| { | |
| "best_metric": 0.6252569868800233, | |
| "best_model_checkpoint": "./CARES/checkpoints/bio-ber-stratified/run-1/checkpoint-3408", | |
| "epoch": 24.170212765957448, | |
| "global_step": 3408, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 0.2541882395744324, | |
| "eval_macro_f1": 0.026901004304160685, | |
| "eval_macro_precision": 0.048828125, | |
| "eval_macro_recall": 0.018564356435643563, | |
| "eval_micro_f1": 0.17543859649122806, | |
| "eval_micro_precision": 0.78125, | |
| "eval_micro_recall": 0.09881422924901186, | |
| "eval_runtime": 9.8993, | |
| "eval_samples_per_second": 98.088, | |
| "eval_steps_per_second": 3.132, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 0.22154481709003448, | |
| "eval_macro_f1": 0.10623663256202004, | |
| "eval_macro_precision": 0.18484581306311026, | |
| "eval_macro_recall": 0.09432864376066943, | |
| "eval_micro_f1": 0.41890639481000924, | |
| "eval_micro_precision": 0.70625, | |
| "eval_micro_recall": 0.2977602108036891, | |
| "eval_runtime": 9.9434, | |
| "eval_samples_per_second": 97.653, | |
| "eval_steps_per_second": 3.118, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_loss": 0.18873167037963867, | |
| "eval_macro_f1": 0.20450440261358283, | |
| "eval_macro_precision": 0.3100231872943203, | |
| "eval_macro_recall": 0.1738961275122085, | |
| "eval_micro_f1": 0.568944099378882, | |
| "eval_micro_precision": 0.7658862876254181, | |
| "eval_micro_recall": 0.4525691699604743, | |
| "eval_runtime": 9.9071, | |
| "eval_samples_per_second": 98.011, | |
| "eval_steps_per_second": 3.129, | |
| "step": 426 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 4.1013980584550545e-05, | |
| "loss": 0.2716, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "eval_loss": 0.17541566491127014, | |
| "eval_macro_f1": 0.2919005682162893, | |
| "eval_macro_precision": 0.3812561939105582, | |
| "eval_macro_recall": 0.25249891158193927, | |
| "eval_micro_f1": 0.6301369863013698, | |
| "eval_micro_precision": 0.7762777242044359, | |
| "eval_micro_recall": 0.5303030303030303, | |
| "eval_runtime": 9.944, | |
| "eval_samples_per_second": 97.647, | |
| "eval_steps_per_second": 3.117, | |
| "step": 568 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "eval_loss": 0.1692640334367752, | |
| "eval_macro_f1": 0.4143233067428529, | |
| "eval_macro_precision": 0.5073151086851961, | |
| "eval_macro_recall": 0.3678263076835473, | |
| "eval_micro_f1": 0.6415811478525276, | |
| "eval_micro_precision": 0.7583108715184187, | |
| "eval_micro_recall": 0.5559947299077734, | |
| "eval_runtime": 9.9093, | |
| "eval_samples_per_second": 97.989, | |
| "eval_steps_per_second": 3.128, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "eval_loss": 0.1651107370853424, | |
| "eval_macro_f1": 0.4093940822021235, | |
| "eval_macro_precision": 0.5308122229847152, | |
| "eval_macro_recall": 0.358915994433547, | |
| "eval_micro_f1": 0.6615737203972498, | |
| "eval_micro_precision": 0.7872727272727272, | |
| "eval_micro_recall": 0.5704874835309618, | |
| "eval_runtime": 9.8986, | |
| "eval_samples_per_second": 98.095, | |
| "eval_steps_per_second": 3.132, | |
| "step": 852 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "eval_loss": 0.1729801893234253, | |
| "eval_macro_f1": 0.42556225648133283, | |
| "eval_macro_precision": 0.58718674661248, | |
| "eval_macro_recall": 0.3723531112715551, | |
| "eval_micro_f1": 0.6549062844542448, | |
| "eval_micro_precision": 0.7406483790523691, | |
| "eval_micro_recall": 0.5869565217391305, | |
| "eval_runtime": 9.912, | |
| "eval_samples_per_second": 97.962, | |
| "eval_steps_per_second": 3.128, | |
| "step": 994 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 3.950611365129501e-05, | |
| "loss": 0.1249, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_loss": 0.17469000816345215, | |
| "eval_macro_f1": 0.4273735214628892, | |
| "eval_macro_precision": 0.6238596091113585, | |
| "eval_macro_recall": 0.375472703111614, | |
| "eval_micro_f1": 0.6517101875689592, | |
| "eval_micro_precision": 0.7377185678601166, | |
| "eval_micro_recall": 0.5836627140974967, | |
| "eval_runtime": 9.8904, | |
| "eval_samples_per_second": 98.176, | |
| "eval_steps_per_second": 3.134, | |
| "step": 1136 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "eval_loss": 0.17877863347530365, | |
| "eval_macro_f1": 0.475201225203798, | |
| "eval_macro_precision": 0.5739651831374122, | |
| "eval_macro_recall": 0.4350214360169413, | |
| "eval_micro_f1": 0.6738227146814404, | |
| "eval_micro_precision": 0.7102189781021898, | |
| "eval_micro_recall": 0.6409749670619236, | |
| "eval_runtime": 9.9151, | |
| "eval_samples_per_second": 97.932, | |
| "eval_steps_per_second": 3.127, | |
| "step": 1278 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "eval_loss": 0.17759235203266144, | |
| "eval_macro_f1": 0.47628022362120415, | |
| "eval_macro_precision": 0.6058635009097308, | |
| "eval_macro_recall": 0.42583522254336503, | |
| "eval_micro_f1": 0.6828591256072173, | |
| "eval_micro_precision": 0.7214076246334311, | |
| "eval_micro_recall": 0.6482213438735178, | |
| "eval_runtime": 9.9061, | |
| "eval_samples_per_second": 98.021, | |
| "eval_steps_per_second": 3.129, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 10.64, | |
| "learning_rate": 3.7998246718039476e-05, | |
| "loss": 0.0762, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "eval_loss": 0.1776473969221115, | |
| "eval_macro_f1": 0.5089905336706273, | |
| "eval_macro_precision": 0.591647351441912, | |
| "eval_macro_recall": 0.47047423582847886, | |
| "eval_micro_f1": 0.688728024819028, | |
| "eval_micro_precision": 0.7223427331887202, | |
| "eval_micro_recall": 0.658102766798419, | |
| "eval_runtime": 9.9428, | |
| "eval_samples_per_second": 97.658, | |
| "eval_steps_per_second": 3.118, | |
| "step": 1562 | |
| }, | |
| { | |
| "epoch": 12.09, | |
| "eval_loss": 0.18189238011837006, | |
| "eval_macro_f1": 0.48004944819202805, | |
| "eval_macro_precision": 0.6458220963071541, | |
| "eval_macro_recall": 0.41943684599646036, | |
| "eval_micro_f1": 0.6883162725026473, | |
| "eval_micro_precision": 0.7414448669201521, | |
| "eval_micro_recall": 0.642292490118577, | |
| "eval_runtime": 9.9058, | |
| "eval_samples_per_second": 98.024, | |
| "eval_steps_per_second": 3.129, | |
| "step": 1704 | |
| }, | |
| { | |
| "epoch": 13.09, | |
| "eval_loss": 0.19002576172351837, | |
| "eval_macro_f1": 0.5199688644954614, | |
| "eval_macro_precision": 0.5810572518525183, | |
| "eval_macro_recall": 0.49026284720966035, | |
| "eval_micro_f1": 0.6896090878717006, | |
| "eval_micro_precision": 0.6996610169491525, | |
| "eval_micro_recall": 0.6798418972332015, | |
| "eval_runtime": 9.9018, | |
| "eval_samples_per_second": 98.063, | |
| "eval_steps_per_second": 3.131, | |
| "step": 1846 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "eval_loss": 0.19194385409355164, | |
| "eval_macro_f1": 0.520691157744967, | |
| "eval_macro_precision": 0.5966466597800141, | |
| "eval_macro_recall": 0.47615158217927345, | |
| "eval_micro_f1": 0.6891228070175438, | |
| "eval_micro_precision": 0.7372372372372372, | |
| "eval_micro_recall": 0.6469038208168643, | |
| "eval_runtime": 9.9153, | |
| "eval_samples_per_second": 97.93, | |
| "eval_steps_per_second": 3.126, | |
| "step": 1988 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "learning_rate": 3.649037978478394e-05, | |
| "loss": 0.0449, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "eval_loss": 0.192080557346344, | |
| "eval_macro_f1": 0.5605505080291033, | |
| "eval_macro_precision": 0.6623700312301602, | |
| "eval_macro_recall": 0.5087278128204601, | |
| "eval_micro_f1": 0.6957736639888229, | |
| "eval_micro_precision": 0.7405204460966542, | |
| "eval_micro_recall": 0.6561264822134387, | |
| "eval_runtime": 9.9139, | |
| "eval_samples_per_second": 97.943, | |
| "eval_steps_per_second": 3.127, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 16.11, | |
| "eval_loss": 0.1960616409778595, | |
| "eval_macro_f1": 0.53306733167955, | |
| "eval_macro_precision": 0.5965001445588101, | |
| "eval_macro_recall": 0.49021955336676715, | |
| "eval_micro_f1": 0.698961937716263, | |
| "eval_micro_precision": 0.7361516034985423, | |
| "eval_micro_recall": 0.6653491436100132, | |
| "eval_runtime": 9.942, | |
| "eval_samples_per_second": 97.667, | |
| "eval_steps_per_second": 3.118, | |
| "step": 2272 | |
| }, | |
| { | |
| "epoch": 17.12, | |
| "eval_loss": 0.20587308704853058, | |
| "eval_macro_f1": 0.5199165971603507, | |
| "eval_macro_precision": 0.5554999099415838, | |
| "eval_macro_recall": 0.501101816922742, | |
| "eval_micro_f1": 0.6840148698884758, | |
| "eval_micro_precision": 0.7022900763358778, | |
| "eval_micro_recall": 0.6666666666666666, | |
| "eval_runtime": 9.9026, | |
| "eval_samples_per_second": 98.055, | |
| "eval_steps_per_second": 3.131, | |
| "step": 2414 | |
| }, | |
| { | |
| "epoch": 17.73, | |
| "learning_rate": 3.498251285152841e-05, | |
| "loss": 0.0256, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 18.13, | |
| "eval_loss": 0.20113714039325714, | |
| "eval_macro_f1": 0.529618760993627, | |
| "eval_macro_precision": 0.6032995030406657, | |
| "eval_macro_recall": 0.4871012589073676, | |
| "eval_micro_f1": 0.6982167352537724, | |
| "eval_micro_precision": 0.7281831187410587, | |
| "eval_micro_recall": 0.6706192358366272, | |
| "eval_runtime": 9.899, | |
| "eval_samples_per_second": 98.091, | |
| "eval_steps_per_second": 3.132, | |
| "step": 2556 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "eval_loss": 0.20477713644504547, | |
| "eval_macro_f1": 0.5176514067210329, | |
| "eval_macro_precision": 0.5857117797598407, | |
| "eval_macro_recall": 0.47641342794341435, | |
| "eval_micro_f1": 0.6941015089163237, | |
| "eval_micro_precision": 0.7238912732474965, | |
| "eval_micro_recall": 0.6666666666666666, | |
| "eval_runtime": 9.9038, | |
| "eval_samples_per_second": 98.043, | |
| "eval_steps_per_second": 3.13, | |
| "step": 2698 | |
| }, | |
| { | |
| "epoch": 20.14, | |
| "eval_loss": 0.2074529379606247, | |
| "eval_macro_f1": 0.5704867306049771, | |
| "eval_macro_precision": 0.6916588415347732, | |
| "eval_macro_recall": 0.5189916858370269, | |
| "eval_micro_f1": 0.7134187457855697, | |
| "eval_micro_precision": 0.7306629834254144, | |
| "eval_micro_recall": 0.696969696969697, | |
| "eval_runtime": 9.8989, | |
| "eval_samples_per_second": 98.092, | |
| "eval_steps_per_second": 3.132, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 21.15, | |
| "eval_loss": 0.21805770695209503, | |
| "eval_macro_f1": 0.5306833888134577, | |
| "eval_macro_precision": 0.6107555764603518, | |
| "eval_macro_recall": 0.4823641731674541, | |
| "eval_micro_f1": 0.709366391184573, | |
| "eval_micro_precision": 0.7431457431457431, | |
| "eval_micro_recall": 0.6785243741765481, | |
| "eval_runtime": 9.95, | |
| "eval_samples_per_second": 97.588, | |
| "eval_steps_per_second": 3.116, | |
| "step": 2982 | |
| }, | |
| { | |
| "epoch": 21.28, | |
| "learning_rate": 3.347464591827287e-05, | |
| "loss": 0.015, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 22.16, | |
| "eval_loss": 0.21425750851631165, | |
| "eval_macro_f1": 0.5900896049733363, | |
| "eval_macro_precision": 0.7164448328913706, | |
| "eval_macro_recall": 0.5339604602685475, | |
| "eval_micro_f1": 0.7065292096219932, | |
| "eval_micro_precision": 0.7385057471264368, | |
| "eval_micro_recall": 0.6772068511198946, | |
| "eval_runtime": 9.9181, | |
| "eval_samples_per_second": 97.901, | |
| "eval_steps_per_second": 3.126, | |
| "step": 3124 | |
| }, | |
| { | |
| "epoch": 23.16, | |
| "eval_loss": 0.21841417253017426, | |
| "eval_macro_f1": 0.5731327712532719, | |
| "eval_macro_precision": 0.6766904212874656, | |
| "eval_macro_recall": 0.5216860842370524, | |
| "eval_micro_f1": 0.7131730443616662, | |
| "eval_micro_precision": 0.7337979094076655, | |
| "eval_micro_recall": 0.6936758893280632, | |
| "eval_runtime": 9.8943, | |
| "eval_samples_per_second": 98.137, | |
| "eval_steps_per_second": 3.133, | |
| "step": 3266 | |
| }, | |
| { | |
| "epoch": 24.17, | |
| "eval_loss": 0.22420497238636017, | |
| "eval_macro_f1": 0.6252569868800233, | |
| "eval_macro_precision": 0.7327575817423139, | |
| "eval_macro_recall": 0.5742398722065618, | |
| "eval_micro_f1": 0.7081471295978001, | |
| "eval_micro_precision": 0.7404744787922358, | |
| "eval_micro_recall": 0.6785243741765481, | |
| "eval_runtime": 9.961, | |
| "eval_samples_per_second": 97.48, | |
| "eval_steps_per_second": 3.112, | |
| "step": 3408 | |
| } | |
| ], | |
| "max_steps": 14100, | |
| "num_train_epochs": 100, | |
| "total_flos": 1.1986190383959552e+16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "adam_epsilon": 1.241521755885265e-07, | |
| "learning_rate": 4.181013432530947e-05, | |
| "per_device_eval_batch_size": 32, | |
| "per_device_train_batch_size": 16, | |
| "seed": 321, | |
| "warmup_steps": 236, | |
| "weight_decay": 3.190500833235664e-11 | |
| } | |
| } | |