| { | |
| "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": 3.928847551345825, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5514, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7218045112781954, | |
| "eval_f1": 0.6545993371027491, | |
| "eval_loss": 0.5084339380264282, | |
| "eval_precision": 0.6600553802562947, | |
| "eval_recall": 0.6506637570467357, | |
| "eval_runtime": 1.7877, | |
| "eval_samples_per_second": 223.186, | |
| "eval_steps_per_second": 27.968, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.589020013809204, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4753, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8170426065162907, | |
| "eval_f1": 0.7662453352594198, | |
| "eval_loss": 0.40073099732398987, | |
| "eval_precision": 0.7870255775577557, | |
| "eval_recall": 0.7530460083651573, | |
| "eval_runtime": 1.8277, | |
| "eval_samples_per_second": 218.302, | |
| "eval_steps_per_second": 27.356, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 7.217240810394287, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3834, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.7805280528052805, | |
| "eval_loss": 0.35415172576904297, | |
| "eval_precision": 0.8448765432098766, | |
| "eval_recall": 0.7540007274049827, | |
| "eval_runtime": 1.8275, | |
| "eval_samples_per_second": 218.333, | |
| "eval_steps_per_second": 27.36, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.0342143774032593, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3188, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8333016825553572, | |
| "eval_loss": 0.32140952348709106, | |
| "eval_precision": 0.8341507249908615, | |
| "eval_recall": 0.8324695399163484, | |
| "eval_runtime": 1.8265, | |
| "eval_samples_per_second": 218.445, | |
| "eval_steps_per_second": 27.374, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.17711341381073, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2981, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.855319904024935, | |
| "eval_loss": 0.2984267771244049, | |
| "eval_precision": 0.862378106322743, | |
| "eval_recall": 0.8491543917075832, | |
| "eval_runtime": 1.8288, | |
| "eval_samples_per_second": 218.18, | |
| "eval_steps_per_second": 27.341, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.153679370880127, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2835, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8368354828562441, | |
| "eval_loss": 0.28104665875434875, | |
| "eval_precision": 0.8520237470480189, | |
| "eval_recall": 0.8252864157119476, | |
| "eval_runtime": 1.8275, | |
| "eval_samples_per_second": 218.335, | |
| "eval_steps_per_second": 27.36, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.0657362937927246, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2517, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8622036668943447, | |
| "eval_loss": 0.28660351037979126, | |
| "eval_precision": 0.8671602787456446, | |
| "eval_recall": 0.8577014002545917, | |
| "eval_runtime": 1.7903, | |
| "eval_samples_per_second": 222.862, | |
| "eval_steps_per_second": 27.928, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.21960711479187, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2374, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8485289465359063, | |
| "eval_loss": 0.2996794879436493, | |
| "eval_precision": 0.8671008040401356, | |
| "eval_recall": 0.8348790689216221, | |
| "eval_runtime": 1.7958, | |
| "eval_samples_per_second": 222.185, | |
| "eval_steps_per_second": 27.843, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.313467025756836, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2293, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8518472677764712, | |
| "eval_loss": 0.29089975357055664, | |
| "eval_precision": 0.8599810186649794, | |
| "eval_recall": 0.844880887434079, | |
| "eval_runtime": 1.7949, | |
| "eval_samples_per_second": 222.299, | |
| "eval_steps_per_second": 27.857, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.5997090339660645, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2091, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8589543987905864, | |
| "eval_loss": 0.29276683926582336, | |
| "eval_precision": 0.8564068100358423, | |
| "eval_recall": 0.8616566648481543, | |
| "eval_runtime": 1.7944, | |
| "eval_samples_per_second": 222.357, | |
| "eval_steps_per_second": 27.864, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 3.3290369510650635, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.198, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8569892473118279, | |
| "eval_loss": 0.28468698263168335, | |
| "eval_precision": 0.8522004241781549, | |
| "eval_recall": 0.8623840698308783, | |
| "eval_runtime": 1.8002, | |
| "eval_samples_per_second": 221.645, | |
| "eval_steps_per_second": 27.775, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 5.568458557128906, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1906, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.843111041207927, | |
| "eval_loss": 0.31200090050697327, | |
| "eval_precision": 0.8585673051692468, | |
| "eval_recall": 0.831332969630842, | |
| "eval_runtime": 1.7936, | |
| "eval_samples_per_second": 222.458, | |
| "eval_steps_per_second": 27.877, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.5475130081176758, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1818, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.850729517396184, | |
| "eval_loss": 0.2906310558319092, | |
| "eval_precision": 0.8535087719298247, | |
| "eval_recall": 0.8481087470449173, | |
| "eval_runtime": 1.795, | |
| "eval_samples_per_second": 222.279, | |
| "eval_steps_per_second": 27.854, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 5.56436014175415, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1756, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8514869535493182, | |
| "eval_loss": 0.28101998567581177, | |
| "eval_precision": 0.8523821128305106, | |
| "eval_recall": 0.8506092016730314, | |
| "eval_runtime": 1.7948, | |
| "eval_samples_per_second": 222.311, | |
| "eval_steps_per_second": 27.859, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.7677656412124634, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.174, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8595070422535211, | |
| "eval_loss": 0.2828814685344696, | |
| "eval_precision": 0.8633733523114054, | |
| "eval_recall": 0.8559283506092017, | |
| "eval_runtime": 1.7987, | |
| "eval_samples_per_second": 221.825, | |
| "eval_steps_per_second": 27.798, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 2.1199889183044434, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1705, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.855319904024935, | |
| "eval_loss": 0.292202889919281, | |
| "eval_precision": 0.862378106322743, | |
| "eval_recall": 0.8491543917075832, | |
| "eval_runtime": 1.7972, | |
| "eval_samples_per_second": 222.011, | |
| "eval_steps_per_second": 27.821, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 2.521127223968506, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1509, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8568221901555235, | |
| "eval_loss": 0.2991296947002411, | |
| "eval_precision": 0.8596491228070176, | |
| "eval_recall": 0.8541553009638116, | |
| "eval_runtime": 1.7965, | |
| "eval_samples_per_second": 222.104, | |
| "eval_steps_per_second": 27.833, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.4539577960968018, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1549, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.855319904024935, | |
| "eval_loss": 0.3000391125679016, | |
| "eval_precision": 0.862378106322743, | |
| "eval_recall": 0.8491543917075832, | |
| "eval_runtime": 1.7991, | |
| "eval_samples_per_second": 221.772, | |
| "eval_steps_per_second": 27.791, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 1.3508776426315308, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1469, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8609292598654301, | |
| "eval_loss": 0.2942558228969574, | |
| "eval_precision": 0.8609292598654301, | |
| "eval_recall": 0.8609292598654301, | |
| "eval_runtime": 1.8067, | |
| "eval_samples_per_second": 220.841, | |
| "eval_steps_per_second": 27.674, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.1654489040374756, | |
| "learning_rate": 0.0, | |
| "loss": 0.1493, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.855319904024935, | |
| "eval_loss": 0.3026413023471832, | |
| "eval_precision": 0.862378106322743, | |
| "eval_recall": 0.8491543917075832, | |
| "eval_runtime": 1.7968, | |
| "eval_samples_per_second": 222.067, | |
| "eval_steps_per_second": 27.828, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8460375995160000.0, | |
| "train_loss": 0.24652244458433056, | |
| "train_runtime": 623.7969, | |
| "train_samples_per_second": 116.865, | |
| "train_steps_per_second": 3.912 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 8460375995160000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |