| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 2862, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5887096774193549, | |
| "eval_loss": 0.19886387884616852, | |
| "eval_runtime": 5.4179, | |
| "eval_samples_per_second": 572.175, | |
| "eval_steps_per_second": 11.997, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.5294517874717712, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 0.3203, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8193548387096774, | |
| "eval_loss": 0.09618715196847916, | |
| "eval_runtime": 6.0744, | |
| "eval_samples_per_second": 510.34, | |
| "eval_steps_per_second": 10.701, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8783870967741936, | |
| "eval_loss": 0.06421981006860733, | |
| "eval_runtime": 5.5189, | |
| "eval_samples_per_second": 561.707, | |
| "eval_steps_per_second": 11.778, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.44660821557044983, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 0.1116, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8996774193548387, | |
| "eval_loss": 0.04962550476193428, | |
| "eval_runtime": 5.5318, | |
| "eval_samples_per_second": 560.394, | |
| "eval_steps_per_second": 11.75, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.3033629357814789, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.0713, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9080645161290323, | |
| "eval_loss": 0.040737707167863846, | |
| "eval_runtime": 6.4166, | |
| "eval_samples_per_second": 483.123, | |
| "eval_steps_per_second": 10.13, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9116129032258065, | |
| "eval_loss": 0.0357489213347435, | |
| "eval_runtime": 5.4265, | |
| "eval_samples_per_second": 571.272, | |
| "eval_steps_per_second": 11.978, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.26608094573020935, | |
| "learning_rate": 6.02375960866527e-06, | |
| "loss": 0.0561, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9187096774193548, | |
| "eval_loss": 0.03307666257023811, | |
| "eval_runtime": 6.4951, | |
| "eval_samples_per_second": 477.285, | |
| "eval_steps_per_second": 10.008, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.861635220125786, | |
| "grad_norm": 0.27350321412086487, | |
| "learning_rate": 2.5296995108315863e-06, | |
| "loss": 0.0492, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9241935483870968, | |
| "eval_loss": 0.031220456585288048, | |
| "eval_runtime": 7.4945, | |
| "eval_samples_per_second": 413.636, | |
| "eval_steps_per_second": 8.673, | |
| "step": 2544 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2862, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "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": 651155886807636.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.2686953507184725, | |
| "num_train_epochs": 9, | |
| "temperature": 11 | |
| } | |
| } | |