| { | |
| "best_metric": 0.7328296703296704, | |
| "best_model_checkpoint": "./priority_model\\checkpoint-13", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 52, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 4.372110366821289, | |
| "learning_rate": 1.7435897435897438e-05, | |
| "loss": 1.0904, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7307692307692307, | |
| "eval_f1": 0.7328296703296704, | |
| "eval_loss": 0.8845496773719788, | |
| "eval_runtime": 0.065, | |
| "eval_samples_per_second": 799.398, | |
| "eval_steps_per_second": 61.492, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 4.574769973754883, | |
| "learning_rate": 1.4871794871794874e-05, | |
| "loss": 0.8551, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6730769230769231, | |
| "eval_f1": 0.6639957264957265, | |
| "eval_loss": 0.777785062789917, | |
| "eval_runtime": 0.062, | |
| "eval_samples_per_second": 838.712, | |
| "eval_steps_per_second": 64.516, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 4.142388820648193, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.6724, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7307692307692307, | |
| "eval_f1": 0.7262200165425973, | |
| "eval_loss": 0.6311762928962708, | |
| "eval_runtime": 0.062, | |
| "eval_samples_per_second": 838.709, | |
| "eval_steps_per_second": 64.516, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 4.405705451965332, | |
| "learning_rate": 9.743589743589744e-06, | |
| "loss": 0.5529, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 4.022765159606934, | |
| "learning_rate": 7.17948717948718e-06, | |
| "loss": 0.484, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7307692307692307, | |
| "eval_f1": 0.7262200165425973, | |
| "eval_loss": 0.60198974609375, | |
| "eval_runtime": 0.063, | |
| "eval_samples_per_second": 825.397, | |
| "eval_steps_per_second": 63.492, | |
| "step": 52 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 78, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 38295335720880.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |