| { | |
| "best_global_step": 49000, | |
| "best_metric": 0.10218960046768188, | |
| "best_model_checkpoint": "myanmar_text_segmentation_model/checkpoint-49000", | |
| "epoch": 4.502434990351925, | |
| "eval_steps": 1000, | |
| "global_step": 49000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04594321418726454, | |
| "grad_norm": 0.577109158039093, | |
| "learning_rate": 1.9816594688964442e-05, | |
| "loss": 0.1144, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09188642837452908, | |
| "grad_norm": 0.8819814324378967, | |
| "learning_rate": 1.9632821832215383e-05, | |
| "loss": 0.1133, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09188642837452908, | |
| "eval_accuracy": 0.9564004775738034, | |
| "eval_f1": 0.7581789101913106, | |
| "eval_loss": 0.11448723077774048, | |
| "eval_precision": 0.7703745996379334, | |
| "eval_recall": 0.7463633395484631, | |
| "eval_runtime": 465.4607, | |
| "eval_samples_per_second": 61.528, | |
| "eval_steps_per_second": 1.231, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1378296425617936, | |
| "grad_norm": 0.5137984156608582, | |
| "learning_rate": 1.9449048975466325e-05, | |
| "loss": 0.1119, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18377285674905816, | |
| "grad_norm": 0.6753482222557068, | |
| "learning_rate": 1.9265276118717266e-05, | |
| "loss": 0.1115, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.18377285674905816, | |
| "eval_accuracy": 0.9565513036953667, | |
| "eval_f1": 0.7631813253052621, | |
| "eval_loss": 0.11509840190410614, | |
| "eval_precision": 0.7619568901058112, | |
| "eval_recall": 0.7644097020785157, | |
| "eval_runtime": 471.4797, | |
| "eval_samples_per_second": 60.743, | |
| "eval_steps_per_second": 1.215, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2297160709363227, | |
| "grad_norm": 0.46862444281578064, | |
| "learning_rate": 1.9081503261968207e-05, | |
| "loss": 0.1083, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2756592851235872, | |
| "grad_norm": 0.6514066457748413, | |
| "learning_rate": 1.889773040521915e-05, | |
| "loss": 0.1085, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.2756592851235872, | |
| "eval_accuracy": 0.955793370748351, | |
| "eval_f1": 0.7617613129080717, | |
| "eval_loss": 0.11627302318811417, | |
| "eval_precision": 0.7514122514122514, | |
| "eval_recall": 0.7723994268768838, | |
| "eval_runtime": 464.541, | |
| "eval_samples_per_second": 61.65, | |
| "eval_steps_per_second": 1.233, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3216024993108518, | |
| "grad_norm": 0.5837205052375793, | |
| "learning_rate": 1.8713957548470093e-05, | |
| "loss": 0.1068, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.3675457134981163, | |
| "grad_norm": 0.590044379234314, | |
| "learning_rate": 1.8530184691721034e-05, | |
| "loss": 0.1076, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.3675457134981163, | |
| "eval_accuracy": 0.9567566300121167, | |
| "eval_f1": 0.7569129075628852, | |
| "eval_loss": 0.11681018769741058, | |
| "eval_precision": 0.7790560842692136, | |
| "eval_recall": 0.7359936967250493, | |
| "eval_runtime": 464.1614, | |
| "eval_samples_per_second": 61.701, | |
| "eval_steps_per_second": 1.234, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.41348892768538087, | |
| "grad_norm": 0.43107616901397705, | |
| "learning_rate": 1.8346411834971975e-05, | |
| "loss": 0.1071, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.4594321418726454, | |
| "grad_norm": 0.6132990121841431, | |
| "learning_rate": 1.8162638978222916e-05, | |
| "loss": 0.1056, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.4594321418726454, | |
| "eval_accuracy": 0.9576032842071108, | |
| "eval_f1": 0.7661190672004459, | |
| "eval_loss": 0.11183962970972061, | |
| "eval_precision": 0.7719609680194174, | |
| "eval_recall": 0.7603649207638403, | |
| "eval_runtime": 473.1438, | |
| "eval_samples_per_second": 60.529, | |
| "eval_steps_per_second": 1.211, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5053753560599099, | |
| "grad_norm": 0.4892466962337494, | |
| "learning_rate": 1.7978866121473858e-05, | |
| "loss": 0.1154, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.5513185702471745, | |
| "grad_norm": 0.8282744884490967, | |
| "learning_rate": 1.7795093264724802e-05, | |
| "loss": 0.1141, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.5513185702471745, | |
| "eval_accuracy": 0.9580804777765948, | |
| "eval_f1": 0.7681685393564336, | |
| "eval_loss": 0.11200755834579468, | |
| "eval_precision": 0.7751412801894542, | |
| "eval_recall": 0.7613201259575663, | |
| "eval_runtime": 463.6002, | |
| "eval_samples_per_second": 61.775, | |
| "eval_steps_per_second": 1.236, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.597261784434439, | |
| "grad_norm": 0.44782352447509766, | |
| "learning_rate": 1.7611320407975743e-05, | |
| "loss": 0.1135, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.6432049986217035, | |
| "grad_norm": 0.5544679760932922, | |
| "learning_rate": 1.7427547551226684e-05, | |
| "loss": 0.1154, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.6432049986217035, | |
| "eval_accuracy": 0.9576146912247081, | |
| "eval_f1": 0.7686338805157804, | |
| "eval_loss": 0.11075501888990402, | |
| "eval_precision": 0.7656366294168357, | |
| "eval_recall": 0.7716546906241484, | |
| "eval_runtime": 472.6577, | |
| "eval_samples_per_second": 60.591, | |
| "eval_steps_per_second": 1.212, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.6891482128089681, | |
| "grad_norm": 0.4101652503013611, | |
| "learning_rate": 1.7243774694477626e-05, | |
| "loss": 0.1123, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.7350914269962326, | |
| "grad_norm": 0.5032928586006165, | |
| "learning_rate": 1.7060001837728567e-05, | |
| "loss": 0.1128, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.7350914269962326, | |
| "eval_accuracy": 0.9583707230021242, | |
| "eval_f1": 0.7702898098589065, | |
| "eval_loss": 0.10922538489103317, | |
| "eval_precision": 0.7744496660220761, | |
| "eval_recall": 0.7661744031991279, | |
| "eval_runtime": 465.4833, | |
| "eval_samples_per_second": 61.525, | |
| "eval_steps_per_second": 1.231, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.7810346411834972, | |
| "grad_norm": 0.5559138059616089, | |
| "learning_rate": 1.687622898097951e-05, | |
| "loss": 0.112, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.8269778553707617, | |
| "grad_norm": 0.7113578915596008, | |
| "learning_rate": 1.6692456124230453e-05, | |
| "loss": 0.1082, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.8269778553707617, | |
| "eval_accuracy": 0.9588998818739956, | |
| "eval_f1": 0.7713698035177239, | |
| "eval_loss": 0.10946337133646011, | |
| "eval_precision": 0.78099547261026, | |
| "eval_recall": 0.7619785159781004, | |
| "eval_runtime": 477.2564, | |
| "eval_samples_per_second": 60.008, | |
| "eval_steps_per_second": 1.201, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.8729210695580263, | |
| "grad_norm": 0.4333978593349457, | |
| "learning_rate": 1.6508683267481394e-05, | |
| "loss": 0.1116, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.9188642837452908, | |
| "grad_norm": 0.3579271137714386, | |
| "learning_rate": 1.6324910410732335e-05, | |
| "loss": 0.1101, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.9188642837452908, | |
| "eval_accuracy": 0.9590893651107495, | |
| "eval_f1": 0.7721416651417115, | |
| "eval_loss": 0.10736154764890671, | |
| "eval_precision": 0.7816850586599052, | |
| "eval_recall": 0.7628284867013311, | |
| "eval_runtime": 478.6898, | |
| "eval_samples_per_second": 59.828, | |
| "eval_steps_per_second": 1.197, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.9648074979325554, | |
| "grad_norm": 0.4602518677711487, | |
| "learning_rate": 1.6141137553983276e-05, | |
| "loss": 0.1101, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.0107507121198198, | |
| "grad_norm": 0.5322664976119995, | |
| "learning_rate": 1.5957364697234217e-05, | |
| "loss": 0.1101, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.0107507121198198, | |
| "eval_accuracy": 0.9591349931811384, | |
| "eval_f1": 0.7744744313952638, | |
| "eval_loss": 0.10792569071054459, | |
| "eval_precision": 0.7768799906601179, | |
| "eval_recall": 0.7720837234653981, | |
| "eval_runtime": 477.7344, | |
| "eval_samples_per_second": 59.948, | |
| "eval_steps_per_second": 1.199, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.0566939263070845, | |
| "grad_norm": 0.5673955678939819, | |
| "learning_rate": 1.5773591840485162e-05, | |
| "loss": 0.1071, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.102637140494349, | |
| "grad_norm": 0.42657607793807983, | |
| "learning_rate": 1.5589818983736103e-05, | |
| "loss": 0.1058, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.102637140494349, | |
| "eval_accuracy": 0.9584847931780965, | |
| "eval_f1": 0.7735292150015388, | |
| "eval_loss": 0.10933595150709152, | |
| "eval_precision": 0.7671773857517072, | |
| "eval_recall": 0.779987102031565, | |
| "eval_runtime": 475.9965, | |
| "eval_samples_per_second": 60.166, | |
| "eval_steps_per_second": 1.204, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.1485803546816136, | |
| "grad_norm": 0.4599572718143463, | |
| "learning_rate": 1.5406046126987044e-05, | |
| "loss": 0.107, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.194523568868878, | |
| "grad_norm": 0.4379059970378876, | |
| "learning_rate": 1.5222273270237986e-05, | |
| "loss": 0.107, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.194523568868878, | |
| "eval_accuracy": 0.959601413456225, | |
| "eval_f1": 0.7737560775183182, | |
| "eval_loss": 0.1075456365942955, | |
| "eval_precision": 0.7856481314090709, | |
| "eval_recall": 0.76221866643641, | |
| "eval_runtime": 478.1835, | |
| "eval_samples_per_second": 59.891, | |
| "eval_steps_per_second": 1.198, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.2404667830561427, | |
| "grad_norm": 0.43774718046188354, | |
| "learning_rate": 1.5038500413488928e-05, | |
| "loss": 0.1055, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.286409997243407, | |
| "grad_norm": 0.4196317195892334, | |
| "learning_rate": 1.485472755673987e-05, | |
| "loss": 0.1055, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.286409997243407, | |
| "eval_accuracy": 0.959860606244962, | |
| "eval_f1": 0.7768597287158228, | |
| "eval_loss": 0.10658087581396103, | |
| "eval_precision": 0.7824100936011823, | |
| "eval_recall": 0.7713875569682759, | |
| "eval_runtime": 464.4067, | |
| "eval_samples_per_second": 61.668, | |
| "eval_steps_per_second": 1.234, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.3323532114306718, | |
| "grad_norm": 0.36208856105804443, | |
| "learning_rate": 1.467095469999081e-05, | |
| "loss": 0.1035, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.3782964256179362, | |
| "grad_norm": 0.5692697763442993, | |
| "learning_rate": 1.4487181843241755e-05, | |
| "loss": 0.1054, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.3782964256179362, | |
| "eval_accuracy": 0.9593396857746885, | |
| "eval_f1": 0.7750922296908528, | |
| "eval_loss": 0.10651089251041412, | |
| "eval_precision": 0.7777847213920075, | |
| "eval_recall": 0.7724183151151778, | |
| "eval_runtime": 484.7629, | |
| "eval_samples_per_second": 59.078, | |
| "eval_steps_per_second": 1.182, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.4242396398052009, | |
| "grad_norm": 0.4805659055709839, | |
| "learning_rate": 1.4303408986492697e-05, | |
| "loss": 0.1043, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.4701828539924653, | |
| "grad_norm": 0.5373257994651794, | |
| "learning_rate": 1.411963612974364e-05, | |
| "loss": 0.1038, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.4701828539924653, | |
| "eval_accuracy": 0.9594822734946539, | |
| "eval_f1": 0.77620383584155, | |
| "eval_loss": 0.10620830953121185, | |
| "eval_precision": 0.7775004264051742, | |
| "eval_recall": 0.7749115625699877, | |
| "eval_runtime": 484.6548, | |
| "eval_samples_per_second": 59.092, | |
| "eval_steps_per_second": 1.182, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.5161260681797297, | |
| "grad_norm": 0.5227932929992676, | |
| "learning_rate": 1.393586327299458e-05, | |
| "loss": 0.1057, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.5620692823669944, | |
| "grad_norm": 0.2855619192123413, | |
| "learning_rate": 1.3752090416245523e-05, | |
| "loss": 0.1052, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.5620692823669944, | |
| "eval_accuracy": 0.9600310777857204, | |
| "eval_f1": 0.7757748564065305, | |
| "eval_loss": 0.10526877641677856, | |
| "eval_precision": 0.7884331924355441, | |
| "eval_recall": 0.7635165582391844, | |
| "eval_runtime": 483.926, | |
| "eval_samples_per_second": 59.181, | |
| "eval_steps_per_second": 1.184, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.608012496554259, | |
| "grad_norm": 0.36337897181510925, | |
| "learning_rate": 1.3568317559496465e-05, | |
| "loss": 0.1058, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.6539557107415235, | |
| "grad_norm": 0.7537007927894592, | |
| "learning_rate": 1.3384544702747406e-05, | |
| "loss": 0.1061, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.6539557107415235, | |
| "eval_accuracy": 0.9598663097537605, | |
| "eval_f1": 0.7781213928303998, | |
| "eval_loss": 0.1061909943819046, | |
| "eval_precision": 0.7795501631707945, | |
| "eval_recall": 0.7766978502486501, | |
| "eval_runtime": 483.1917, | |
| "eval_samples_per_second": 59.27, | |
| "eval_steps_per_second": 1.186, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.699898924928788, | |
| "grad_norm": 0.7511648535728455, | |
| "learning_rate": 1.3200771845998349e-05, | |
| "loss": 0.1035, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.7458421391160526, | |
| "grad_norm": 0.5765513181686401, | |
| "learning_rate": 1.301699898924929e-05, | |
| "loss": 0.1038, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.7458421391160526, | |
| "eval_accuracy": 0.9596578148210112, | |
| "eval_f1": 0.7782083489589002, | |
| "eval_loss": 0.10631386935710907, | |
| "eval_precision": 0.7753205917958497, | |
| "eval_recall": 0.7811176980094495, | |
| "eval_runtime": 481.5937, | |
| "eval_samples_per_second": 59.467, | |
| "eval_steps_per_second": 1.19, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.7917853533033172, | |
| "grad_norm": 1.3671506643295288, | |
| "learning_rate": 1.2833226132500231e-05, | |
| "loss": 0.1037, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.8377285674905817, | |
| "grad_norm": 0.6475801467895508, | |
| "learning_rate": 1.2649453275751174e-05, | |
| "loss": 0.1017, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.8377285674905817, | |
| "eval_accuracy": 0.9603023113152545, | |
| "eval_f1": 0.7790918283030356, | |
| "eval_loss": 0.10478387773036957, | |
| "eval_precision": 0.784845306908137, | |
| "eval_recall": 0.7734220900645168, | |
| "eval_runtime": 483.6282, | |
| "eval_samples_per_second": 59.217, | |
| "eval_steps_per_second": 1.185, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.883671781677846, | |
| "grad_norm": 0.898071825504303, | |
| "learning_rate": 1.2465680419002115e-05, | |
| "loss": 0.1027, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.9296149958651108, | |
| "grad_norm": 0.6122450232505798, | |
| "learning_rate": 1.2281907562253056e-05, | |
| "loss": 0.1029, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.9296149958651108, | |
| "eval_accuracy": 0.96032068928805, | |
| "eval_f1": 0.7800494094554808, | |
| "eval_loss": 0.10455518215894699, | |
| "eval_precision": 0.7826737729678044, | |
| "eval_recall": 0.7774425865013855, | |
| "eval_runtime": 483.8492, | |
| "eval_samples_per_second": 59.19, | |
| "eval_steps_per_second": 1.184, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.9755582100523754, | |
| "grad_norm": 0.6479480266571045, | |
| "learning_rate": 1.2098134705504e-05, | |
| "loss": 0.1032, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.0215014242396396, | |
| "grad_norm": 0.9193481206893921, | |
| "learning_rate": 1.191436184875494e-05, | |
| "loss": 0.1018, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.0215014242396396, | |
| "eval_accuracy": 0.9604075093664289, | |
| "eval_f1": 0.7798349386244161, | |
| "eval_loss": 0.10434387624263763, | |
| "eval_precision": 0.7843734048157843, | |
| "eval_recall": 0.7753486903705062, | |
| "eval_runtime": 482.4895, | |
| "eval_samples_per_second": 59.357, | |
| "eval_steps_per_second": 1.188, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.0674446384269043, | |
| "grad_norm": 0.41459354758262634, | |
| "learning_rate": 1.1730588992005882e-05, | |
| "loss": 0.1015, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.113387852614169, | |
| "grad_norm": 0.5848496556282043, | |
| "learning_rate": 1.1546816135256825e-05, | |
| "loss": 0.1007, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.113387852614169, | |
| "eval_accuracy": 0.960530451667199, | |
| "eval_f1": 0.7787409881226326, | |
| "eval_loss": 0.10433882474899292, | |
| "eval_precision": 0.7888630191217876, | |
| "eval_recall": 0.7688754212751719, | |
| "eval_runtime": 481.4982, | |
| "eval_samples_per_second": 59.479, | |
| "eval_steps_per_second": 1.19, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.1593310668014336, | |
| "grad_norm": 0.5300314426422119, | |
| "learning_rate": 1.1363043278507766e-05, | |
| "loss": 0.1008, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.205274280988698, | |
| "grad_norm": 0.34308552742004395, | |
| "learning_rate": 1.1179270421758709e-05, | |
| "loss": 0.1008, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.205274280988698, | |
| "eval_accuracy": 0.9605203120960014, | |
| "eval_f1": 0.7802636835797044, | |
| "eval_loss": 0.10480789840221405, | |
| "eval_precision": 0.7859456144721461, | |
| "eval_recall": 0.7746633171524092, | |
| "eval_runtime": 482.2132, | |
| "eval_samples_per_second": 59.391, | |
| "eval_steps_per_second": 1.188, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.2512174951759625, | |
| "grad_norm": 0.668408215045929, | |
| "learning_rate": 1.099549756500965e-05, | |
| "loss": 0.101, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.297160709363227, | |
| "grad_norm": 0.6432192921638489, | |
| "learning_rate": 1.0811724708260591e-05, | |
| "loss": 0.1015, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.297160709363227, | |
| "eval_accuracy": 0.9606768417263634, | |
| "eval_f1": 0.7810441317665336, | |
| "eval_loss": 0.10385506600141525, | |
| "eval_precision": 0.78621070532204, | |
| "eval_recall": 0.7759450190366459, | |
| "eval_runtime": 480.9669, | |
| "eval_samples_per_second": 59.545, | |
| "eval_steps_per_second": 1.191, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.3431039235504914, | |
| "grad_norm": 0.6759513020515442, | |
| "learning_rate": 1.0627951851511534e-05, | |
| "loss": 0.1007, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.389047137737756, | |
| "grad_norm": 0.9079370498657227, | |
| "learning_rate": 1.0444178994762475e-05, | |
| "loss": 0.1008, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.389047137737756, | |
| "eval_accuracy": 0.960218026129675, | |
| "eval_f1": 0.7791951377937008, | |
| "eval_loss": 0.10625031590461731, | |
| "eval_precision": 0.7831873122488701, | |
| "eval_recall": 0.7752434559000111, | |
| "eval_runtime": 480.1994, | |
| "eval_samples_per_second": 59.64, | |
| "eval_steps_per_second": 1.193, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.4349903519250207, | |
| "grad_norm": 0.5865328907966614, | |
| "learning_rate": 1.0260406138013416e-05, | |
| "loss": 0.1009, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.4809335661122853, | |
| "grad_norm": 0.4314309060573578, | |
| "learning_rate": 1.0076633281264359e-05, | |
| "loss": 0.1003, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.4809335661122853, | |
| "eval_accuracy": 0.9607592257423434, | |
| "eval_f1": 0.7807987830560178, | |
| "eval_loss": 0.10385860502719879, | |
| "eval_precision": 0.7889295945350374, | |
| "eval_recall": 0.7728338563576461, | |
| "eval_runtime": 480.1646, | |
| "eval_samples_per_second": 59.644, | |
| "eval_steps_per_second": 1.193, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.52687678029955, | |
| "grad_norm": 0.5162900686264038, | |
| "learning_rate": 9.8928604245153e-06, | |
| "loss": 0.102, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.572819994486814, | |
| "grad_norm": 0.8881911039352417, | |
| "learning_rate": 9.709087567766241e-06, | |
| "loss": 0.0998, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.572819994486814, | |
| "eval_accuracy": 0.9608099235983311, | |
| "eval_f1": 0.7834099891185191, | |
| "eval_loss": 0.10304800420999527, | |
| "eval_precision": 0.7829613019033707, | |
| "eval_recall": 0.7838591908818379, | |
| "eval_runtime": 481.0354, | |
| "eval_samples_per_second": 59.536, | |
| "eval_steps_per_second": 1.191, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.618763208674079, | |
| "grad_norm": 0.36895298957824707, | |
| "learning_rate": 9.525314711017184e-06, | |
| "loss": 0.0989, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.6647064228613435, | |
| "grad_norm": 0.4097418785095215, | |
| "learning_rate": 9.341541854268126e-06, | |
| "loss": 0.0991, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.6647064228613435, | |
| "eval_accuracy": 0.9608473132671219, | |
| "eval_f1": 0.7830652507074841, | |
| "eval_loss": 0.10377328097820282, | |
| "eval_precision": 0.7843967022702352, | |
| "eval_recall": 0.7817383115533957, | |
| "eval_runtime": 494.1609, | |
| "eval_samples_per_second": 57.955, | |
| "eval_steps_per_second": 1.16, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.7106496370486077, | |
| "grad_norm": 0.4404696524143219, | |
| "learning_rate": 9.157768997519067e-06, | |
| "loss": 0.0977, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.7565928512358724, | |
| "grad_norm": 0.45993557572364807, | |
| "learning_rate": 8.97399614077001e-06, | |
| "loss": 0.0971, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.7565928512358724, | |
| "eval_accuracy": 0.9610475697982732, | |
| "eval_f1": 0.7825106312355632, | |
| "eval_loss": 0.10452987253665924, | |
| "eval_precision": 0.7890627357701161, | |
| "eval_recall": 0.7760664434256789, | |
| "eval_runtime": 494.0836, | |
| "eval_samples_per_second": 57.964, | |
| "eval_steps_per_second": 1.16, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.802536065423137, | |
| "grad_norm": 0.506994366645813, | |
| "learning_rate": 8.79022328402095e-06, | |
| "loss": 0.0965, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.8484792796104017, | |
| "grad_norm": 0.6458877325057983, | |
| "learning_rate": 8.606450427271894e-06, | |
| "loss": 0.0961, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.8484792796104017, | |
| "eval_accuracy": 0.9609030809087084, | |
| "eval_f1": 0.7821163878725582, | |
| "eval_loss": 0.10469945520162582, | |
| "eval_precision": 0.788239130911403, | |
| "eval_recall": 0.7760880299837292, | |
| "eval_runtime": 493.3004, | |
| "eval_samples_per_second": 58.056, | |
| "eval_steps_per_second": 1.162, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.894422493797666, | |
| "grad_norm": 0.5674518942832947, | |
| "learning_rate": 8.422677570522835e-06, | |
| "loss": 0.0968, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.9403657079849306, | |
| "grad_norm": 0.6595680117607117, | |
| "learning_rate": 8.238904713773776e-06, | |
| "loss": 0.0958, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.9403657079849306, | |
| "eval_accuracy": 0.9610779885118659, | |
| "eval_f1": 0.7836532860065301, | |
| "eval_loss": 0.10300234705209732, | |
| "eval_precision": 0.7869283784408849, | |
| "eval_recall": 0.7804053415937895, | |
| "eval_runtime": 493.1751, | |
| "eval_samples_per_second": 58.071, | |
| "eval_steps_per_second": 1.162, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.9863089221721952, | |
| "grad_norm": 0.5502172112464905, | |
| "learning_rate": 8.055131857024719e-06, | |
| "loss": 0.0971, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.03225213635946, | |
| "grad_norm": 0.5948086380958557, | |
| "learning_rate": 7.87135900027566e-06, | |
| "loss": 0.0963, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.03225213635946, | |
| "eval_accuracy": 0.9608828017663134, | |
| "eval_f1": 0.7828877352895777, | |
| "eval_loss": 0.10367628186941147, | |
| "eval_precision": 0.7857005030479105, | |
| "eval_recall": 0.7800950348218164, | |
| "eval_runtime": 492.9957, | |
| "eval_samples_per_second": 58.092, | |
| "eval_steps_per_second": 1.162, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.078195350546724, | |
| "grad_norm": 0.5747931599617004, | |
| "learning_rate": 7.687586143526601e-06, | |
| "loss": 0.0961, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.1241385647339888, | |
| "grad_norm": 0.5392933487892151, | |
| "learning_rate": 7.503813286777543e-06, | |
| "loss": 0.0972, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.1241385647339888, | |
| "eval_accuracy": 0.9612465588830248, | |
| "eval_f1": 0.7833850043316463, | |
| "eval_loss": 0.10355565696954727, | |
| "eval_precision": 0.7910026930303056, | |
| "eval_recall": 0.7759126391995704, | |
| "eval_runtime": 493.4779, | |
| "eval_samples_per_second": 58.035, | |
| "eval_steps_per_second": 1.161, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.1700817789212534, | |
| "grad_norm": 0.47470977902412415, | |
| "learning_rate": 7.3200404300284854e-06, | |
| "loss": 0.0963, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.216024993108518, | |
| "grad_norm": 0.4675300121307373, | |
| "learning_rate": 7.1362675732794275e-06, | |
| "loss": 0.0966, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.216024993108518, | |
| "eval_accuracy": 0.9612091692142339, | |
| "eval_f1": 0.7841735299101847, | |
| "eval_loss": 0.1029718816280365, | |
| "eval_precision": 0.7879947905314094, | |
| "eval_recall": 0.7803891516752518, | |
| "eval_runtime": 493.9167, | |
| "eval_samples_per_second": 57.983, | |
| "eval_steps_per_second": 1.16, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.2619682072957823, | |
| "grad_norm": 0.65328049659729, | |
| "learning_rate": 6.9524947165303695e-06, | |
| "loss": 0.0981, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 3.307911421483047, | |
| "grad_norm": 0.8536853194236755, | |
| "learning_rate": 6.768721859781311e-06, | |
| "loss": 0.0972, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.307911421483047, | |
| "eval_accuracy": 0.9613213382206066, | |
| "eval_f1": 0.7841876799047575, | |
| "eval_loss": 0.10238692909479141, | |
| "eval_precision": 0.7899809695924207, | |
| "eval_recall": 0.7784787412878, | |
| "eval_runtime": 493.9879, | |
| "eval_samples_per_second": 57.975, | |
| "eval_steps_per_second": 1.16, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.3538546356703116, | |
| "grad_norm": 0.4792385995388031, | |
| "learning_rate": 6.584949003032253e-06, | |
| "loss": 0.0961, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 3.399797849857576, | |
| "grad_norm": 0.5664140582084656, | |
| "learning_rate": 6.401176146283195e-06, | |
| "loss": 0.0961, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.399797849857576, | |
| "eval_accuracy": 0.9612731752574184, | |
| "eval_f1": 0.7838260321119623, | |
| "eval_loss": 0.10364777594804764, | |
| "eval_precision": 0.7897582688488667, | |
| "eval_recall": 0.7779822504526431, | |
| "eval_runtime": 487.6212, | |
| "eval_samples_per_second": 58.732, | |
| "eval_steps_per_second": 1.175, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.4457410640448405, | |
| "grad_norm": 0.4458509087562561, | |
| "learning_rate": 6.217403289534136e-06, | |
| "loss": 0.0956, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 3.491684278232105, | |
| "grad_norm": 1.0309734344482422, | |
| "learning_rate": 6.033630432785078e-06, | |
| "loss": 0.0968, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.491684278232105, | |
| "eval_accuracy": 0.9613644313981962, | |
| "eval_f1": 0.7849622795571747, | |
| "eval_loss": 0.10237538069486618, | |
| "eval_precision": 0.7891754053427515, | |
| "eval_recall": 0.7807938996386949, | |
| "eval_runtime": 489.0957, | |
| "eval_samples_per_second": 58.555, | |
| "eval_steps_per_second": 1.172, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.53762749241937, | |
| "grad_norm": 0.6991232633590698, | |
| "learning_rate": 5.84985757603602e-06, | |
| "loss": 0.0964, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 3.5835707066066345, | |
| "grad_norm": 0.6872137784957886, | |
| "learning_rate": 5.666084719286962e-06, | |
| "loss": 0.0956, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 3.5835707066066345, | |
| "eval_accuracy": 0.9613866117101908, | |
| "eval_f1": 0.7850481992416146, | |
| "eval_loss": 0.1031898632645607, | |
| "eval_precision": 0.789164428569481, | |
| "eval_recall": 0.7809746870623663, | |
| "eval_runtime": 488.2001, | |
| "eval_samples_per_second": 58.662, | |
| "eval_steps_per_second": 1.174, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 3.6295139207938987, | |
| "grad_norm": 0.5371155738830566, | |
| "learning_rate": 5.482311862537903e-06, | |
| "loss": 0.0947, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 3.6754571349811633, | |
| "grad_norm": 0.5987829566001892, | |
| "learning_rate": 5.298539005788845e-06, | |
| "loss": 0.0946, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.6754571349811633, | |
| "eval_accuracy": 0.9612731752574184, | |
| "eval_f1": 0.7850010409935133, | |
| "eval_loss": 0.10323267430067062, | |
| "eval_precision": 0.7866217618545723, | |
| "eval_recall": 0.7833869849244876, | |
| "eval_runtime": 488.7815, | |
| "eval_samples_per_second": 58.593, | |
| "eval_steps_per_second": 1.172, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.7214003491684275, | |
| "grad_norm": 0.4283543825149536, | |
| "learning_rate": 5.114766149039787e-06, | |
| "loss": 0.0931, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 3.767343563355692, | |
| "grad_norm": 1.0369564294815063, | |
| "learning_rate": 4.9309932922907285e-06, | |
| "loss": 0.0933, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 3.767343563355692, | |
| "eval_accuracy": 0.9615013156093629, | |
| "eval_f1": 0.7854287067997648, | |
| "eval_loss": 0.10276814550161362, | |
| "eval_precision": 0.7903677069273602, | |
| "eval_recall": 0.780551050860629, | |
| "eval_runtime": 488.7183, | |
| "eval_samples_per_second": 58.6, | |
| "eval_steps_per_second": 1.172, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 3.813286777542957, | |
| "grad_norm": 0.5058717131614685, | |
| "learning_rate": 4.7472204355416705e-06, | |
| "loss": 0.0914, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 3.8592299917302215, | |
| "grad_norm": 0.5976991057395935, | |
| "learning_rate": 4.5634475787926126e-06, | |
| "loss": 0.0928, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.8592299917302215, | |
| "eval_accuracy": 0.9614994144397634, | |
| "eval_f1": 0.784492448549999, | |
| "eval_loss": 0.1031479462981224, | |
| "eval_precision": 0.7923046862099713, | |
| "eval_recall": 0.7768327662364646, | |
| "eval_runtime": 489.9123, | |
| "eval_samples_per_second": 58.457, | |
| "eval_steps_per_second": 1.17, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.905173205917486, | |
| "grad_norm": 0.45505064725875854, | |
| "learning_rate": 4.379674722043555e-06, | |
| "loss": 0.0922, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 3.9511164201047504, | |
| "grad_norm": 0.5285763740539551, | |
| "learning_rate": 4.195901865294497e-06, | |
| "loss": 0.0924, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 3.9511164201047504, | |
| "eval_accuracy": 0.961282681105416, | |
| "eval_f1": 0.7860457849952998, | |
| "eval_loss": 0.10335655510425568, | |
| "eval_precision": 0.7846593154850732, | |
| "eval_recall": 0.7874371628786755, | |
| "eval_runtime": 488.6406, | |
| "eval_samples_per_second": 58.61, | |
| "eval_steps_per_second": 1.173, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 3.997059634292015, | |
| "grad_norm": 0.5644091367721558, | |
| "learning_rate": 4.012129008545439e-06, | |
| "loss": 0.0931, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 4.043002848479279, | |
| "grad_norm": 0.6456932425498962, | |
| "learning_rate": 3.82835615179638e-06, | |
| "loss": 0.0939, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 4.043002848479279, | |
| "eval_accuracy": 0.9612351518654276, | |
| "eval_f1": 0.786335258531395, | |
| "eval_loss": 0.10406021773815155, | |
| "eval_precision": 0.7830351743734916, | |
| "eval_recall": 0.7896632766776128, | |
| "eval_runtime": 488.3357, | |
| "eval_samples_per_second": 58.646, | |
| "eval_steps_per_second": 1.173, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 4.088946062666544, | |
| "grad_norm": 0.4968123137950897, | |
| "learning_rate": 3.644583295047322e-06, | |
| "loss": 0.0946, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 4.134889276853809, | |
| "grad_norm": 0.6218472719192505, | |
| "learning_rate": 3.460810438298264e-06, | |
| "loss": 0.0946, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 4.134889276853809, | |
| "eval_accuracy": 0.9616508742845266, | |
| "eval_f1": 0.7865272415737647, | |
| "eval_loss": 0.10293179005384445, | |
| "eval_precision": 0.790283169206881, | |
| "eval_recall": 0.7828068461768857, | |
| "eval_runtime": 489.0208, | |
| "eval_samples_per_second": 58.564, | |
| "eval_steps_per_second": 1.172, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 4.180832491041073, | |
| "grad_norm": 0.5280947685241699, | |
| "learning_rate": 3.2770375815492056e-06, | |
| "loss": 0.0948, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 4.226775705228338, | |
| "grad_norm": 0.759973406791687, | |
| "learning_rate": 3.0932647248001476e-06, | |
| "loss": 0.0939, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 4.226775705228338, | |
| "eval_accuracy": 0.9616014438749385, | |
| "eval_f1": 0.7859887986330535, | |
| "eval_loss": 0.1021941676735878, | |
| "eval_precision": 0.7900514981911567, | |
| "eval_recall": 0.7819676687326802, | |
| "eval_runtime": 489.6378, | |
| "eval_samples_per_second": 58.49, | |
| "eval_steps_per_second": 1.17, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 4.272718919415603, | |
| "grad_norm": 0.555004358291626, | |
| "learning_rate": 2.909491868051089e-06, | |
| "loss": 0.0953, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 4.318662133602867, | |
| "grad_norm": 0.792995035648346, | |
| "learning_rate": 2.725719011302031e-06, | |
| "loss": 0.0931, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 4.318662133602867, | |
| "eval_accuracy": 0.9615406064477533, | |
| "eval_f1": 0.7852703901371545, | |
| "eval_loss": 0.10258933901786804, | |
| "eval_precision": 0.7906673304593299, | |
| "eval_recall": 0.7799466272352207, | |
| "eval_runtime": 489.6821, | |
| "eval_samples_per_second": 58.485, | |
| "eval_steps_per_second": 1.17, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 4.364605347790131, | |
| "grad_norm": 0.40561413764953613, | |
| "learning_rate": 2.541946154552973e-06, | |
| "loss": 0.0943, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 4.410548561977396, | |
| "grad_norm": 0.5800179243087769, | |
| "learning_rate": 2.3581732978039145e-06, | |
| "loss": 0.0927, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 4.410548561977396, | |
| "eval_accuracy": 0.9616705197037217, | |
| "eval_f1": 0.7859697371567711, | |
| "eval_loss": 0.10238471627235413, | |
| "eval_precision": 0.7916168287482448, | |
| "eval_recall": 0.7804026432740333, | |
| "eval_runtime": 489.7062, | |
| "eval_samples_per_second": 58.482, | |
| "eval_steps_per_second": 1.17, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 4.45649177616466, | |
| "grad_norm": 0.599105179309845, | |
| "learning_rate": 2.1744004410548565e-06, | |
| "loss": 0.0944, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 4.502434990351925, | |
| "grad_norm": 0.8176274299621582, | |
| "learning_rate": 1.990627584305798e-06, | |
| "loss": 0.0942, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 4.502434990351925, | |
| "eval_accuracy": 0.9617782526476956, | |
| "eval_f1": 0.7869475259848198, | |
| "eval_loss": 0.10218960046768188, | |
| "eval_precision": 0.7908678412296121, | |
| "eval_recall": 0.7830658848734893, | |
| "eval_runtime": 488.7041, | |
| "eval_samples_per_second": 58.602, | |
| "eval_steps_per_second": 1.172, | |
| "step": 49000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 54415, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.696790502219255e+17, | |
| "train_batch_size": 50, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |