| { | |
| "best_metric": 0.4466552734375, | |
| "best_model_checkpoint": "/tmp/model/checkpoint-12", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 12, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.5942, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6104, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.8108, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.2256, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.42857142857142855, | |
| "eval_auc": 1.0, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6931501030921936, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 0.1781, | |
| "eval_samples_per_second": 39.307, | |
| "eval_steps_per_second": 5.615, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5964, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4e-05, | |
| "loss": 0.633, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.6581, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.4944, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5714285714285714, | |
| "eval_auc": 1.0, | |
| "eval_f1": 0.4, | |
| "eval_loss": 0.5654993653297424, | |
| "eval_precision": 1.0, | |
| "eval_recall": 0.25, | |
| "eval_runtime": 0.1786, | |
| "eval_samples_per_second": 39.192, | |
| "eval_steps_per_second": 5.599, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 3e-05, | |
| "loss": 0.429, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.4291, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5811, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.637, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 1.0, | |
| "eval_auc": 1.0, | |
| "eval_f1": 1.0, | |
| "eval_loss": 0.4466552734375, | |
| "eval_precision": 1.0, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 0.1756, | |
| "eval_samples_per_second": 39.861, | |
| "eval_steps_per_second": 5.694, | |
| "step": 12 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 12, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 6276851158781952.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |