| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.043847560882568, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5619, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7167919799498746, | |
| "eval_f1": 0.6483759017352311, | |
| "eval_loss": 0.5126742124557495, | |
| "eval_precision": 0.653631914482581, | |
| "eval_recall": 0.6446172031278414, | |
| "eval_runtime": 5.1413, | |
| "eval_samples_per_second": 77.606, | |
| "eval_steps_per_second": 9.725, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.8154773712158203, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5059, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7343358395989975, | |
| "eval_f1": 0.7022388059701492, | |
| "eval_loss": 0.4967426359653473, | |
| "eval_precision": 0.695631275368511, | |
| "eval_recall": 0.7220403709765412, | |
| "eval_runtime": 5.0554, | |
| "eval_samples_per_second": 78.925, | |
| "eval_steps_per_second": 9.89, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.306382656097412, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4822, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7468671679197995, | |
| "eval_f1": 0.7064830258490717, | |
| "eval_loss": 0.45057693123817444, | |
| "eval_precision": 0.7006054632497888, | |
| "eval_recall": 0.7159028914348063, | |
| "eval_runtime": 5.0477, | |
| "eval_samples_per_second": 79.046, | |
| "eval_steps_per_second": 9.906, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.528663158416748, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4402, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8195488721804511, | |
| "eval_f1": 0.7727934198038595, | |
| "eval_loss": 0.39838677644729614, | |
| "eval_precision": 0.7875606352581567, | |
| "eval_recall": 0.76232042189489, | |
| "eval_runtime": 5.0575, | |
| "eval_samples_per_second": 78.893, | |
| "eval_steps_per_second": 9.886, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.9896986484527588, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4068, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7869674185463659, | |
| "eval_f1": 0.7560721246862345, | |
| "eval_loss": 0.4135726988315582, | |
| "eval_precision": 0.7472972972972973, | |
| "eval_recall": 0.7717766866703037, | |
| "eval_runtime": 5.0743, | |
| "eval_samples_per_second": 78.631, | |
| "eval_steps_per_second": 9.853, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.2119059562683105, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3791, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.7979441442122369, | |
| "eval_loss": 0.3770710527896881, | |
| "eval_precision": 0.7972133421798662, | |
| "eval_recall": 0.7986906710310966, | |
| "eval_runtime": 5.083, | |
| "eval_samples_per_second": 78.497, | |
| "eval_steps_per_second": 9.837, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.8872367143630981, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3635, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8195488721804511, | |
| "eval_f1": 0.7912087912087913, | |
| "eval_loss": 0.39156973361968994, | |
| "eval_precision": 0.7821969696969697, | |
| "eval_recall": 0.8048281505728314, | |
| "eval_runtime": 5.0823, | |
| "eval_samples_per_second": 78.507, | |
| "eval_steps_per_second": 9.838, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.425744533538818, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3433, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.7974014336917563, | |
| "eval_loss": 0.37988749146461487, | |
| "eval_precision": 0.7934340756451043, | |
| "eval_recall": 0.801918530641935, | |
| "eval_runtime": 5.0721, | |
| "eval_samples_per_second": 78.666, | |
| "eval_steps_per_second": 9.858, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.274303436279297, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3379, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_f1": 0.7958625291958625, | |
| "eval_loss": 0.3713899850845337, | |
| "eval_precision": 0.7902930402930403, | |
| "eval_recall": 0.802645935624659, | |
| "eval_runtime": 5.0569, | |
| "eval_samples_per_second": 78.902, | |
| "eval_steps_per_second": 9.887, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 6.189886569976807, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3296, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8039756622954537, | |
| "eval_loss": 0.3635365664958954, | |
| "eval_precision": 0.8032299897460643, | |
| "eval_recall": 0.8047372249499909, | |
| "eval_runtime": 5.0487, | |
| "eval_samples_per_second": 79.03, | |
| "eval_steps_per_second": 9.903, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 4.368456840515137, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.3105, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.7983530530947143, | |
| "eval_loss": 0.36521923542022705, | |
| "eval_precision": 0.7932554744525547, | |
| "eval_recall": 0.8044189852700492, | |
| "eval_runtime": 5.0638, | |
| "eval_samples_per_second": 78.795, | |
| "eval_steps_per_second": 9.874, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 11.303323745727539, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3024, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.8069354838709677, | |
| "eval_loss": 0.3701755404472351, | |
| "eval_precision": 0.798820781588242, | |
| "eval_recall": 0.8179669030732861, | |
| "eval_runtime": 5.0521, | |
| "eval_samples_per_second": 78.976, | |
| "eval_steps_per_second": 9.897, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.7624099254608154, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.309, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8039756622954537, | |
| "eval_loss": 0.35115164518356323, | |
| "eval_precision": 0.8032299897460643, | |
| "eval_recall": 0.8047372249499909, | |
| "eval_runtime": 5.0555, | |
| "eval_samples_per_second": 78.925, | |
| "eval_steps_per_second": 9.89, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 20.14969825744629, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.3021, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.8074672778816575, | |
| "eval_loss": 0.35049960017204285, | |
| "eval_precision": 0.8059873949579832, | |
| "eval_recall": 0.8090107292234952, | |
| "eval_runtime": 5.0531, | |
| "eval_samples_per_second": 78.961, | |
| "eval_steps_per_second": 9.895, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.1024099588394165, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2903, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8136136136136136, | |
| "eval_loss": 0.3553332984447479, | |
| "eval_precision": 0.8076923076923077, | |
| "eval_recall": 0.820785597381342, | |
| "eval_runtime": 5.0992, | |
| "eval_samples_per_second": 78.247, | |
| "eval_steps_per_second": 9.805, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 7.029820442199707, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2834, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.8119476846942383, | |
| "eval_loss": 0.3529706597328186, | |
| "eval_precision": 0.8045650301464256, | |
| "eval_recall": 0.8215130023640662, | |
| "eval_runtime": 5.0628, | |
| "eval_samples_per_second": 78.81, | |
| "eval_steps_per_second": 9.876, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 4.537665367126465, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2811, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8446115288220551, | |
| "eval_f1": 0.8134839254478557, | |
| "eval_loss": 0.34711357951164246, | |
| "eval_precision": 0.8119747899159664, | |
| "eval_recall": 0.8150572831423895, | |
| "eval_runtime": 5.0737, | |
| "eval_samples_per_second": 78.642, | |
| "eval_steps_per_second": 9.855, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.4649994373321533, | |
| "learning_rate": 5e-06, | |
| "loss": 0.288, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8446115288220551, | |
| "eval_f1": 0.8161454307628278, | |
| "eval_loss": 0.350491464138031, | |
| "eval_precision": 0.8107299270072992, | |
| "eval_recall": 0.8225586470267321, | |
| "eval_runtime": 5.0512, | |
| "eval_samples_per_second": 78.991, | |
| "eval_steps_per_second": 9.899, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.690791130065918, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.277, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.8084033613445378, | |
| "eval_loss": 0.3479045629501343, | |
| "eval_precision": 0.8055472976990309, | |
| "eval_recall": 0.8115111838516094, | |
| "eval_runtime": 5.063, | |
| "eval_samples_per_second": 78.807, | |
| "eval_steps_per_second": 9.876, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 7.5570807456970215, | |
| "learning_rate": 0.0, | |
| "loss": 0.2775, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.8084033613445378, | |
| "eval_loss": 0.3485976755619049, | |
| "eval_precision": 0.8055472976990309, | |
| "eval_recall": 0.8115111838516094, | |
| "eval_runtime": 5.0507, | |
| "eval_samples_per_second": 79.0, | |
| "eval_steps_per_second": 9.9, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7597037114448000.0, | |
| "train_loss": 0.35358311856379276, | |
| "train_runtime": 1950.2117, | |
| "train_samples_per_second": 37.309, | |
| "train_steps_per_second": 1.251 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7597037114448000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |