| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 375, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 2.1940720081329346, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0435, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 2.7920234203338623, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6718, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 2.9994211196899414, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5008, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 7.913583755493164, | |
| "learning_rate": 2e-05, | |
| "loss": 0.4437, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 3.560945987701416, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3628, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 6.318229675292969, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3767, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 2.496079444885254, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3398, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 4.720012664794922, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3219, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 4.714549541473389, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3112, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 4.124231338500977, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3029, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 5.186863899230957, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2997, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 4.2964582443237305, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2656, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8880294192239412, | |
| "eval_f1": 0.8036882807852468, | |
| "eval_loss": 0.3132246732711792, | |
| "eval_precision": 0.7543271915131211, | |
| "eval_recall": 0.8599618077657543, | |
| "eval_runtime": 0.4252, | |
| "eval_samples_per_second": 352.805, | |
| "eval_steps_per_second": 23.52, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 3.693039655685425, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2821, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 5.225029468536377, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2464, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 7.333633899688721, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2913, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 3.361424446105957, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2501, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 3.7213943004608154, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2563, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 4.98430061340332, | |
| "learning_rate": 2e-05, | |
| "loss": 0.234, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 5.262845993041992, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2414, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 6.322588920593262, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2807, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 3.358001470565796, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2247, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 3.0363667011260986, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2259, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 3.240709066390991, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2361, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 3.994675874710083, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2121, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.988025665283203, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2175, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.898173979203652, | |
| "eval_f1": 0.8382264829239066, | |
| "eval_loss": 0.2903146743774414, | |
| "eval_precision": 0.7917374080362196, | |
| "eval_recall": 0.890515595162317, | |
| "eval_runtime": 0.4109, | |
| "eval_samples_per_second": 365.087, | |
| "eval_steps_per_second": 24.339, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 2.051712989807129, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1549, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 4.846020698547363, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2409, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 2.303395986557007, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1975, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 3.298722743988037, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1776, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 2.484005928039551, | |
| "learning_rate": 2e-05, | |
| "loss": 0.189, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 3.962454319000244, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1937, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 4.414576053619385, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1799, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 4.953629970550537, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2084, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 4.017173767089844, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1637, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 3.566938638687134, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2825, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 4.922001838684082, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1963, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 5.1235270500183105, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2167, | |
| "step": 370 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 375, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 473453787413376.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |