| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "global_step": 2655, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6506, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6680805938494168, | |
| "eval_loss": 0.6074808239936829, | |
| "eval_runtime": 3.0467, | |
| "eval_samples_per_second": 309.518, | |
| "eval_steps_per_second": 9.847, | |
| "step": 531 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 5e-05, | |
| "loss": 0.575, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6977730646871686, | |
| "eval_loss": 0.5816474556922913, | |
| "eval_runtime": 3.0782, | |
| "eval_samples_per_second": 306.344, | |
| "eval_steps_per_second": 9.746, | |
| "step": 1062 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 3.489425981873112e-05, | |
| "loss": 0.4397, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7253446447507953, | |
| "eval_loss": 0.6136603951454163, | |
| "eval_runtime": 3.1046, | |
| "eval_samples_per_second": 303.739, | |
| "eval_steps_per_second": 9.663, | |
| "step": 1593 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 1.9788519637462235e-05, | |
| "loss": 0.2524, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7465535524920467, | |
| "eval_loss": 0.8123661279678345, | |
| "eval_runtime": 3.1044, | |
| "eval_samples_per_second": 303.766, | |
| "eval_steps_per_second": 9.664, | |
| "step": 2124 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 4.682779456193353e-06, | |
| "loss": 0.126, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7370095440084835, | |
| "eval_loss": 1.1437411308288574, | |
| "eval_runtime": 3.1169, | |
| "eval_samples_per_second": 302.541, | |
| "eval_steps_per_second": 9.625, | |
| "step": 2655 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 2655, | |
| "total_flos": 4866474386987520.0, | |
| "train_loss": 0.3910632814166685, | |
| "train_runtime": 432.5593, | |
| "train_samples_per_second": 98.067, | |
| "train_steps_per_second": 6.138 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7370095440084835, | |
| "eval_loss": 1.1437411308288574, | |
| "eval_runtime": 3.1329, | |
| "eval_samples_per_second": 301.0, | |
| "eval_steps_per_second": 9.576, | |
| "step": 2655 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7314984709480122, | |
| "eval_loss": 1.2070553302764893, | |
| "eval_runtime": 10.6939, | |
| "eval_samples_per_second": 305.782, | |
| "eval_steps_per_second": 9.632, | |
| "step": 2655 | |
| } | |
| ], | |
| "max_steps": 2655, | |
| "num_train_epochs": 5, | |
| "total_flos": 4866474386987520.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |