| { | |
| "best_metric": 1.0, | |
| "best_model_checkpoint": "model/checkpoint-30", | |
| "epoch": 7.466666666666667, | |
| "eval_steps": 500, | |
| "global_step": 56, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.93, | |
| "eval_accuracy": 0.6190476190476191, | |
| "eval_loss": 0.5468170046806335, | |
| "eval_runtime": 12.9206, | |
| "eval_samples_per_second": 8.127, | |
| "eval_steps_per_second": 0.31, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.00048, | |
| "loss": 1.0237, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9619047619047619, | |
| "eval_loss": 0.11131703853607178, | |
| "eval_runtime": 12.3741, | |
| "eval_samples_per_second": 8.485, | |
| "eval_steps_per_second": 0.323, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.0004, | |
| "loss": 0.5329, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_accuracy": 0.9714285714285714, | |
| "eval_loss": 0.046230871230363846, | |
| "eval_runtime": 12.3758, | |
| "eval_samples_per_second": 8.484, | |
| "eval_steps_per_second": 0.323, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.00032, | |
| "loss": 0.335, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.015091076493263245, | |
| "eval_runtime": 12.4003, | |
| "eval_samples_per_second": 8.468, | |
| "eval_steps_per_second": 0.323, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 0.00024, | |
| "loss": 0.3359, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "eval_accuracy": 0.9428571428571428, | |
| "eval_loss": 0.1429169476032257, | |
| "eval_runtime": 12.3608, | |
| "eval_samples_per_second": 8.495, | |
| "eval_steps_per_second": 0.324, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 0.00016, | |
| "loss": 0.2794, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9619047619047619, | |
| "eval_loss": 0.060580089688301086, | |
| "eval_runtime": 12.4196, | |
| "eval_samples_per_second": 8.454, | |
| "eval_steps_per_second": 0.322, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 8e-05, | |
| "loss": 0.2937, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "eval_accuracy": 0.9904761904761905, | |
| "eval_loss": 0.01593722589313984, | |
| "eval_runtime": 12.3709, | |
| "eval_samples_per_second": 8.488, | |
| "eval_steps_per_second": 0.323, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "learning_rate": 0.0, | |
| "loss": 0.249, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "eval_accuracy": 0.9904761904761905, | |
| "eval_loss": 0.01840730383992195, | |
| "eval_runtime": 12.4229, | |
| "eval_samples_per_second": 8.452, | |
| "eval_steps_per_second": 0.322, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "step": 56, | |
| "total_flos": 5.438387167052636e+17, | |
| "train_loss": 0.4356716935123716, | |
| "train_runtime": 2360.1881, | |
| "train_samples_per_second": 3.183, | |
| "train_steps_per_second": 0.024 | |
| } | |
| ], | |
| "logging_steps": 8, | |
| "max_steps": 56, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "total_flos": 5.438387167052636e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |