93,
"eval_steps_per_second": 58.757,
"step": 7200
},
{
"epoch": 1.94,
"eval_accuracy": 0.8220833539962769,
"eval_f1": 0.7814926463871381,
"eval_loss": 0.39430883526802063,
"eval_mcc": 0.6326604221713003,
"eval_runtime": 57.1848,
"eval_samples_per_second": 470.213,
"eval_steps_per_second": 58.792,
"step": 7400
},
{
"epoch": 1.97,
"learning_rate": 4.0144546649145865e-05,
"loss": 0.3619,
"step": 7500
},
{
"epoch": 2.0,
"eval_accuracy": 0.8156495094299316,
"eval_f1": 0.7624478842191019,
"eval_loss": 0.41225436329841614,
"eval_mcc": 0.619232101302901,
"eval_runtime": 57.0196,
"eval_samples_per_second": 471.575,
"eval_steps_per_second": 58.962,
"step": 7600
},
{
"epoch": 2.05,
"eval_accuracy": 0.8209304809570312,
"eval_f1": 0.7901320664254892,
"eval_loss": 0.4239133894443512,
"eval_mcc": 0.6342483559433012,
"eval_runtime": 57.2383,
"eval_samples_per_second": 469.773,
"eval_steps_per_second": 58.737,
"step": 7800
},
{
"epoch": 2.1,
"learning_rate": 3.948751642575559e-05,
"loss": 0.2886,
"step": 8000
},
{
"epoch": 2.1,
"eval_accuracy": 0.821525514125824,
"eval_f1": 0.790097537506014,
"eval_loss": 0.4144342243671417,
"eval_mcc": 0.6350233886842492,
"eval_runtime": 57.1057,
"eval_samples_per_second": 470.864,
"eval_steps_per_second": 58.873,
"step": 8000
},
{
"epoch": 2.16,
"eval_accuracy": 0.8241288065910339,
"eval_f1": 0.7847617313731737,
"eval_loss": 0.40638136863708496,
"eval_mcc": 0.6370477000221836,
"eval_runtime": 57.2228,
"eval_samples_per_second": 469.9,
"eval_steps_per_second": 58.753,
"step": 8200
},
{
"epoch": 2.21,
"eval_accuracy": 0.8260998725891113,
"eval_f1": 0.7945337903154934,
"eval_loss": 0.42089590430259705,
"eval_mcc": 0.6438486028810766,
"eval_runtime": 57.1363,
"eval_samples_per_second": 470.612,
"eval_steps_per_second": 58.842,
"step": 8400
},
{
"epoch": 2.23,
"learning_rate": 3.8830486202365314e-05,
"loss": 0.2766,
"step": 8500
},
{
"epoch": 2.26,
"eval_accuracy": 0.8188478350639343,
"eval_f1": 0.7956709593523219,
"eval_loss": 0.43640926480293274,
"eval_mcc": 0.63650625614698,
"eval_runtime": 56.9978,
"eval_samples_per_second": 471.755,
"eval_steps_per_second": 58.985,
"step": 8600
},
{
"epoch": 2.31,
"eval_accuracy": 0.8274387121200562,
"eval_f1": 0.7945811935540995,
"eval_loss": 0.405335932970047,
"eval_mcc": 0.6458161739050261,
"eval_runtime": 57.0558,
"eval_samples_per_second": 471.275,
"eval_steps_per_second": 58.925,
"step": 8800
},
{
"epoch": 2.37,
"learning_rate": 3.817345597897504e-05,
"loss": 0.2797,
"step": 9000
},
{
"epoch": 2.37,
"eval_accuracy": 0.8264718055725098,
"eval_f1": 0.7906684611933603,
"eval_loss": 0.4281841516494751,
"eval_mcc": 0.6427212326252381,
"eval_runtime": 57.0173,
"eval_samples_per_second": 471.593,
"eval_steps_per_second": 58.965,
"step": 9000
},
{
"epoch": 2.42,
"eval_accuracy": 0.8209676742553711,
"eval_f1": 0.775926270713089,
"eval_loss": 0.4169626533985138,
"eval_mcc": 0.6299012875382383,
"eval_runtime": 57.2133,
"eval_samples_per_second": 469.978,
"eval_steps_per_second": 58.763,
"step": 9200
},
{
"epoch": 2.47,
"eval_accuracy": 0.8285172581672668,
"eval_f1": 0.7999826486791307,
"eval_loss": 0.40043607354164124,
"eval_mcc": 0.6503868538586679,
"eval_runtime": 57.1097,
"eval_samples_per_second": 470.831,
"eval_steps_per_second": 58.869,
"step": 9400
},
{
"epoch": 2.5,
"learning_rate": 3.7516425755584764e-05,
"loss": 0.2866,
"step": 9500
},
{
"epoch": 2.52,
"eval_accuracy": 0.8278850317001343,
"eval_f1": 0.7914375844975214,
"eval_loss": 0.4065154492855072,
"eval_mcc": 0.6453385200418826,
"eval_runtime": 57.1041,
"eval_samples_per_second": 470.877,
"eval_steps_per_second": 58.875,
"step": 9600
},
{
"epoch": 2.58,
"eval_accuracy": 0.8289635181427002,
"eval_f1": 0.791702522759183,
"eval_loss": 0.39962634444236755,
"eval_mcc": 0.6472846692602012,
"eval_runtime": 57.1663,
"eval_samples_per_second": 470.364,
"eval_steps_per_second": 58.811,
"step": 9800
},
{
"epoch": 2.63,
"learning_rate": 3.685939553219449e-05,
"loss": 0.2838,
"step": 10000
},
{
"epoch": 2.63,
"eval_accuracy": 0.8313064575195312,
"eval_f1": 0.7954914337240757,
"eval_loss": 0.41669055819511414,
"eval_mcc": 0.6523724926921574,
"eval_runtime": 57.0331,
"eval_samples_per_second": 471.463,
"eval_steps_per_second": 58.948,
"step": 10000
},
{
"epoch": 2.68,
"eval_accuracy": 0.8318271636962891,
"eval_f1": 0.7962328767123288,
"eval_loss": 0.428965300321579,
"eval_mcc": 0.6534785544952427,
"eval_runtime": 57.2333,
"eval_samples_per_second": 469.814,
"eval_steps_per_second": 58.742,
"step": 10200
},
{
"epoch": 2.73,
"eval_accuracy": 0.8308601975440979,
"eval_f1": 0.7983863817714337,
"eval_loss": 0.4066237211227417,
"eval_mcc": 0.6527156031218291,
"eval_runtime": 57.2722,
"eval_samples_per_second": 469.495,
"eval_steps_per_second": 58.702,
"step": 10400
},
{
"epoch": 2.76,
"learning_rate": 3.6202365308804206e-05,
"loss": 0.2807,
"step": 10500
},
{
"epoch": 2.79,
"eval_accuracy": 0.8308601975440979,
"eval_f1": 0.8048906048906048,
"eval_loss": 0.3991719186306,
"eval_mcc": 0.6567986450071405,
"eval_runtime": 57.1392,
"eval_samples_per_second": 470.588,
"eval_steps_per_second": 58.839,
"step": 10600
},
{
"epoch": 2.84,
"eval_accuracy": 0.830302357673645,
"eval_f1": 0.7989779285431076,
"eval_loss": 0.41573354601860046,
"eval_mcc": 0.6521800844709269,
"eval_runtime": 57.1598,
"eval_samples_per_second": 470.418,
"eval_steps_per_second": 58.818,
"step": 10800
},
{
"epoch": 2.89,
"learning_rate": 3.554533508541393e-05,
"loss": 0.2854,
"step": 11000
},
{
"epoch": 2.89,
"eval_accuracy": 0.827996551990509,
"eval_f1": 0.7814375502102926,
"eval_loss": 0.4512079060077667,
"eval_mcc": 0.6447879159418899,
"eval_runtime": 57.062,
"eval_samples_per_second": 471.224,
"eval_steps_per_second": 58.918,
"step": 11000
},
{
"epoch": 2.94,
"eval_accuracy": 0.8250957727432251,
"eval_f1": 0.7756095233551219,
"eval_loss": 0.426098495721817,
"eval_mcc": 0.6390143935892307,
"eval_runtime": 57.1498,
"eval_samples_per_second": 470.5,
"eval_steps_per_second": 58.828,
"step": 11200
},
{
"epoch": 3.0,
"eval_accuracy": 0.8325709104537964,
"eval_f1": 0.798478066248881,
"eval_loss": 0.4026671350002289,
"eval_mcc": 0.655443685564215,
"eval_runtime": 57.2209,
"eval_samples_per_second": 469.916,
"eval_steps_per_second": 58.755,
"step": 11400
},
{
"epoch": 3.02,
"learning_rate": 3.4888304862023655e-05,
"loss": 0.2705,
"step": 11500
},
{
"epoch": 3.05,
"eval_accuracy": 0.8301535844802856,
"eval_f1": 0.7850317721816898,
"eval_loss": 0.5190560817718506,
"eval_mcc": 0.6492185236914582,
"eval_runtime": 57.1491,
"eval_samples_per_second": 470.506,
"eval_steps_per_second": 58.829,
"step": 11600
},
{
"epoch": 3.1,
"eval_accuracy": 0.8269924521446228,
"eval_f1": 0.8008220585716732,
"eval_loss": 0.4870125353336334,
"eval_mcc": 0.6492371751318124,
"eval_runtime": 57.0005,
"eval_samples_per_second": 471.733,
"eval_steps_per_second": 58.982,
"step": 11800
},
{
"epoch": 3.15,
"learning_rate": 3.423127463863337e-05,
"loss": 0.2011,
"step": 12000
},
{
"epoch": 3.15,
"eval_accuracy": 0.830599844455719,
"eval_f1": 0.7876357872161873,
"eval_loss": 0.47779834270477295,
"eval_mcc": 0.6500219154781351,
"eval_runtime": 57.4254,
"eval_samples_per_second": 468.243,
"eval_steps_per_second": 58.546,
"step": 12000
},
{
"epoch": 3.21,
"eval_accuracy": 0.8338726162910461,
"eval_f1": 0.7996771155657204,
"eval_loss": 0.43881040811538696,
"eval_mcc": 0.6579979480473281,
"eval_runtime": 57.1566,
"eval_samples_per_second": 470.445,
"eval_steps_per_second": 58.821,
"step": 12200
},
{
"epoch": 3.26,
"eval_accuracy": 0.834430456161499,
"eval_f1": 0.795854732208364,
"eval_loss": 0.4549373984336853,
"eval_mcc": 0.658150442576307,
"eval_runtime": 57.0319,
"eval_samples_per_second": 471.473,
"eval_steps_per_second": 58.949,
"step": 12400
},
{
"epoch": 3.29,
"learning_rate": 3.35742444152431e-05,
"loss": 0.2077,
"step": 12500
},
{
"epoch": 3.31,
"eval_accuracy": 0.830599844455719,
"eval_f1": 0.79477359765713,
"eval_loss": 0.45785439014434814,
"eval_mcc": 0.6509541866022467,
"eval_runtime": 57.1908,
"eval_samples_per_second": 470.163,
"eval_steps_per_second": 58.786,
"step": 12600
},
{
"epoch": 3.31,
"step": 12600,
"total_flos": 7.386343728182784e+16,
"train_loss": 0.36493192960345555,
"train_runtime": 7909.6164,
"train_samples_per_second": 307.851,
"train_steps_per_second": 4.811
}
],
"max_steps": 38050,
"num_train_epochs": 10,
"total_flos": 7.386343728182784e+16,
"trial_name": null,
"trial_params": null
}