| { | |
| "best_metric": 0.1244530975818634, | |
| "best_model_checkpoint": "/home1/datawork/mcontini/models/multilabel/huggingface/dinov2-large-2024_05_27-_batch-size32_epochs150_freeze/checkpoint-24570", | |
| "epoch": 100.0, | |
| "eval_steps": 500, | |
| "global_step": 27300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.232726022688209, | |
| "eval_f1_macro": 0.4712591871520079, | |
| "eval_f1_micro": 0.7380235658381353, | |
| "eval_loss": 0.1701383888721466, | |
| "eval_roc_auc": 0.8226166045759734, | |
| "eval_runtime": 579.6646, | |
| "eval_samples_per_second": 5.018, | |
| "eval_steps_per_second": 0.157, | |
| "learning_rate": 0.001, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 1.8315018315018317, | |
| "grad_norm": 0.3410358726978302, | |
| "learning_rate": 0.001, | |
| "loss": 0.2748, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.24888277758679958, | |
| "eval_f1_macro": 0.5722657636852799, | |
| "eval_f1_micro": 0.7568708574323469, | |
| "eval_loss": 0.15890291333198547, | |
| "eval_roc_auc": 0.8372263575023127, | |
| "eval_runtime": 572.4604, | |
| "eval_samples_per_second": 5.082, | |
| "eval_steps_per_second": 0.159, | |
| "learning_rate": 0.001, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.24785149535922998, | |
| "eval_f1_macro": 0.6104117366005594, | |
| "eval_f1_micro": 0.7723932964583505, | |
| "eval_loss": 0.15134122967720032, | |
| "eval_roc_auc": 0.851560035472562, | |
| "eval_runtime": 581.798, | |
| "eval_samples_per_second": 5.0, | |
| "eval_steps_per_second": 0.156, | |
| "learning_rate": 0.001, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 3.663003663003663, | |
| "grad_norm": 0.28447234630584717, | |
| "learning_rate": 0.001, | |
| "loss": 0.1714, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.24853901684427637, | |
| "eval_f1_macro": 0.599745200497783, | |
| "eval_f1_micro": 0.7608496532472631, | |
| "eval_loss": 0.15164224803447723, | |
| "eval_roc_auc": 0.8311391237980399, | |
| "eval_runtime": 574.55, | |
| "eval_samples_per_second": 5.063, | |
| "eval_steps_per_second": 0.158, | |
| "learning_rate": 0.001, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.24750773461670678, | |
| "eval_f1_macro": 0.5935106518877853, | |
| "eval_f1_micro": 0.7692371752165224, | |
| "eval_loss": 0.15243493020534515, | |
| "eval_roc_auc": 0.8475084947536117, | |
| "eval_runtime": 565.8483, | |
| "eval_samples_per_second": 5.141, | |
| "eval_steps_per_second": 0.161, | |
| "learning_rate": 0.001, | |
| "step": 1365 | |
| }, | |
| { | |
| "epoch": 5.4945054945054945, | |
| "grad_norm": 0.32554125785827637, | |
| "learning_rate": 0.001, | |
| "loss": 0.1661, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.24269508422138192, | |
| "eval_f1_macro": 0.613788061665736, | |
| "eval_f1_micro": 0.7718080548414739, | |
| "eval_loss": 0.14673969149589539, | |
| "eval_roc_auc": 0.8414237469351327, | |
| "eval_runtime": 563.6053, | |
| "eval_samples_per_second": 5.161, | |
| "eval_steps_per_second": 0.161, | |
| "learning_rate": 0.001, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.2509453420419388, | |
| "eval_f1_macro": 0.6309898748773293, | |
| "eval_f1_micro": 0.7732481363152289, | |
| "eval_loss": 0.1506606936454773, | |
| "eval_roc_auc": 0.8436874105670831, | |
| "eval_runtime": 563.2783, | |
| "eval_samples_per_second": 5.164, | |
| "eval_steps_per_second": 0.162, | |
| "learning_rate": 0.001, | |
| "step": 1911 | |
| }, | |
| { | |
| "epoch": 7.326007326007326, | |
| "grad_norm": 0.22656740248203278, | |
| "learning_rate": 0.001, | |
| "loss": 0.1611, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.2609144035751117, | |
| "eval_f1_macro": 0.6403740520777896, | |
| "eval_f1_micro": 0.7828014555188422, | |
| "eval_loss": 0.14430351555347443, | |
| "eval_roc_auc": 0.8536999365685817, | |
| "eval_runtime": 570.8881, | |
| "eval_samples_per_second": 5.096, | |
| "eval_steps_per_second": 0.159, | |
| "learning_rate": 0.001, | |
| "step": 2184 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.25128910278446204, | |
| "eval_f1_macro": 0.6366498775226099, | |
| "eval_f1_micro": 0.781416038551835, | |
| "eval_loss": 0.14617429673671722, | |
| "eval_roc_auc": 0.8571263621918263, | |
| "eval_runtime": 575.7471, | |
| "eval_samples_per_second": 5.053, | |
| "eval_steps_per_second": 0.158, | |
| "learning_rate": 0.001, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 9.157509157509157, | |
| "grad_norm": 0.1823957860469818, | |
| "learning_rate": 0.001, | |
| "loss": 0.1606, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.2688209006531454, | |
| "eval_f1_macro": 0.6254472467805158, | |
| "eval_f1_micro": 0.7794745970641737, | |
| "eval_loss": 0.14414818584918976, | |
| "eval_roc_auc": 0.8488544375964642, | |
| "eval_runtime": 577.67, | |
| "eval_samples_per_second": 5.036, | |
| "eval_steps_per_second": 0.158, | |
| "learning_rate": 0.001, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 10.989010989010989, | |
| "grad_norm": 0.1724083572626114, | |
| "learning_rate": 0.001, | |
| "loss": 0.1592, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.2595393606050189, | |
| "eval_f1_macro": 0.6357434835994208, | |
| "eval_f1_micro": 0.7780349253103302, | |
| "eval_loss": 0.14589238166809082, | |
| "eval_roc_auc": 0.8494945098415877, | |
| "eval_runtime": 576.9242, | |
| "eval_samples_per_second": 5.042, | |
| "eval_steps_per_second": 0.158, | |
| "learning_rate": 0.001, | |
| "step": 3003 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.2554142316947405, | |
| "eval_f1_macro": 0.638389910481932, | |
| "eval_f1_micro": 0.7823495795575149, | |
| "eval_loss": 0.14458976686000824, | |
| "eval_roc_auc": 0.8601197613283508, | |
| "eval_runtime": 579.3816, | |
| "eval_samples_per_second": 5.021, | |
| "eval_steps_per_second": 0.157, | |
| "learning_rate": 0.001, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 12.820512820512821, | |
| "grad_norm": 0.18148617446422577, | |
| "learning_rate": 0.001, | |
| "loss": 0.1582, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.2561017531797869, | |
| "eval_f1_macro": 0.6574365741219022, | |
| "eval_f1_micro": 0.786284091383703, | |
| "eval_loss": 0.14142437279224396, | |
| "eval_roc_auc": 0.8559800911193071, | |
| "eval_runtime": 580.1191, | |
| "eval_samples_per_second": 5.014, | |
| "eval_steps_per_second": 0.157, | |
| "learning_rate": 0.001, | |
| "step": 3549 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.24682021313166036, | |
| "eval_f1_macro": 0.6245865910731833, | |
| "eval_f1_micro": 0.7766990291262137, | |
| "eval_loss": 0.1581379920244217, | |
| "eval_roc_auc": 0.8490862605596214, | |
| "eval_runtime": 579.658, | |
| "eval_samples_per_second": 5.018, | |
| "eval_steps_per_second": 0.157, | |
| "learning_rate": 0.001, | |
| "step": 3822 | |
| }, | |
| { | |
| "epoch": 14.652014652014651, | |
| "grad_norm": 0.17000725865364075, | |
| "learning_rate": 0.001, | |
| "loss": 0.1575, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.2598831213475421, | |
| "eval_f1_macro": 0.6552072842486797, | |
| "eval_f1_micro": 0.7859620485615181, | |
| "eval_loss": 0.1447945237159729, | |
| "eval_roc_auc": 0.8622108111394943, | |
| "eval_runtime": 578.0771, | |
| "eval_samples_per_second": 5.032, | |
| "eval_steps_per_second": 0.157, | |
| "learning_rate": 0.001, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.2605706428325885, | |
| "eval_f1_macro": 0.6495757946819554, | |
| "eval_f1_micro": 0.7853051058530511, | |
| "eval_loss": 0.1438169628381729, | |
| "eval_roc_auc": 0.8571207689487904, | |
| "eval_runtime": 577.2113, | |
| "eval_samples_per_second": 5.04, | |
| "eval_steps_per_second": 0.158, | |
| "learning_rate": 0.001, | |
| "step": 4368 | |
| }, | |
| { | |
| "epoch": 16.483516483516482, | |
| "grad_norm": 0.16623707115650177, | |
| "learning_rate": 0.001, | |
| "loss": 0.158, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.2506015812994156, | |
| "eval_f1_macro": 0.6310969679900952, | |
| "eval_f1_micro": 0.7824457675812967, | |
| "eval_loss": 0.14359386265277863, | |
| "eval_roc_auc": 0.8547290741339532, | |
| "eval_runtime": 583.6271, | |
| "eval_samples_per_second": 4.984, | |
| "eval_steps_per_second": 0.156, | |
| "learning_rate": 0.001, | |
| "step": 4641 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.2564455139223101, | |
| "eval_f1_macro": 0.6531950395959965, | |
| "eval_f1_micro": 0.7848311343456975, | |
| "eval_loss": 0.1412857472896576, | |
| "eval_roc_auc": 0.8569977513721464, | |
| "eval_runtime": 571.2179, | |
| "eval_samples_per_second": 5.093, | |
| "eval_steps_per_second": 0.159, | |
| "learning_rate": 0.001, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 18.315018315018314, | |
| "grad_norm": 0.14722026884555817, | |
| "learning_rate": 0.001, | |
| "loss": 0.1571, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.26022688209006534, | |
| "eval_f1_macro": 0.6486819478687708, | |
| "eval_f1_micro": 0.7833830386020918, | |
| "eval_loss": 0.14079046249389648, | |
| "eval_roc_auc": 0.852859282271344, | |
| "eval_runtime": 574.5699, | |
| "eval_samples_per_second": 5.063, | |
| "eval_steps_per_second": 0.158, | |
| "learning_rate": 0.001, | |
| "step": 5187 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.26400825025782054, | |
| "eval_f1_macro": 0.6262168341318395, | |
| "eval_f1_micro": 0.7775968460747342, | |
| "eval_loss": 0.14640754461288452, | |
| "eval_roc_auc": 0.8444495064377587, | |
| "eval_runtime": 579.1715, | |
| "eval_samples_per_second": 5.023, | |
| "eval_steps_per_second": 0.157, | |
| "learning_rate": 0.001, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 20.146520146520146, | |
| "grad_norm": 0.16411201655864716, | |
| "learning_rate": 0.001, | |
| "loss": 0.1579, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.2653832932279134, | |
| "eval_f1_macro": 0.6582044080070929, | |
| "eval_f1_micro": 0.7890916719110552, | |
| "eval_loss": 0.1412632316350937, | |
| "eval_roc_auc": 0.8562690934879753, | |
| "eval_runtime": 565.162, | |
| "eval_samples_per_second": 5.147, | |
| "eval_steps_per_second": 0.161, | |
| "learning_rate": 0.001, | |
| "step": 5733 | |
| }, | |
| { | |
| "epoch": 21.978021978021978, | |
| "grad_norm": 0.14830045402050018, | |
| "learning_rate": 0.001, | |
| "loss": 0.1564, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.2543829494671708, | |
| "eval_f1_macro": 0.6586947128782558, | |
| "eval_f1_micro": 0.7871090517954659, | |
| "eval_loss": 0.14168681204319, | |
| "eval_roc_auc": 0.8590452124964612, | |
| "eval_runtime": 558.0731, | |
| "eval_samples_per_second": 5.213, | |
| "eval_steps_per_second": 0.163, | |
| "learning_rate": 0.001, | |
| "step": 6006 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.269852182880715, | |
| "eval_f1_macro": 0.6427873985434494, | |
| "eval_f1_micro": 0.7863651704353696, | |
| "eval_loss": 0.1393543779850006, | |
| "eval_roc_auc": 0.8532977785132612, | |
| "eval_runtime": 555.9424, | |
| "eval_samples_per_second": 5.233, | |
| "eval_steps_per_second": 0.164, | |
| "learning_rate": 0.001, | |
| "step": 6279 | |
| }, | |
| { | |
| "epoch": 23.80952380952381, | |
| "grad_norm": 0.1764509379863739, | |
| "learning_rate": 0.001, | |
| "loss": 0.1554, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.2588518391199725, | |
| "eval_f1_macro": 0.6618962794412713, | |
| "eval_f1_micro": 0.7857706852844616, | |
| "eval_loss": 0.14052371680736542, | |
| "eval_roc_auc": 0.8570680235127297, | |
| "eval_runtime": 561.2023, | |
| "eval_samples_per_second": 5.184, | |
| "eval_steps_per_second": 0.162, | |
| "learning_rate": 0.001, | |
| "step": 6552 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.2653832932279134, | |
| "eval_f1_macro": 0.653320279245233, | |
| "eval_f1_micro": 0.7897693920335429, | |
| "eval_loss": 0.1392364352941513, | |
| "eval_roc_auc": 0.8568084813100703, | |
| "eval_runtime": 557.349, | |
| "eval_samples_per_second": 5.219, | |
| "eval_steps_per_second": 0.163, | |
| "learning_rate": 0.001, | |
| "step": 6825 | |
| }, | |
| { | |
| "epoch": 25.641025641025642, | |
| "grad_norm": 0.17609645426273346, | |
| "learning_rate": 0.001, | |
| "loss": 0.1554, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.27019594362323823, | |
| "eval_f1_macro": 0.6529431984792132, | |
| "eval_f1_micro": 0.7838044308632545, | |
| "eval_loss": 0.14239099621772766, | |
| "eval_roc_auc": 0.8503066581053809, | |
| "eval_runtime": 557.4603, | |
| "eval_samples_per_second": 5.218, | |
| "eval_steps_per_second": 0.163, | |
| "learning_rate": 0.001, | |
| "step": 7098 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.2671020969405294, | |
| "eval_f1_macro": 0.6810613208979668, | |
| "eval_f1_micro": 0.7974886125815585, | |
| "eval_loss": 0.1386287957429886, | |
| "eval_roc_auc": 0.8670442949969421, | |
| "eval_runtime": 549.4206, | |
| "eval_samples_per_second": 5.295, | |
| "eval_steps_per_second": 0.166, | |
| "learning_rate": 0.001, | |
| "step": 7371 | |
| }, | |
| { | |
| "epoch": 27.47252747252747, | |
| "grad_norm": 0.1621919423341751, | |
| "learning_rate": 0.001, | |
| "loss": 0.156, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.2650395324853902, | |
| "eval_f1_macro": 0.6474807711800876, | |
| "eval_f1_micro": 0.7791304347826087, | |
| "eval_loss": 0.15519200265407562, | |
| "eval_roc_auc": 0.8544148384115767, | |
| "eval_runtime": 547.1571, | |
| "eval_samples_per_second": 5.317, | |
| "eval_steps_per_second": 0.166, | |
| "learning_rate": 0.001, | |
| "step": 7644 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.27019594362323823, | |
| "eval_f1_macro": 0.6550381793679035, | |
| "eval_f1_micro": 0.7913651213762871, | |
| "eval_loss": 0.14190098643302917, | |
| "eval_roc_auc": 0.8614869597044164, | |
| "eval_runtime": 559.1951, | |
| "eval_samples_per_second": 5.202, | |
| "eval_steps_per_second": 0.163, | |
| "learning_rate": 0.001, | |
| "step": 7917 | |
| }, | |
| { | |
| "epoch": 29.304029304029303, | |
| "grad_norm": 0.16609038412570953, | |
| "learning_rate": 0.001, | |
| "loss": 0.1548, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.2767273977311791, | |
| "eval_f1_macro": 0.663185953977854, | |
| "eval_f1_micro": 0.7857173292428311, | |
| "eval_loss": 0.13986903429031372, | |
| "eval_roc_auc": 0.8502714393738423, | |
| "eval_runtime": 549.5664, | |
| "eval_samples_per_second": 5.293, | |
| "eval_steps_per_second": 0.166, | |
| "learning_rate": 0.001, | |
| "step": 8190 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.27260226882090066, | |
| "eval_f1_macro": 0.6554744698272669, | |
| "eval_f1_micro": 0.7881844380403459, | |
| "eval_loss": 0.13765402138233185, | |
| "eval_roc_auc": 0.8530413436678441, | |
| "eval_runtime": 541.1802, | |
| "eval_samples_per_second": 5.375, | |
| "eval_steps_per_second": 0.168, | |
| "learning_rate": 0.001, | |
| "step": 8463 | |
| }, | |
| { | |
| "epoch": 31.135531135531135, | |
| "grad_norm": 0.15226389467716217, | |
| "learning_rate": 0.001, | |
| "loss": 0.1554, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.2677896184255758, | |
| "eval_f1_macro": 0.6596978272887946, | |
| "eval_f1_micro": 0.7914770376499792, | |
| "eval_loss": 0.13866138458251953, | |
| "eval_roc_auc": 0.861421701170076, | |
| "eval_runtime": 550.2376, | |
| "eval_samples_per_second": 5.287, | |
| "eval_steps_per_second": 0.165, | |
| "learning_rate": 0.001, | |
| "step": 8736 | |
| }, | |
| { | |
| "epoch": 32.967032967032964, | |
| "grad_norm": 0.15690498054027557, | |
| "learning_rate": 0.001, | |
| "loss": 0.1551, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.2605706428325885, | |
| "eval_f1_macro": 0.6583814800932023, | |
| "eval_f1_micro": 0.7887546855476885, | |
| "eval_loss": 0.13930276036262512, | |
| "eval_roc_auc": 0.8580923964636631, | |
| "eval_runtime": 550.269, | |
| "eval_samples_per_second": 5.287, | |
| "eval_steps_per_second": 0.165, | |
| "learning_rate": 0.001, | |
| "step": 9009 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.2763836369886559, | |
| "eval_f1_macro": 0.6636727922636001, | |
| "eval_f1_micro": 0.795303262082937, | |
| "eval_loss": 0.1374826431274414, | |
| "eval_roc_auc": 0.8637208325253699, | |
| "eval_runtime": 543.5944, | |
| "eval_samples_per_second": 5.351, | |
| "eval_steps_per_second": 0.167, | |
| "learning_rate": 0.001, | |
| "step": 9282 | |
| }, | |
| { | |
| "epoch": 34.798534798534796, | |
| "grad_norm": 0.166019469499588, | |
| "learning_rate": 0.001, | |
| "loss": 0.1544, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.25850807837744927, | |
| "eval_f1_macro": 0.6442491093834092, | |
| "eval_f1_micro": 0.7860775988902434, | |
| "eval_loss": 0.14001137018203735, | |
| "eval_roc_auc": 0.8541119565138938, | |
| "eval_runtime": 537.9679, | |
| "eval_samples_per_second": 5.407, | |
| "eval_steps_per_second": 0.169, | |
| "learning_rate": 0.001, | |
| "step": 9555 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.26916466139566864, | |
| "eval_f1_macro": 0.6541220211466265, | |
| "eval_f1_micro": 0.7890085033301218, | |
| "eval_loss": 0.13899104297161102, | |
| "eval_roc_auc": 0.856659816065419, | |
| "eval_runtime": 532.0356, | |
| "eval_samples_per_second": 5.468, | |
| "eval_steps_per_second": 0.171, | |
| "learning_rate": 0.001, | |
| "step": 9828 | |
| }, | |
| { | |
| "epoch": 36.63003663003663, | |
| "grad_norm": 0.1606360822916031, | |
| "learning_rate": 0.001, | |
| "loss": 0.1555, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.2667583361980062, | |
| "eval_f1_macro": 0.6602790790864311, | |
| "eval_f1_micro": 0.788356222091162, | |
| "eval_loss": 0.14101693034172058, | |
| "eval_roc_auc": 0.8547597772428587, | |
| "eval_runtime": 536.869, | |
| "eval_samples_per_second": 5.418, | |
| "eval_steps_per_second": 0.17, | |
| "learning_rate": 0.001, | |
| "step": 10101 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.2633207287727741, | |
| "eval_f1_macro": 0.6508514926081754, | |
| "eval_f1_micro": 0.7864065343433915, | |
| "eval_loss": 0.13849563896656036, | |
| "eval_roc_auc": 0.8524644165509108, | |
| "eval_runtime": 538.0729, | |
| "eval_samples_per_second": 5.406, | |
| "eval_steps_per_second": 0.169, | |
| "learning_rate": 0.001, | |
| "step": 10374 | |
| }, | |
| { | |
| "epoch": 38.46153846153846, | |
| "grad_norm": 0.1459091752767563, | |
| "learning_rate": 0.001, | |
| "loss": 0.1547, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.26263320728772777, | |
| "eval_f1_macro": 0.6513021077089046, | |
| "eval_f1_micro": 0.7819844457738655, | |
| "eval_loss": 0.14249388873577118, | |
| "eval_roc_auc": 0.8475227906130532, | |
| "eval_runtime": 528.7471, | |
| "eval_samples_per_second": 5.502, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 0.001, | |
| "step": 10647 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.2633207287727741, | |
| "eval_f1_macro": 0.6421624481517915, | |
| "eval_f1_micro": 0.7819497946916141, | |
| "eval_loss": 0.1512959599494934, | |
| "eval_roc_auc": 0.850003278224088, | |
| "eval_runtime": 526.9207, | |
| "eval_samples_per_second": 5.521, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.001, | |
| "step": 10920 | |
| }, | |
| { | |
| "epoch": 40.29304029304029, | |
| "grad_norm": 0.16056223213672638, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1527, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.27157098659333107, | |
| "eval_f1_macro": 0.6708412782877394, | |
| "eval_f1_micro": 0.795353889863792, | |
| "eval_loss": 0.1416281908750534, | |
| "eval_roc_auc": 0.8608108400991562, | |
| "eval_runtime": 520.8949, | |
| "eval_samples_per_second": 5.585, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 11193 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.2811962873839807, | |
| "eval_f1_macro": 0.6820172839356666, | |
| "eval_f1_micro": 0.8014906832298136, | |
| "eval_loss": 0.13480685651302338, | |
| "eval_roc_auc": 0.8666871757284966, | |
| "eval_runtime": 521.536, | |
| "eval_samples_per_second": 5.578, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 11466 | |
| }, | |
| { | |
| "epoch": 42.124542124542124, | |
| "grad_norm": 0.13122691214084625, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1455, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.2756961155036095, | |
| "eval_f1_macro": 0.681931169239128, | |
| "eval_f1_micro": 0.8014919187733112, | |
| "eval_loss": 0.1342025101184845, | |
| "eval_roc_auc": 0.8664528250448902, | |
| "eval_runtime": 522.5262, | |
| "eval_samples_per_second": 5.567, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 11739 | |
| }, | |
| { | |
| "epoch": 43.956043956043956, | |
| "grad_norm": 0.15952740609645844, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1416, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.2811962873839807, | |
| "eval_f1_macro": 0.683693351140427, | |
| "eval_f1_micro": 0.8019789631231031, | |
| "eval_loss": 0.1327475756406784, | |
| "eval_roc_auc": 0.8658142028998129, | |
| "eval_runtime": 518.6048, | |
| "eval_samples_per_second": 5.609, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 12012 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.2811962873839807, | |
| "eval_f1_macro": 0.6900135704395078, | |
| "eval_f1_micro": 0.8049446006284108, | |
| "eval_loss": 0.1318245828151703, | |
| "eval_roc_auc": 0.8690323816564128, | |
| "eval_runtime": 519.4737, | |
| "eval_samples_per_second": 5.6, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 12285 | |
| }, | |
| { | |
| "epoch": 45.78754578754579, | |
| "grad_norm": 0.18243736028671265, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1402, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.28910278446201443, | |
| "eval_f1_macro": 0.6920134474185277, | |
| "eval_f1_micro": 0.8063969585520062, | |
| "eval_loss": 0.13027183711528778, | |
| "eval_roc_auc": 0.869961249113055, | |
| "eval_runtime": 520.8317, | |
| "eval_samples_per_second": 5.585, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 12558 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.284977655551736, | |
| "eval_f1_macro": 0.6938459582689339, | |
| "eval_f1_micro": 0.8065087538619978, | |
| "eval_loss": 0.12985946238040924, | |
| "eval_roc_auc": 0.8709245066726377, | |
| "eval_runtime": 520.349, | |
| "eval_samples_per_second": 5.59, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 12831 | |
| }, | |
| { | |
| "epoch": 47.61904761904762, | |
| "grad_norm": 0.20866894721984863, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1387, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.2853214162942592, | |
| "eval_f1_macro": 0.6917397436201066, | |
| "eval_f1_micro": 0.8031727379553465, | |
| "eval_loss": 0.12981055676937103, | |
| "eval_roc_auc": 0.8638141011256728, | |
| "eval_runtime": 515.7048, | |
| "eval_samples_per_second": 5.641, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 0.0001, | |
| "step": 13104 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.2839463733241664, | |
| "eval_f1_macro": 0.6980761423122126, | |
| "eval_f1_micro": 0.8081048867699644, | |
| "eval_loss": 0.1301460713148117, | |
| "eval_roc_auc": 0.8724715693697419, | |
| "eval_runtime": 519.0148, | |
| "eval_samples_per_second": 5.605, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 13377 | |
| }, | |
| { | |
| "epoch": 49.45054945054945, | |
| "grad_norm": 0.16819314658641815, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1391, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.2829150910965968, | |
| "eval_f1_macro": 0.6968263757426811, | |
| "eval_f1_micro": 0.8056895691232739, | |
| "eval_loss": 0.1294524371623993, | |
| "eval_roc_auc": 0.8677359534045406, | |
| "eval_runtime": 516.1382, | |
| "eval_samples_per_second": 5.636, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 0.0001, | |
| "step": 13650 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.2846338948092128, | |
| "eval_f1_macro": 0.6981227572539419, | |
| "eval_f1_micro": 0.8078541374474054, | |
| "eval_loss": 0.12989668548107147, | |
| "eval_roc_auc": 0.8712741682837021, | |
| "eval_runtime": 519.4542, | |
| "eval_samples_per_second": 5.6, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 13923 | |
| }, | |
| { | |
| "epoch": 51.282051282051285, | |
| "grad_norm": 0.2057354748249054, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1374, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.284977655551736, | |
| "eval_f1_macro": 0.7032059573412642, | |
| "eval_f1_micro": 0.809621541745341, | |
| "eval_loss": 0.13097986578941345, | |
| "eval_roc_auc": 0.8736802406111642, | |
| "eval_runtime": 517.4492, | |
| "eval_samples_per_second": 5.622, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 0.0001, | |
| "step": 14196 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.288415262976968, | |
| "eval_f1_macro": 0.6952081515364695, | |
| "eval_f1_micro": 0.8082875892525485, | |
| "eval_loss": 0.12910524010658264, | |
| "eval_roc_auc": 0.8713581964654615, | |
| "eval_runtime": 515.3537, | |
| "eval_samples_per_second": 5.645, | |
| "eval_steps_per_second": 0.177, | |
| "learning_rate": 0.0001, | |
| "step": 14469 | |
| }, | |
| { | |
| "epoch": 53.11355311355312, | |
| "grad_norm": 0.19681566953659058, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1367, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.2860089377793056, | |
| "eval_f1_macro": 0.6914506394370794, | |
| "eval_f1_micro": 0.8055729885778838, | |
| "eval_loss": 0.1276824176311493, | |
| "eval_roc_auc": 0.8659785304346413, | |
| "eval_runtime": 517.0408, | |
| "eval_samples_per_second": 5.626, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 0.0001, | |
| "step": 14742 | |
| }, | |
| { | |
| "epoch": 54.94505494505494, | |
| "grad_norm": 0.21626819670200348, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1364, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.28979030594706084, | |
| "eval_f1_macro": 0.7051415507931676, | |
| "eval_f1_micro": 0.8091508143727464, | |
| "eval_loss": 0.12751279771327972, | |
| "eval_roc_auc": 0.8706409740591669, | |
| "eval_runtime": 516.7044, | |
| "eval_samples_per_second": 5.63, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 0.0001, | |
| "step": 15015 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.2911653489171537, | |
| "eval_f1_macro": 0.6990943862949641, | |
| "eval_f1_micro": 0.8077718065316246, | |
| "eval_loss": 0.12798655033111572, | |
| "eval_roc_auc": 0.8706103127849166, | |
| "eval_runtime": 520.6769, | |
| "eval_samples_per_second": 5.587, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 0.0001, | |
| "step": 15288 | |
| }, | |
| { | |
| "epoch": 56.776556776556774, | |
| "grad_norm": 0.2569683790206909, | |
| "learning_rate": 0.0001, | |
| "loss": 0.135, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.29150910965967686, | |
| "eval_f1_macro": 0.7001268142729874, | |
| "eval_f1_micro": 0.8107930240210597, | |
| "eval_loss": 0.1279618740081787, | |
| "eval_roc_auc": 0.8737307073185808, | |
| "eval_runtime": 522.0888, | |
| "eval_samples_per_second": 5.572, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 15561 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.290134066689584, | |
| "eval_f1_macro": 0.7039327958876614, | |
| "eval_f1_micro": 0.8108946874106743, | |
| "eval_loss": 0.1280883550643921, | |
| "eval_roc_auc": 0.8759047076910806, | |
| "eval_runtime": 524.5592, | |
| "eval_samples_per_second": 5.546, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 15834 | |
| }, | |
| { | |
| "epoch": 58.608058608058606, | |
| "grad_norm": 0.209602490067482, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1345, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.2873839807493984, | |
| "eval_f1_macro": 0.699653006099352, | |
| "eval_f1_micro": 0.8071845383437488, | |
| "eval_loss": 0.1287168562412262, | |
| "eval_roc_auc": 0.8699666770335968, | |
| "eval_runtime": 523.8253, | |
| "eval_samples_per_second": 5.553, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 16107 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.28875902371949125, | |
| "eval_f1_macro": 0.7042073996338176, | |
| "eval_f1_micro": 0.8103491168421926, | |
| "eval_loss": 0.1270500272512436, | |
| "eval_roc_auc": 0.8711666058400855, | |
| "eval_runtime": 535.2866, | |
| "eval_samples_per_second": 5.434, | |
| "eval_steps_per_second": 0.17, | |
| "learning_rate": 0.0001, | |
| "step": 16380 | |
| }, | |
| { | |
| "epoch": 60.43956043956044, | |
| "grad_norm": 0.2388932704925537, | |
| "learning_rate": 0.0001, | |
| "loss": 0.134, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.28944654520453766, | |
| "eval_f1_macro": 0.6994480698947442, | |
| "eval_f1_micro": 0.8072888368788399, | |
| "eval_loss": 0.1269637793302536, | |
| "eval_roc_auc": 0.8672195202075096, | |
| "eval_runtime": 529.882, | |
| "eval_samples_per_second": 5.49, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 0.0001, | |
| "step": 16653 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.28979030594706084, | |
| "eval_f1_macro": 0.7105518005302388, | |
| "eval_f1_micro": 0.8124407826982492, | |
| "eval_loss": 0.12639474868774414, | |
| "eval_roc_auc": 0.8741971299571496, | |
| "eval_runtime": 525.1851, | |
| "eval_samples_per_second": 5.539, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 16926 | |
| }, | |
| { | |
| "epoch": 62.27106227106227, | |
| "grad_norm": 0.2082633525133133, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1331, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.2918528704022001, | |
| "eval_f1_macro": 0.7042257858113937, | |
| "eval_f1_micro": 0.8093336660843524, | |
| "eval_loss": 0.12643341720104218, | |
| "eval_roc_auc": 0.8697033641675914, | |
| "eval_runtime": 525.5019, | |
| "eval_samples_per_second": 5.536, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 17199 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.2918528704022001, | |
| "eval_f1_macro": 0.7054117610081568, | |
| "eval_f1_micro": 0.8119739624362535, | |
| "eval_loss": 0.12570597231388092, | |
| "eval_roc_auc": 0.8720692041572721, | |
| "eval_runtime": 522.3213, | |
| "eval_samples_per_second": 5.569, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 17472 | |
| }, | |
| { | |
| "epoch": 64.1025641025641, | |
| "grad_norm": 0.25887489318847656, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1327, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.29322791337229287, | |
| "eval_f1_macro": 0.7040599127700347, | |
| "eval_f1_micro": 0.8103770839396333, | |
| "eval_loss": 0.12599390745162964, | |
| "eval_roc_auc": 0.8704049661183646, | |
| "eval_runtime": 523.7812, | |
| "eval_samples_per_second": 5.554, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 17745 | |
| }, | |
| { | |
| "epoch": 65.93406593406593, | |
| "grad_norm": 0.2609516382217407, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1319, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.29769680302509455, | |
| "eval_f1_macro": 0.7083351143800681, | |
| "eval_f1_micro": 0.8141795311606633, | |
| "eval_loss": 0.12674611806869507, | |
| "eval_roc_auc": 0.8776967697557255, | |
| "eval_runtime": 525.4493, | |
| "eval_samples_per_second": 5.536, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 18018 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.28979030594706084, | |
| "eval_f1_macro": 0.6998024530144022, | |
| "eval_f1_micro": 0.8090950582963362, | |
| "eval_loss": 0.12676431238651276, | |
| "eval_roc_auc": 0.8702765206211787, | |
| "eval_runtime": 528.6841, | |
| "eval_samples_per_second": 5.502, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 0.0001, | |
| "step": 18291 | |
| }, | |
| { | |
| "epoch": 67.76556776556777, | |
| "grad_norm": 0.31347450613975525, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1319, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.2928841526297697, | |
| "eval_f1_macro": 0.7034736625177254, | |
| "eval_f1_micro": 0.8127327032445482, | |
| "eval_loss": 0.12638631463050842, | |
| "eval_roc_auc": 0.8763513248964829, | |
| "eval_runtime": 524.2249, | |
| "eval_samples_per_second": 5.549, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 18564 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.2952904778274321, | |
| "eval_f1_macro": 0.7078892431331377, | |
| "eval_f1_micro": 0.8131967584022379, | |
| "eval_loss": 0.12608103454113007, | |
| "eval_roc_auc": 0.8750227852703141, | |
| "eval_runtime": 527.0116, | |
| "eval_samples_per_second": 5.52, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 0.0001, | |
| "step": 18837 | |
| }, | |
| { | |
| "epoch": 69.59706959706959, | |
| "grad_norm": 0.24166111648082733, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1308, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.29150910965967686, | |
| "eval_f1_macro": 0.7081157868651535, | |
| "eval_f1_micro": 0.8136722606120435, | |
| "eval_loss": 0.12582050263881683, | |
| "eval_roc_auc": 0.875321565859663, | |
| "eval_runtime": 523.4787, | |
| "eval_samples_per_second": 5.557, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 0.0001, | |
| "step": 19110 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.2918528704022001, | |
| "eval_f1_macro": 0.7044517956080781, | |
| "eval_f1_micro": 0.8123295595405339, | |
| "eval_loss": 0.12533149123191833, | |
| "eval_roc_auc": 0.8732599343798731, | |
| "eval_runtime": 528.0987, | |
| "eval_samples_per_second": 5.508, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 1e-05, | |
| "step": 19383 | |
| }, | |
| { | |
| "epoch": 71.42857142857143, | |
| "grad_norm": 0.2781914174556732, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1294, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.2966655207975249, | |
| "eval_f1_macro": 0.7099295458861072, | |
| "eval_f1_micro": 0.8159506713723581, | |
| "eval_loss": 0.1258901059627533, | |
| "eval_roc_auc": 0.8806755842224759, | |
| "eval_runtime": 528.1564, | |
| "eval_samples_per_second": 5.508, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 1e-05, | |
| "step": 19656 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.2949467170849089, | |
| "eval_f1_macro": 0.7116557450872655, | |
| "eval_f1_micro": 0.8159496670343587, | |
| "eval_loss": 0.12526649236679077, | |
| "eval_roc_auc": 0.8785730989334561, | |
| "eval_runtime": 529.1223, | |
| "eval_samples_per_second": 5.498, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 1e-05, | |
| "step": 19929 | |
| }, | |
| { | |
| "epoch": 73.26007326007326, | |
| "grad_norm": 0.278796911239624, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1287, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.29769680302509455, | |
| "eval_f1_macro": 0.7159515864206437, | |
| "eval_f1_micro": 0.8156100747030249, | |
| "eval_loss": 0.12490212172269821, | |
| "eval_roc_auc": 0.8786256372123677, | |
| "eval_runtime": 527.6257, | |
| "eval_samples_per_second": 5.513, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 1e-05, | |
| "step": 20202 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.2966655207975249, | |
| "eval_f1_macro": 0.7082306828309269, | |
| "eval_f1_micro": 0.8135426082669078, | |
| "eval_loss": 0.12504002451896667, | |
| "eval_roc_auc": 0.8755625874067082, | |
| "eval_runtime": 534.3244, | |
| "eval_samples_per_second": 5.444, | |
| "eval_steps_per_second": 0.17, | |
| "learning_rate": 1e-05, | |
| "step": 20475 | |
| }, | |
| { | |
| "epoch": 75.0915750915751, | |
| "grad_norm": 0.27821090817451477, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1282, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.2966655207975249, | |
| "eval_f1_macro": 0.6998917153140419, | |
| "eval_f1_micro": 0.8099675513769865, | |
| "eval_loss": 0.12634462118148804, | |
| "eval_roc_auc": 0.8700019375997908, | |
| "eval_runtime": 530.2495, | |
| "eval_samples_per_second": 5.486, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 1e-05, | |
| "step": 20748 | |
| }, | |
| { | |
| "epoch": 76.92307692307692, | |
| "grad_norm": 0.2594936192035675, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1285, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.2966655207975249, | |
| "eval_f1_macro": 0.7104044870773909, | |
| "eval_f1_micro": 0.8142915811088296, | |
| "eval_loss": 0.1249643936753273, | |
| "eval_roc_auc": 0.8761213663476708, | |
| "eval_runtime": 520.8557, | |
| "eval_samples_per_second": 5.585, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 1e-05, | |
| "step": 21021 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.2939154348573393, | |
| "eval_f1_macro": 0.7076718539561497, | |
| "eval_f1_micro": 0.812339968613199, | |
| "eval_loss": 0.12509745359420776, | |
| "eval_roc_auc": 0.8734096354102718, | |
| "eval_runtime": 523.1264, | |
| "eval_samples_per_second": 5.561, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 1e-05, | |
| "step": 21294 | |
| }, | |
| { | |
| "epoch": 78.75457875457876, | |
| "grad_norm": 0.24959908425807953, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1281, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.29838432451014096, | |
| "eval_f1_macro": 0.7097766100728804, | |
| "eval_f1_micro": 0.8147326016360423, | |
| "eval_loss": 0.12465520948171616, | |
| "eval_roc_auc": 0.876088219187627, | |
| "eval_runtime": 536.4997, | |
| "eval_samples_per_second": 5.422, | |
| "eval_steps_per_second": 0.17, | |
| "learning_rate": 1e-05, | |
| "step": 21567 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.2990718459951873, | |
| "eval_f1_macro": 0.7133791911991404, | |
| "eval_f1_micro": 0.8166140393490405, | |
| "eval_loss": 0.12526248395442963, | |
| "eval_roc_auc": 0.8816640751357672, | |
| "eval_runtime": 530.0362, | |
| "eval_samples_per_second": 5.488, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 1e-05, | |
| "step": 21840 | |
| }, | |
| { | |
| "epoch": 80.58608058608058, | |
| "grad_norm": 0.2447408139705658, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1281, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.2952904778274321, | |
| "eval_f1_macro": 0.705898272950067, | |
| "eval_f1_micro": 0.8121923983622152, | |
| "eval_loss": 0.12510864436626434, | |
| "eval_roc_auc": 0.8729086155116128, | |
| "eval_runtime": 529.107, | |
| "eval_samples_per_second": 5.498, | |
| "eval_steps_per_second": 0.172, | |
| "learning_rate": 1e-05, | |
| "step": 22113 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.29975936748023374, | |
| "eval_f1_macro": 0.7095032932540235, | |
| "eval_f1_micro": 0.8150326797385622, | |
| "eval_loss": 0.12532733380794525, | |
| "eval_roc_auc": 0.8780991768028508, | |
| "eval_runtime": 520.4428, | |
| "eval_samples_per_second": 5.589, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 1e-05, | |
| "step": 22386 | |
| }, | |
| { | |
| "epoch": 82.41758241758242, | |
| "grad_norm": 0.29986944794654846, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1269, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.29597799931247853, | |
| "eval_f1_macro": 0.7124383950303705, | |
| "eval_f1_micro": 0.815855206584497, | |
| "eval_loss": 0.12474868446588516, | |
| "eval_roc_auc": 0.87784419205748, | |
| "eval_runtime": 521.522, | |
| "eval_samples_per_second": 5.578, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 1e-05, | |
| "step": 22659 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.3007906497078034, | |
| "eval_f1_macro": 0.7138847615465347, | |
| "eval_f1_micro": 0.8175330467926365, | |
| "eval_loss": 0.12511858344078064, | |
| "eval_roc_auc": 0.8816953578814657, | |
| "eval_runtime": 527.0352, | |
| "eval_samples_per_second": 5.52, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 1e-05, | |
| "step": 22932 | |
| }, | |
| { | |
| "epoch": 84.24908424908425, | |
| "grad_norm": 0.23814542591571808, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1267, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.2966655207975249, | |
| "eval_f1_macro": 0.7054571621251418, | |
| "eval_f1_micro": 0.8132141082960754, | |
| "eval_loss": 0.12457013875246048, | |
| "eval_roc_auc": 0.8731936400597855, | |
| "eval_runtime": 570.3692, | |
| "eval_samples_per_second": 5.1, | |
| "eval_steps_per_second": 0.16, | |
| "learning_rate": 1e-05, | |
| "step": 23205 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.2946029563423857, | |
| "eval_f1_macro": 0.7142702846379808, | |
| "eval_f1_micro": 0.8143732269868025, | |
| "eval_loss": 0.1251869946718216, | |
| "eval_roc_auc": 0.875841355405741, | |
| "eval_runtime": 518.8454, | |
| "eval_samples_per_second": 5.607, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 1e-05, | |
| "step": 23478 | |
| }, | |
| { | |
| "epoch": 86.08058608058609, | |
| "grad_norm": 0.30401352047920227, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1274, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.2935716741148161, | |
| "eval_f1_macro": 0.7081357577756824, | |
| "eval_f1_micro": 0.8135328455150868, | |
| "eval_loss": 0.12492978572845459, | |
| "eval_roc_auc": 0.8754128863763097, | |
| "eval_runtime": 516.882, | |
| "eval_samples_per_second": 5.628, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 1e-05, | |
| "step": 23751 | |
| }, | |
| { | |
| "epoch": 87.91208791208791, | |
| "grad_norm": 0.25764307379722595, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1263, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.2990718459951873, | |
| "eval_f1_macro": 0.7099379006276698, | |
| "eval_f1_micro": 0.815831263487927, | |
| "eval_loss": 0.12513719499111176, | |
| "eval_roc_auc": 0.8794975742553038, | |
| "eval_runtime": 519.5458, | |
| "eval_samples_per_second": 5.599, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 1e-05, | |
| "step": 24024 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.2963217600550017, | |
| "eval_f1_macro": 0.7092910188720426, | |
| "eval_f1_micro": 0.8143914473684211, | |
| "eval_loss": 0.12514576315879822, | |
| "eval_roc_auc": 0.8758152934234348, | |
| "eval_runtime": 518.776, | |
| "eval_samples_per_second": 5.607, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 1e-05, | |
| "step": 24297 | |
| }, | |
| { | |
| "epoch": 89.74358974358974, | |
| "grad_norm": 0.25485533475875854, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1272, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.2942591955998625, | |
| "eval_f1_macro": 0.7121664381657501, | |
| "eval_f1_micro": 0.8134516195584898, | |
| "eval_loss": 0.1244530975818634, | |
| "eval_roc_auc": 0.8742738658040219, | |
| "eval_runtime": 522.4184, | |
| "eval_samples_per_second": 5.568, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 1e-05, | |
| "step": 24570 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.2990718459951873, | |
| "eval_f1_macro": 0.7106178930468596, | |
| "eval_f1_micro": 0.8153902768123646, | |
| "eval_loss": 0.12501013278961182, | |
| "eval_roc_auc": 0.8780205765416332, | |
| "eval_runtime": 521.4209, | |
| "eval_samples_per_second": 5.579, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 1e-05, | |
| "step": 24843 | |
| }, | |
| { | |
| "epoch": 91.57509157509158, | |
| "grad_norm": 0.24377481639385223, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1275, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.2973530422825713, | |
| "eval_f1_macro": 0.7140173113811211, | |
| "eval_f1_micro": 0.8163049232398094, | |
| "eval_loss": 0.12525025010108948, | |
| "eval_roc_auc": 0.8797377671737511, | |
| "eval_runtime": 518.7634, | |
| "eval_samples_per_second": 5.608, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 1e-05, | |
| "step": 25116 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.29872808525266414, | |
| "eval_f1_macro": 0.7129019083206937, | |
| "eval_f1_micro": 0.8148661314641998, | |
| "eval_loss": 0.12471849471330643, | |
| "eval_roc_auc": 0.8787177863576121, | |
| "eval_runtime": 519.6644, | |
| "eval_samples_per_second": 5.598, | |
| "eval_steps_per_second": 0.175, | |
| "learning_rate": 1e-05, | |
| "step": 25389 | |
| }, | |
| { | |
| "epoch": 93.4065934065934, | |
| "grad_norm": 0.2000974863767624, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1257, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.2980405637676177, | |
| "eval_f1_macro": 0.7053935701419592, | |
| "eval_f1_micro": 0.8141884924726748, | |
| "eval_loss": 0.12515641748905182, | |
| "eval_roc_auc": 0.8748001190563892, | |
| "eval_runtime": 516.6712, | |
| "eval_samples_per_second": 5.63, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 1e-05, | |
| "step": 25662 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.30147817119284975, | |
| "eval_f1_macro": 0.7134995447430972, | |
| "eval_f1_micro": 0.8165906870726147, | |
| "eval_loss": 0.12481416761875153, | |
| "eval_roc_auc": 0.8800302912139601, | |
| "eval_runtime": 518.3317, | |
| "eval_samples_per_second": 5.612, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 1e-05, | |
| "step": 25935 | |
| }, | |
| { | |
| "epoch": 95.23809523809524, | |
| "grad_norm": 0.284708708524704, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1271, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.2980405637676177, | |
| "eval_f1_macro": 0.7110442004495683, | |
| "eval_f1_micro": 0.8160666176830762, | |
| "eval_loss": 0.12492986023426056, | |
| "eval_roc_auc": 0.8788918921131211, | |
| "eval_runtime": 518.2372, | |
| "eval_samples_per_second": 5.613, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 1e-05, | |
| "step": 26208 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.30147817119284975, | |
| "eval_f1_macro": 0.7158597011246477, | |
| "eval_f1_micro": 0.8168590473093806, | |
| "eval_loss": 0.12459924072027206, | |
| "eval_roc_auc": 0.8806157536086776, | |
| "eval_runtime": 516.8314, | |
| "eval_samples_per_second": 5.629, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "step": 26481 | |
| }, | |
| { | |
| "epoch": 97.06959706959707, | |
| "grad_norm": 0.2898052930831909, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.1272, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.29975936748023374, | |
| "eval_f1_macro": 0.707122866121441, | |
| "eval_f1_micro": 0.8149457415323906, | |
| "eval_loss": 0.12447398155927658, | |
| "eval_roc_auc": 0.8762181147848592, | |
| "eval_runtime": 516.1798, | |
| "eval_samples_per_second": 5.636, | |
| "eval_steps_per_second": 0.176, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "step": 26754 | |
| }, | |
| { | |
| "epoch": 98.9010989010989, | |
| "grad_norm": 0.2835540473461151, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.126, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.30216569267789617, | |
| "eval_f1_macro": 0.7182970295785608, | |
| "eval_f1_micro": 0.8165748111859562, | |
| "eval_loss": 0.12462905794382095, | |
| "eval_roc_auc": 0.8791385487378576, | |
| "eval_runtime": 525.6968, | |
| "eval_samples_per_second": 5.534, | |
| "eval_steps_per_second": 0.173, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "step": 27027 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.30147817119284975, | |
| "eval_f1_macro": 0.7136275002413193, | |
| "eval_f1_micro": 0.8161644284310514, | |
| "eval_loss": 0.12463195621967316, | |
| "eval_roc_auc": 0.8780520664444814, | |
| "eval_runtime": 523.9869, | |
| "eval_samples_per_second": 5.552, | |
| "eval_steps_per_second": 0.174, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "step": 27300, | |
| "total_flos": 1.2912305794345248e+21, | |
| "train_loss": 0.1443542043106023, | |
| "train_runtime": 217230.0758, | |
| "train_samples_per_second": 6.013, | |
| "train_steps_per_second": 0.189 | |
| } | |
| ], | |
| "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": 1.2912305794345248e+21, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |