| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 5625, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 3.5154, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.146, | |
| "eval_f1": 0.05673406159055716, | |
| "eval_loss": 3.445145606994629, | |
| "eval_runtime": 16.8588, | |
| "eval_samples_per_second": 59.316, | |
| "eval_steps_per_second": 7.415, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.2e-05, | |
| "loss": 3.2778, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.197, | |
| "eval_f1": 0.09626799246503706, | |
| "eval_loss": 3.1801486015319824, | |
| "eval_runtime": 16.85, | |
| "eval_samples_per_second": 59.347, | |
| "eval_steps_per_second": 7.418, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 3.0802, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.174, | |
| "eval_f1": 0.11203679919237455, | |
| "eval_loss": 3.1373538970947266, | |
| "eval_runtime": 16.8552, | |
| "eval_samples_per_second": 59.329, | |
| "eval_steps_per_second": 7.416, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.8756, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.191, | |
| "eval_f1": 0.1229927332781174, | |
| "eval_loss": 3.123941421508789, | |
| "eval_runtime": 16.904, | |
| "eval_samples_per_second": 59.158, | |
| "eval_steps_per_second": 7.395, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.6657, | |
| "step": 5625 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.196, | |
| "eval_f1": 0.1399460092685649, | |
| "eval_loss": 3.178866147994995, | |
| "eval_runtime": 16.8498, | |
| "eval_samples_per_second": 59.348, | |
| "eval_steps_per_second": 7.419, | |
| "step": 5625 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 5625, | |
| "total_flos": 4.030110203721907e+16, | |
| "train_loss": 3.0829483940972224, | |
| "train_runtime": 3366.8478, | |
| "train_samples_per_second": 13.366, | |
| "train_steps_per_second": 1.671 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5625, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 4.030110203721907e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |