| { | |
| "best_metric": 0.12050338089466095, | |
| "best_model_checkpoint": "/home/datawork-iot-nos/Seatizen/models/multilabel/fine_scale/DinoVdeau-large-2024_09_05-batch-size32_epochs150_freeze/checkpoint-22386", | |
| "epoch": 92.0, | |
| "eval_steps": 500, | |
| "global_step": 25116, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.22314622314622315, | |
| "eval_f1_macro": 0.5430112866470752, | |
| "eval_f1_micro": 0.7516596896274684, | |
| "eval_loss": 0.16899551451206207, | |
| "eval_roc_auc": 0.8384250127967259, | |
| "eval_runtime": 514.8139, | |
| "eval_samples_per_second": 5.606, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.001, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 1.8315018315018317, | |
| "grad_norm": 0.9695320725440979, | |
| "learning_rate": 0.001, | |
| "loss": 0.2719, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.24012474012474014, | |
| "eval_f1_macro": 0.5721428312627432, | |
| "eval_f1_micro": 0.765669700910273, | |
| "eval_loss": 0.153842031955719, | |
| "eval_roc_auc": 0.8396070197954885, | |
| "eval_runtime": 520.9151, | |
| "eval_samples_per_second": 5.54, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.001, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.23458073458073458, | |
| "eval_f1_macro": 0.6137585525531024, | |
| "eval_f1_micro": 0.7772688719253604, | |
| "eval_loss": 0.14828726649284363, | |
| "eval_roc_auc": 0.851613165443153, | |
| "eval_runtime": 519.0164, | |
| "eval_samples_per_second": 5.561, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.001, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 3.663003663003663, | |
| "grad_norm": 0.24952150881290436, | |
| "learning_rate": 0.001, | |
| "loss": 0.1694, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.2494802494802495, | |
| "eval_f1_macro": 0.6224730910908008, | |
| "eval_f1_micro": 0.7722737615963591, | |
| "eval_loss": 0.1479637324810028, | |
| "eval_roc_auc": 0.8406560025496872, | |
| "eval_runtime": 512.3298, | |
| "eval_samples_per_second": 5.633, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 0.001, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.2494802494802495, | |
| "eval_f1_macro": 0.6302307709949958, | |
| "eval_f1_micro": 0.779738930569409, | |
| "eval_loss": 0.14575305581092834, | |
| "eval_roc_auc": 0.8469713476749664, | |
| "eval_runtime": 509.5723, | |
| "eval_samples_per_second": 5.664, | |
| "eval_steps_per_second": 0.179, | |
| "learning_rate": 0.001, | |
| "step": 1365 | |
| }, | |
| { | |
| "epoch": 5.4945054945054945, | |
| "grad_norm": 0.17697261273860931, | |
| "learning_rate": 0.001, | |
| "loss": 0.1625, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.2480942480942481, | |
| "eval_f1_macro": 0.6092591780781843, | |
| "eval_f1_micro": 0.7798061948433986, | |
| "eval_loss": 0.14499613642692566, | |
| "eval_roc_auc": 0.8476621294180898, | |
| "eval_runtime": 511.5766, | |
| "eval_samples_per_second": 5.641, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 0.001, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.2525987525987526, | |
| "eval_f1_macro": 0.624806622732382, | |
| "eval_f1_micro": 0.7767369242779079, | |
| "eval_loss": 0.1474585235118866, | |
| "eval_roc_auc": 0.8453623673245133, | |
| "eval_runtime": 510.6821, | |
| "eval_samples_per_second": 5.651, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 0.001, | |
| "step": 1911 | |
| }, | |
| { | |
| "epoch": 7.326007326007326, | |
| "grad_norm": 0.24790136516094208, | |
| "learning_rate": 0.001, | |
| "loss": 0.1592, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.25744975744975745, | |
| "eval_f1_macro": 0.6249401475720361, | |
| "eval_f1_micro": 0.7803859753759638, | |
| "eval_loss": 0.14568069577217102, | |
| "eval_roc_auc": 0.8520784887308331, | |
| "eval_runtime": 514.1071, | |
| "eval_samples_per_second": 5.614, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.001, | |
| "step": 2184 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.25744975744975745, | |
| "eval_f1_macro": 0.652642904607388, | |
| "eval_f1_micro": 0.7868685150535805, | |
| "eval_loss": 0.14169421792030334, | |
| "eval_roc_auc": 0.8560913219420118, | |
| "eval_runtime": 518.322, | |
| "eval_samples_per_second": 5.568, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 0.001, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 9.157509157509157, | |
| "grad_norm": 0.2022881656885147, | |
| "learning_rate": 0.001, | |
| "loss": 0.157, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.25467775467775466, | |
| "eval_f1_macro": 0.6289931868767601, | |
| "eval_f1_micro": 0.7757335098168984, | |
| "eval_loss": 0.1436299830675125, | |
| "eval_roc_auc": 0.8403493908543921, | |
| "eval_runtime": 514.5559, | |
| "eval_samples_per_second": 5.609, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.001, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 10.989010989010989, | |
| "grad_norm": 0.16650938987731934, | |
| "learning_rate": 0.001, | |
| "loss": 0.1563, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.26403326403326405, | |
| "eval_f1_macro": 0.6447870111639475, | |
| "eval_f1_micro": 0.7886988341417751, | |
| "eval_loss": 0.1428152322769165, | |
| "eval_roc_auc": 0.8569209092596786, | |
| "eval_runtime": 515.3716, | |
| "eval_samples_per_second": 5.6, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.001, | |
| "step": 3003 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.25814275814275817, | |
| "eval_f1_macro": 0.6493205009564239, | |
| "eval_f1_micro": 0.7904845227679873, | |
| "eval_loss": 0.1438700556755066, | |
| "eval_roc_auc": 0.8637698939454586, | |
| "eval_runtime": 512.0567, | |
| "eval_samples_per_second": 5.636, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 0.001, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 12.820512820512821, | |
| "grad_norm": 0.163461372256279, | |
| "learning_rate": 0.001, | |
| "loss": 0.1558, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.2713097713097713, | |
| "eval_f1_macro": 0.6561811626743236, | |
| "eval_f1_micro": 0.7906956746065871, | |
| "eval_loss": 0.13913600146770477, | |
| "eval_roc_auc": 0.8551388511813229, | |
| "eval_runtime": 511.5092, | |
| "eval_samples_per_second": 5.642, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 0.001, | |
| "step": 3549 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.2643797643797644, | |
| "eval_f1_macro": 0.6337626365639194, | |
| "eval_f1_micro": 0.783810807286006, | |
| "eval_loss": 0.14094506204128265, | |
| "eval_roc_auc": 0.8484895839481307, | |
| "eval_runtime": 513.4311, | |
| "eval_samples_per_second": 5.621, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.001, | |
| "step": 3822 | |
| }, | |
| { | |
| "epoch": 14.652014652014651, | |
| "grad_norm": 0.17725127935409546, | |
| "learning_rate": 0.001, | |
| "loss": 0.1543, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.2577962577962578, | |
| "eval_f1_macro": 0.6463067634895379, | |
| "eval_f1_micro": 0.7907172995780591, | |
| "eval_loss": 0.1396123319864273, | |
| "eval_roc_auc": 0.8603407738558333, | |
| "eval_runtime": 520.2063, | |
| "eval_samples_per_second": 5.548, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.001, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.2654192654192654, | |
| "eval_f1_macro": 0.6593840515969085, | |
| "eval_f1_micro": 0.7913274487959551, | |
| "eval_loss": 0.13904806971549988, | |
| "eval_roc_auc": 0.856418510343081, | |
| "eval_runtime": 522.3782, | |
| "eval_samples_per_second": 5.525, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.001, | |
| "step": 4368 | |
| }, | |
| { | |
| "epoch": 16.483516483516482, | |
| "grad_norm": 0.16505596041679382, | |
| "learning_rate": 0.001, | |
| "loss": 0.1535, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.2564102564102564, | |
| "eval_f1_macro": 0.6585824628325464, | |
| "eval_f1_micro": 0.7939832128313804, | |
| "eval_loss": 0.1418265849351883, | |
| "eval_roc_auc": 0.8664525383660324, | |
| "eval_runtime": 520.8828, | |
| "eval_samples_per_second": 5.541, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.001, | |
| "step": 4641 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.26576576576576577, | |
| "eval_f1_macro": 0.6560187518750095, | |
| "eval_f1_micro": 0.7957187827911858, | |
| "eval_loss": 0.14155420660972595, | |
| "eval_roc_auc": 0.864558649781785, | |
| "eval_runtime": 521.9656, | |
| "eval_samples_per_second": 5.529, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.001, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 18.315018315018314, | |
| "grad_norm": 0.176731139421463, | |
| "learning_rate": 0.001, | |
| "loss": 0.1549, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.262993762993763, | |
| "eval_f1_macro": 0.6524018082903621, | |
| "eval_f1_micro": 0.7885625699767461, | |
| "eval_loss": 0.14027266204357147, | |
| "eval_roc_auc": 0.8535729424099051, | |
| "eval_runtime": 525.594, | |
| "eval_samples_per_second": 5.491, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.001, | |
| "step": 5187 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.26126126126126126, | |
| "eval_f1_macro": 0.6558190248610255, | |
| "eval_f1_micro": 0.7910696719558615, | |
| "eval_loss": 0.14759798347949982, | |
| "eval_roc_auc": 0.8567849608157283, | |
| "eval_runtime": 533.1376, | |
| "eval_samples_per_second": 5.413, | |
| "eval_steps_per_second": 0.171, | |
| "learning_rate": 0.001, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 20.146520146520146, | |
| "grad_norm": 0.15767891705036163, | |
| "learning_rate": 0.001, | |
| "loss": 0.154, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.26576576576576577, | |
| "eval_f1_macro": 0.6397027546064713, | |
| "eval_f1_micro": 0.7879767016708474, | |
| "eval_loss": 0.14285211265087128, | |
| "eval_roc_auc": 0.8567511447301636, | |
| "eval_runtime": 527.0011, | |
| "eval_samples_per_second": 5.476, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.001, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 21.978021978021978, | |
| "grad_norm": 0.18300685286521912, | |
| "learning_rate": 0.001, | |
| "loss": 0.1529, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.26126126126126126, | |
| "eval_f1_macro": 0.650810186340724, | |
| "eval_f1_micro": 0.7936799099512236, | |
| "eval_loss": 0.141402930021286, | |
| "eval_roc_auc": 0.8653510005054305, | |
| "eval_runtime": 525.9127, | |
| "eval_samples_per_second": 5.488, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.001, | |
| "step": 6006 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.26853776853776856, | |
| "eval_f1_macro": 0.6618136826297922, | |
| "eval_f1_micro": 0.7975794766896787, | |
| "eval_loss": 0.1415141373872757, | |
| "eval_roc_auc": 0.8613092204030781, | |
| "eval_runtime": 530.5247, | |
| "eval_samples_per_second": 5.44, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 0.0001, | |
| "step": 6279 | |
| }, | |
| { | |
| "epoch": 23.80952380952381, | |
| "grad_norm": 0.16848017275333405, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1449, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.27893277893277896, | |
| "eval_f1_macro": 0.6750686264509598, | |
| "eval_f1_micro": 0.8044778018063861, | |
| "eval_loss": 0.13230843842029572, | |
| "eval_roc_auc": 0.8664561198395929, | |
| "eval_runtime": 521.5756, | |
| "eval_samples_per_second": 5.533, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 6552 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.27927927927927926, | |
| "eval_f1_macro": 0.6724022117445357, | |
| "eval_f1_micro": 0.8044072500946213, | |
| "eval_loss": 0.13101588189601898, | |
| "eval_roc_auc": 0.868781233937024, | |
| "eval_runtime": 523.3306, | |
| "eval_samples_per_second": 5.515, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 6825 | |
| }, | |
| { | |
| "epoch": 25.641025641025642, | |
| "grad_norm": 0.16336454451084137, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1416, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.28205128205128205, | |
| "eval_f1_macro": 0.6689442300740391, | |
| "eval_f1_micro": 0.8035965398218775, | |
| "eval_loss": 0.13268393278121948, | |
| "eval_roc_auc": 0.8645798435204571, | |
| "eval_runtime": 532.8406, | |
| "eval_samples_per_second": 5.416, | |
| "eval_steps_per_second": 0.171, | |
| "learning_rate": 0.0001, | |
| "step": 7098 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.2817047817047817, | |
| "eval_f1_macro": 0.679681812643572, | |
| "eval_f1_micro": 0.8068647969861867, | |
| "eval_loss": 0.1317097693681717, | |
| "eval_roc_auc": 0.8714747032608311, | |
| "eval_runtime": 527.4278, | |
| "eval_samples_per_second": 5.472, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 7371 | |
| }, | |
| { | |
| "epoch": 27.47252747252747, | |
| "grad_norm": 0.1572931855916977, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1391, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.27754677754677753, | |
| "eval_f1_macro": 0.6818462300001074, | |
| "eval_f1_micro": 0.8072126727334008, | |
| "eval_loss": 0.12880520522594452, | |
| "eval_roc_auc": 0.8697994857701482, | |
| "eval_runtime": 536.9046, | |
| "eval_samples_per_second": 5.375, | |
| "eval_steps_per_second": 0.169, | |
| "learning_rate": 0.0001, | |
| "step": 7644 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.2844767844767845, | |
| "eval_f1_macro": 0.6807929806344717, | |
| "eval_f1_micro": 0.8038088702067427, | |
| "eval_loss": 0.12942521274089813, | |
| "eval_roc_auc": 0.8628519636133017, | |
| "eval_runtime": 520.5065, | |
| "eval_samples_per_second": 5.545, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 7917 | |
| }, | |
| { | |
| "epoch": 29.304029304029303, | |
| "grad_norm": 0.19199338555335999, | |
| "learning_rate": 0.0001, | |
| "loss": 0.138, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.28586278586278585, | |
| "eval_f1_macro": 0.6825529208005033, | |
| "eval_f1_micro": 0.8077149835761811, | |
| "eval_loss": 0.12943296134471893, | |
| "eval_roc_auc": 0.8701959964759374, | |
| "eval_runtime": 543.5755, | |
| "eval_samples_per_second": 5.309, | |
| "eval_steps_per_second": 0.167, | |
| "learning_rate": 0.0001, | |
| "step": 8190 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.28794178794178793, | |
| "eval_f1_macro": 0.6779122940127521, | |
| "eval_f1_micro": 0.8073808915025994, | |
| "eval_loss": 0.12738928198814392, | |
| "eval_roc_auc": 0.8666172459085354, | |
| "eval_runtime": 521.4164, | |
| "eval_samples_per_second": 5.535, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 8463 | |
| }, | |
| { | |
| "epoch": 31.135531135531135, | |
| "grad_norm": 0.1997932642698288, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1364, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.2882882882882883, | |
| "eval_f1_macro": 0.6868638344898197, | |
| "eval_f1_micro": 0.8104185890445432, | |
| "eval_loss": 0.12775012850761414, | |
| "eval_roc_auc": 0.8728485806633693, | |
| "eval_runtime": 519.8308, | |
| "eval_samples_per_second": 5.552, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 8736 | |
| }, | |
| { | |
| "epoch": 32.967032967032964, | |
| "grad_norm": 0.19476589560508728, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1359, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.2869022869022869, | |
| "eval_f1_macro": 0.6810807224403135, | |
| "eval_f1_micro": 0.8077248140635565, | |
| "eval_loss": 0.12765593826770782, | |
| "eval_roc_auc": 0.8692062891212271, | |
| "eval_runtime": 514.7142, | |
| "eval_samples_per_second": 5.607, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.0001, | |
| "step": 9009 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.2882882882882883, | |
| "eval_f1_macro": 0.687361527737602, | |
| "eval_f1_micro": 0.8108837797932926, | |
| "eval_loss": 0.12660712003707886, | |
| "eval_roc_auc": 0.8714320206807965, | |
| "eval_runtime": 514.9645, | |
| "eval_samples_per_second": 5.604, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.0001, | |
| "step": 9282 | |
| }, | |
| { | |
| "epoch": 34.798534798534796, | |
| "grad_norm": 0.2034957855939865, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1341, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.29036729036729036, | |
| "eval_f1_macro": 0.688483181989703, | |
| "eval_f1_micro": 0.8103963941193815, | |
| "eval_loss": 0.1262102574110031, | |
| "eval_roc_auc": 0.8715800817488106, | |
| "eval_runtime": 525.0872, | |
| "eval_samples_per_second": 5.496, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 9555 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.28274428274428276, | |
| "eval_f1_macro": 0.6876394944988364, | |
| "eval_f1_micro": 0.8070400273399119, | |
| "eval_loss": 0.12687553465366364, | |
| "eval_roc_auc": 0.8657418371913091, | |
| "eval_runtime": 513.0757, | |
| "eval_samples_per_second": 5.625, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.0001, | |
| "step": 9828 | |
| }, | |
| { | |
| "epoch": 36.63003663003663, | |
| "grad_norm": 0.20557202398777008, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1339, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.28655578655578656, | |
| "eval_f1_macro": 0.6833930255395054, | |
| "eval_f1_micro": 0.8081597960050999, | |
| "eval_loss": 0.12656189501285553, | |
| "eval_roc_auc": 0.8678163688633396, | |
| "eval_runtime": 515.3436, | |
| "eval_samples_per_second": 5.6, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.0001, | |
| "step": 10101 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.2955647955647956, | |
| "eval_f1_macro": 0.6936175483283518, | |
| "eval_f1_micro": 0.8106371284826448, | |
| "eval_loss": 0.12547720968723297, | |
| "eval_roc_auc": 0.8706625538294134, | |
| "eval_runtime": 512.1358, | |
| "eval_samples_per_second": 5.635, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 0.0001, | |
| "step": 10374 | |
| }, | |
| { | |
| "epoch": 38.46153846153846, | |
| "grad_norm": 0.2112371176481247, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1307, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.2927927927927928, | |
| "eval_f1_macro": 0.6985657340894045, | |
| "eval_f1_micro": 0.8141880626875626, | |
| "eval_loss": 0.12485096603631973, | |
| "eval_roc_auc": 0.8767653445350737, | |
| "eval_runtime": 512.6109, | |
| "eval_samples_per_second": 5.63, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 0.0001, | |
| "step": 10647 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.2934857934857935, | |
| "eval_f1_macro": 0.6989554260935754, | |
| "eval_f1_micro": 0.8138017044273539, | |
| "eval_loss": 0.1257668137550354, | |
| "eval_roc_auc": 0.8773247787534647, | |
| "eval_runtime": 513.8833, | |
| "eval_samples_per_second": 5.616, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.0001, | |
| "step": 10920 | |
| }, | |
| { | |
| "epoch": 40.29304029304029, | |
| "grad_norm": 0.23032954335212708, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1317, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.29244629244629244, | |
| "eval_f1_macro": 0.6923923602014324, | |
| "eval_f1_micro": 0.8101351925856646, | |
| "eval_loss": 0.12528541684150696, | |
| "eval_roc_auc": 0.8687915491174283, | |
| "eval_runtime": 513.0005, | |
| "eval_samples_per_second": 5.626, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.0001, | |
| "step": 11193 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.3004158004158004, | |
| "eval_f1_macro": 0.6970236383039276, | |
| "eval_f1_micro": 0.8138018093835474, | |
| "eval_loss": 0.12443084269762039, | |
| "eval_roc_auc": 0.8737649281720051, | |
| "eval_runtime": 525.5315, | |
| "eval_samples_per_second": 5.492, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 11466 | |
| }, | |
| { | |
| "epoch": 42.124542124542124, | |
| "grad_norm": 0.23487386107444763, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1308, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.2948717948717949, | |
| "eval_f1_macro": 0.6956334056896907, | |
| "eval_f1_micro": 0.8131470414948238, | |
| "eval_loss": 0.12451612949371338, | |
| "eval_roc_auc": 0.8733690344991142, | |
| "eval_runtime": 514.3778, | |
| "eval_samples_per_second": 5.611, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.0001, | |
| "step": 11739 | |
| }, | |
| { | |
| "epoch": 43.956043956043956, | |
| "grad_norm": 0.25621357560157776, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1307, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.2966042966042966, | |
| "eval_f1_macro": 0.6915470420512126, | |
| "eval_f1_micro": 0.812950847173293, | |
| "eval_loss": 0.12501148879528046, | |
| "eval_roc_auc": 0.8742664283667729, | |
| "eval_runtime": 519.3764, | |
| "eval_samples_per_second": 5.557, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 12012 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.29625779625779625, | |
| "eval_f1_macro": 0.7050548840380568, | |
| "eval_f1_micro": 0.8136846971798428, | |
| "eval_loss": 0.12397606670856476, | |
| "eval_roc_auc": 0.8740443367647517, | |
| "eval_runtime": 515.8997, | |
| "eval_samples_per_second": 5.594, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 0.0001, | |
| "step": 12285 | |
| }, | |
| { | |
| "epoch": 45.78754578754579, | |
| "grad_norm": 0.22914335131645203, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1295, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.29764379764379767, | |
| "eval_f1_macro": 0.6987723620069867, | |
| "eval_f1_micro": 0.8130628734954971, | |
| "eval_loss": 0.12409698963165283, | |
| "eval_roc_auc": 0.8733228777555885, | |
| "eval_runtime": 516.6269, | |
| "eval_samples_per_second": 5.586, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 0.0001, | |
| "step": 12558 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.2955647955647956, | |
| "eval_f1_macro": 0.6957628076563835, | |
| "eval_f1_micro": 0.811911298838437, | |
| "eval_loss": 0.12429661303758621, | |
| "eval_roc_auc": 0.8716271908692008, | |
| "eval_runtime": 518.0917, | |
| "eval_samples_per_second": 5.57, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 0.0001, | |
| "step": 12831 | |
| }, | |
| { | |
| "epoch": 47.61904761904762, | |
| "grad_norm": 0.25639113783836365, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1293, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.2955647955647956, | |
| "eval_f1_macro": 0.6990296569974817, | |
| "eval_f1_micro": 0.8135280295401142, | |
| "eval_loss": 0.12393072247505188, | |
| "eval_roc_auc": 0.874436809929186, | |
| "eval_runtime": 517.4634, | |
| "eval_samples_per_second": 5.577, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 0.0001, | |
| "step": 13104 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.29972279972279975, | |
| "eval_f1_macro": 0.7007060102949784, | |
| "eval_f1_micro": 0.8152993625265614, | |
| "eval_loss": 0.1242954283952713, | |
| "eval_roc_auc": 0.8774914581184896, | |
| "eval_runtime": 511.3749, | |
| "eval_samples_per_second": 5.644, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 0.0001, | |
| "step": 13377 | |
| }, | |
| { | |
| "epoch": 49.45054945054945, | |
| "grad_norm": 0.27197974920272827, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1274, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.29799029799029797, | |
| "eval_f1_macro": 0.6999734070385492, | |
| "eval_f1_micro": 0.8151919866444074, | |
| "eval_loss": 0.12405084818601608, | |
| "eval_roc_auc": 0.8769273693258459, | |
| "eval_runtime": 509.3276, | |
| "eval_samples_per_second": 5.666, | |
| "eval_steps_per_second": 0.179, | |
| "learning_rate": 0.0001, | |
| "step": 13650 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.3011088011088011, | |
| "eval_f1_macro": 0.7055935576453343, | |
| "eval_f1_micro": 0.8153039745759215, | |
| "eval_loss": 0.12483017891645432, | |
| "eval_roc_auc": 0.8803007418345086, | |
| "eval_runtime": 511.0056, | |
| "eval_samples_per_second": 5.648, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 0.0001, | |
| "step": 13923 | |
| }, | |
| { | |
| "epoch": 51.282051282051285, | |
| "grad_norm": 0.23091430962085724, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1271, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.3049203049203049, | |
| "eval_f1_macro": 0.7035566403965832, | |
| "eval_f1_micro": 0.8157241959217996, | |
| "eval_loss": 0.12426182627677917, | |
| "eval_roc_auc": 0.8750656737623661, | |
| "eval_runtime": 511.1647, | |
| "eval_samples_per_second": 5.646, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 0.0001, | |
| "step": 14196 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.30214830214830213, | |
| "eval_f1_macro": 0.7031528349086803, | |
| "eval_f1_micro": 0.8152648882600192, | |
| "eval_loss": 0.12408608943223953, | |
| "eval_roc_auc": 0.8778170234547618, | |
| "eval_runtime": 520.4, | |
| "eval_samples_per_second": 5.546, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 14469 | |
| }, | |
| { | |
| "epoch": 53.11355311355312, | |
| "grad_norm": 0.23177389800548553, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1275, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.30214830214830213, | |
| "eval_f1_macro": 0.7067666695453366, | |
| "eval_f1_micro": 0.8152251458307105, | |
| "eval_loss": 0.12344320118427277, | |
| "eval_roc_auc": 0.8753333050750151, | |
| "eval_runtime": 522.8329, | |
| "eval_samples_per_second": 5.52, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 14742 | |
| }, | |
| { | |
| "epoch": 54.94505494505494, | |
| "grad_norm": 0.3403611481189728, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1256, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.30180180180180183, | |
| "eval_f1_macro": 0.7075536762185066, | |
| "eval_f1_micro": 0.8166332665330662, | |
| "eval_loss": 0.12307523190975189, | |
| "eval_roc_auc": 0.8776256091187804, | |
| "eval_runtime": 513.5394, | |
| "eval_samples_per_second": 5.62, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.0001, | |
| "step": 15015 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.30665280665280664, | |
| "eval_f1_macro": 0.7087921855865761, | |
| "eval_f1_micro": 0.8189626693095475, | |
| "eval_loss": 0.12282071262598038, | |
| "eval_roc_auc": 0.8821854285803199, | |
| "eval_runtime": 519.2592, | |
| "eval_samples_per_second": 5.558, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 15288 | |
| }, | |
| { | |
| "epoch": 56.776556776556774, | |
| "grad_norm": 0.28649473190307617, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1258, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.306999306999307, | |
| "eval_f1_macro": 0.7079839879234633, | |
| "eval_f1_micro": 0.8160328019748128, | |
| "eval_loss": 0.12259934842586517, | |
| "eval_roc_auc": 0.8766650096203477, | |
| "eval_runtime": 523.8952, | |
| "eval_samples_per_second": 5.509, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 15561 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.30214830214830213, | |
| "eval_f1_macro": 0.7072503847729165, | |
| "eval_f1_micro": 0.8170145133631687, | |
| "eval_loss": 0.12334763258695602, | |
| "eval_roc_auc": 0.8773053153896588, | |
| "eval_runtime": 522.7463, | |
| "eval_samples_per_second": 5.521, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 15834 | |
| }, | |
| { | |
| "epoch": 58.608058608058606, | |
| "grad_norm": 0.2677023112773895, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1258, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.30214830214830213, | |
| "eval_f1_macro": 0.713532815646164, | |
| "eval_f1_micro": 0.8172105834237543, | |
| "eval_loss": 0.12272054702043533, | |
| "eval_roc_auc": 0.8780682765680952, | |
| "eval_runtime": 524.8476, | |
| "eval_samples_per_second": 5.499, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 16107 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.30214830214830213, | |
| "eval_f1_macro": 0.7039801220819605, | |
| "eval_f1_micro": 0.8142579609764339, | |
| "eval_loss": 0.12334387749433517, | |
| "eval_roc_auc": 0.8729462194126062, | |
| "eval_runtime": 526.97, | |
| "eval_samples_per_second": 5.477, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 16380 | |
| }, | |
| { | |
| "epoch": 60.43956043956044, | |
| "grad_norm": 0.273879736661911, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1252, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.3042273042273042, | |
| "eval_f1_macro": 0.7120578542808926, | |
| "eval_f1_micro": 0.816814564846061, | |
| "eval_loss": 0.12339764833450317, | |
| "eval_roc_auc": 0.8783554248995846, | |
| "eval_runtime": 524.4656, | |
| "eval_samples_per_second": 5.503, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 16653 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.3049203049203049, | |
| "eval_f1_macro": 0.7124854785684515, | |
| "eval_f1_micro": 0.8169309505831026, | |
| "eval_loss": 0.12234435975551605, | |
| "eval_roc_auc": 0.876382515863111, | |
| "eval_runtime": 518.389, | |
| "eval_samples_per_second": 5.567, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 0.0001, | |
| "step": 16926 | |
| }, | |
| { | |
| "epoch": 62.27106227106227, | |
| "grad_norm": 0.2836596667766571, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1238, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.30353430353430355, | |
| "eval_f1_macro": 0.709030237195192, | |
| "eval_f1_micro": 0.8151443922095366, | |
| "eval_loss": 0.12311259657144547, | |
| "eval_roc_auc": 0.875227363209172, | |
| "eval_runtime": 523.1406, | |
| "eval_samples_per_second": 5.517, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 17199 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.30665280665280664, | |
| "eval_f1_macro": 0.7114197657112039, | |
| "eval_f1_micro": 0.8183222681531587, | |
| "eval_loss": 0.12282687425613403, | |
| "eval_roc_auc": 0.8785221042646094, | |
| "eval_runtime": 525.9879, | |
| "eval_samples_per_second": 5.487, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 17472 | |
| }, | |
| { | |
| "epoch": 64.1025641025641, | |
| "grad_norm": 0.327009916305542, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1247, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.30353430353430355, | |
| "eval_f1_macro": 0.715610525327271, | |
| "eval_f1_micro": 0.8185065204751224, | |
| "eval_loss": 0.12305620312690735, | |
| "eval_roc_auc": 0.8802214933483853, | |
| "eval_runtime": 527.6963, | |
| "eval_samples_per_second": 5.469, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 0.0001, | |
| "step": 17745 | |
| }, | |
| { | |
| "epoch": 65.93406593406593, | |
| "grad_norm": 0.3439556360244751, | |
| "learning_rate": 0.0001, | |
| "loss": 0.123, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.30214830214830213, | |
| "eval_f1_macro": 0.7083957677770276, | |
| "eval_f1_micro": 0.8193021036471515, | |
| "eval_loss": 0.12252139300107956, | |
| "eval_roc_auc": 0.8809488409975973, | |
| "eval_runtime": 523.6027, | |
| "eval_samples_per_second": 5.512, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 18018 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.3031878031878032, | |
| "eval_f1_macro": 0.713563304331985, | |
| "eval_f1_micro": 0.8185542268382505, | |
| "eval_loss": 0.12215397506952286, | |
| "eval_roc_auc": 0.8813502879665707, | |
| "eval_runtime": 528.5406, | |
| "eval_samples_per_second": 5.46, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 0.0001, | |
| "step": 18291 | |
| }, | |
| { | |
| "epoch": 67.76556776556777, | |
| "grad_norm": 0.3434881269931793, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1224, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.3090783090783091, | |
| "eval_f1_macro": 0.7169216330412181, | |
| "eval_f1_micro": 0.8201218248870841, | |
| "eval_loss": 0.12200037389993668, | |
| "eval_roc_auc": 0.8818022645643908, | |
| "eval_runtime": 525.6971, | |
| "eval_samples_per_second": 5.49, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 18564 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.30180180180180183, | |
| "eval_f1_macro": 0.7165157275423649, | |
| "eval_f1_micro": 0.8171493231633209, | |
| "eval_loss": 0.12282921373844147, | |
| "eval_roc_auc": 0.8767867663076429, | |
| "eval_runtime": 539.1574, | |
| "eval_samples_per_second": 5.353, | |
| "eval_steps_per_second": 0.169, | |
| "learning_rate": 0.0001, | |
| "step": 18837 | |
| }, | |
| { | |
| "epoch": 69.59706959706959, | |
| "grad_norm": 0.2773456275463104, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1228, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.3042273042273042, | |
| "eval_f1_macro": 0.7130922408537738, | |
| "eval_f1_micro": 0.8176893032631977, | |
| "eval_loss": 0.12265007942914963, | |
| "eval_roc_auc": 0.8764658555456234, | |
| "eval_runtime": 532.0042, | |
| "eval_samples_per_second": 5.425, | |
| "eval_steps_per_second": 0.171, | |
| "learning_rate": 0.0001, | |
| "step": 19110 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.29799029799029797, | |
| "eval_f1_macro": 0.7123118599173115, | |
| "eval_f1_micro": 0.8155257705805251, | |
| "eval_loss": 0.12318737804889679, | |
| "eval_roc_auc": 0.8733064995562728, | |
| "eval_runtime": 512.5227, | |
| "eval_samples_per_second": 5.631, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 0.0001, | |
| "step": 19383 | |
| }, | |
| { | |
| "epoch": 71.42857142857143, | |
| "grad_norm": 0.32921841740608215, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1224, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.30561330561330563, | |
| "eval_f1_macro": 0.7181217472368024, | |
| "eval_f1_micro": 0.8177146438270315, | |
| "eval_loss": 0.12224896252155304, | |
| "eval_roc_auc": 0.8780131460200304, | |
| "eval_runtime": 526.8353, | |
| "eval_samples_per_second": 5.478, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 19656 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.3076923076923077, | |
| "eval_f1_macro": 0.7046690012290543, | |
| "eval_f1_micro": 0.8161570403926011, | |
| "eval_loss": 0.12214501202106476, | |
| "eval_roc_auc": 0.8759937448960649, | |
| "eval_runtime": 523.4325, | |
| "eval_samples_per_second": 5.514, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 19929 | |
| }, | |
| { | |
| "epoch": 73.26007326007326, | |
| "grad_norm": 0.27500712871551514, | |
| "learning_rate": 0.0001, | |
| "loss": 0.122, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.2972972972972973, | |
| "eval_f1_macro": 0.7070482653980339, | |
| "eval_f1_micro": 0.8147835269271382, | |
| "eval_loss": 0.12297073751688004, | |
| "eval_roc_auc": 0.8731965201490751, | |
| "eval_runtime": 521.6748, | |
| "eval_samples_per_second": 5.532, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 20202 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.3049203049203049, | |
| "eval_f1_macro": 0.7123584497861349, | |
| "eval_f1_micro": 0.8175831550689987, | |
| "eval_loss": 0.12141965329647064, | |
| "eval_roc_auc": 0.876778409536002, | |
| "eval_runtime": 521.9582, | |
| "eval_samples_per_second": 5.529, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 1e-05, | |
| "step": 20475 | |
| }, | |
| { | |
| "epoch": 75.0915750915751, | |
| "grad_norm": 0.34586936235427856, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1201, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.30665280665280664, | |
| "eval_f1_macro": 0.7265282519195887, | |
| "eval_f1_micro": 0.8212704324436167, | |
| "eval_loss": 0.12091591954231262, | |
| "eval_roc_auc": 0.8828403151052873, | |
| "eval_runtime": 515.687, | |
| "eval_samples_per_second": 5.596, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 1e-05, | |
| "step": 20748 | |
| }, | |
| { | |
| "epoch": 76.92307692307692, | |
| "grad_norm": 0.3650946617126465, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1192, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.30734580734580735, | |
| "eval_f1_macro": 0.7249141687532618, | |
| "eval_f1_micro": 0.8221009885557243, | |
| "eval_loss": 0.12162773311138153, | |
| "eval_roc_auc": 0.88597146196019, | |
| "eval_runtime": 505.8066, | |
| "eval_samples_per_second": 5.706, | |
| "eval_steps_per_second": 0.18, | |
| "learning_rate": 1e-05, | |
| "step": 21021 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.30561330561330563, | |
| "eval_f1_macro": 0.7232913822219021, | |
| "eval_f1_micro": 0.821013443640124, | |
| "eval_loss": 0.12114103883504868, | |
| "eval_roc_auc": 0.8828214151193448, | |
| "eval_runtime": 515.423, | |
| "eval_samples_per_second": 5.599, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 1e-05, | |
| "step": 21294 | |
| }, | |
| { | |
| "epoch": 78.75457875457876, | |
| "grad_norm": 0.3805921673774719, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1178, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.30561330561330563, | |
| "eval_f1_macro": 0.7157592534107864, | |
| "eval_f1_micro": 0.8181284095677717, | |
| "eval_loss": 0.1210767850279808, | |
| "eval_roc_auc": 0.8769422854254683, | |
| "eval_runtime": 524.7026, | |
| "eval_samples_per_second": 5.5, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 1e-05, | |
| "step": 21567 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.3090783090783091, | |
| "eval_f1_macro": 0.7196736600383237, | |
| "eval_f1_micro": 0.8200463116109824, | |
| "eval_loss": 0.12099559605121613, | |
| "eval_roc_auc": 0.8823936101146178, | |
| "eval_runtime": 518.5996, | |
| "eval_samples_per_second": 5.565, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 1e-05, | |
| "step": 21840 | |
| }, | |
| { | |
| "epoch": 80.58608058608058, | |
| "grad_norm": 0.38496658205986023, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1178, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.31046431046431044, | |
| "eval_f1_macro": 0.7194056763702963, | |
| "eval_f1_micro": 0.8189727287937092, | |
| "eval_loss": 0.12053155153989792, | |
| "eval_roc_auc": 0.8783734261636972, | |
| "eval_runtime": 517.5249, | |
| "eval_samples_per_second": 5.577, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 1e-05, | |
| "step": 22113 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.306999306999307, | |
| "eval_f1_macro": 0.7212694332008583, | |
| "eval_f1_micro": 0.8186875235267054, | |
| "eval_loss": 0.12050338089466095, | |
| "eval_roc_auc": 0.8782284502601733, | |
| "eval_runtime": 511.5713, | |
| "eval_samples_per_second": 5.641, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 1e-05, | |
| "step": 22386 | |
| }, | |
| { | |
| "epoch": 82.41758241758242, | |
| "grad_norm": 0.29807013273239136, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1162, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.3049203049203049, | |
| "eval_f1_macro": 0.7136069207682542, | |
| "eval_f1_micro": 0.817129142279675, | |
| "eval_loss": 0.12153622508049011, | |
| "eval_roc_auc": 0.8753921914755026, | |
| "eval_runtime": 514.5554, | |
| "eval_samples_per_second": 5.609, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 1e-05, | |
| "step": 22659 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.3115038115038115, | |
| "eval_f1_macro": 0.72263281374496, | |
| "eval_f1_micro": 0.8212135055442501, | |
| "eval_loss": 0.12091034650802612, | |
| "eval_roc_auc": 0.8817381602117871, | |
| "eval_runtime": 514.2801, | |
| "eval_samples_per_second": 5.612, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 1e-05, | |
| "step": 22932 | |
| }, | |
| { | |
| "epoch": 84.24908424908425, | |
| "grad_norm": 0.4926730692386627, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1174, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.30942480942480943, | |
| "eval_f1_macro": 0.7219026145386024, | |
| "eval_f1_micro": 0.8212908842183808, | |
| "eval_loss": 0.12058679759502411, | |
| "eval_roc_auc": 0.8823288887291161, | |
| "eval_runtime": 513.2258, | |
| "eval_samples_per_second": 5.623, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 1e-05, | |
| "step": 23205 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.30838530838530837, | |
| "eval_f1_macro": 0.7255503995321377, | |
| "eval_f1_micro": 0.8206727371003285, | |
| "eval_loss": 0.1210218220949173, | |
| "eval_roc_auc": 0.8810894976708349, | |
| "eval_runtime": 516.6336, | |
| "eval_samples_per_second": 5.586, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 1e-05, | |
| "step": 23478 | |
| }, | |
| { | |
| "epoch": 86.08058608058609, | |
| "grad_norm": 0.3941400647163391, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1167, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.30734580734580735, | |
| "eval_f1_macro": 0.7163464112504625, | |
| "eval_f1_micro": 0.81919187715867, | |
| "eval_loss": 0.12097787857055664, | |
| "eval_roc_auc": 0.8800195357981024, | |
| "eval_runtime": 514.309, | |
| "eval_samples_per_second": 5.611, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 1e-05, | |
| "step": 23751 | |
| }, | |
| { | |
| "epoch": 87.91208791208791, | |
| "grad_norm": 0.30446189641952515, | |
| "learning_rate": 1e-05, | |
| "loss": 0.116, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.30942480942480943, | |
| "eval_f1_macro": 0.7179611359738045, | |
| "eval_f1_micro": 0.8219223445649475, | |
| "eval_loss": 0.12078534066677094, | |
| "eval_roc_auc": 0.8831166385335435, | |
| "eval_runtime": 513.9044, | |
| "eval_samples_per_second": 5.616, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 1e-05, | |
| "step": 24024 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.3125433125433125, | |
| "eval_f1_macro": 0.7293063087262872, | |
| "eval_f1_micro": 0.8235824319895118, | |
| "eval_loss": 0.1213160827755928, | |
| "eval_roc_auc": 0.8871674997505042, | |
| "eval_runtime": 514.1086, | |
| "eval_samples_per_second": 5.614, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "step": 24297 | |
| }, | |
| { | |
| "epoch": 89.74358974358974, | |
| "grad_norm": 0.2996889054775238, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.1161, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.3108108108108108, | |
| "eval_f1_macro": 0.7249894355418997, | |
| "eval_f1_micro": 0.8228019165403988, | |
| "eval_loss": 0.12110408395528793, | |
| "eval_roc_auc": 0.8868651536304606, | |
| "eval_runtime": 511.9928, | |
| "eval_samples_per_second": 5.637, | |
| "eval_steps_per_second": 0.178, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "step": 24570 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.31046431046431044, | |
| "eval_f1_macro": 0.7187027508297176, | |
| "eval_f1_micro": 0.8191074795725959, | |
| "eval_loss": 0.1205781027674675, | |
| "eval_roc_auc": 0.8779146622039986, | |
| "eval_runtime": 513.1929, | |
| "eval_samples_per_second": 5.624, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "step": 24843 | |
| }, | |
| { | |
| "epoch": 91.57509157509158, | |
| "grad_norm": 0.30445897579193115, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.1162, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.31046431046431044, | |
| "eval_f1_macro": 0.7150284118631205, | |
| "eval_f1_micro": 0.8196009683612989, | |
| "eval_loss": 0.12076584249734879, | |
| "eval_roc_auc": 0.8793405313350767, | |
| "eval_runtime": 520.7671, | |
| "eval_samples_per_second": 5.542, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "step": 25116 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "step": 25116, | |
| "total_flos": 1.1890234809282512e+21, | |
| "train_loss": 0.1360613288991788, | |
| "train_runtime": 194834.2342, | |
| "train_samples_per_second": 6.71, | |
| "train_steps_per_second": 0.21 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 40950, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 150, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 10, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1890234809282512e+21, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |