| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9929078014184397, | |
| "eval_steps": 10, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 2.1404, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_accuracy": 0.629, | |
| "eval_loss": 1.6121153831481934, | |
| "eval_runtime": 10.809, | |
| "eval_samples_per_second": 92.515, | |
| "eval_steps_per_second": 2.96, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 1.3709, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_accuracy": 0.882, | |
| "eval_loss": 0.8628067374229431, | |
| "eval_runtime": 10.9181, | |
| "eval_samples_per_second": 91.591, | |
| "eval_steps_per_second": 2.931, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.8772, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.933, | |
| "eval_loss": 0.5010371208190918, | |
| "eval_runtime": 11.2214, | |
| "eval_samples_per_second": 89.116, | |
| "eval_steps_per_second": 2.852, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.6335, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_accuracy": 0.941, | |
| "eval_loss": 0.33985650539398193, | |
| "eval_runtime": 11.3617, | |
| "eval_samples_per_second": 88.015, | |
| "eval_steps_per_second": 2.816, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.5387, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_accuracy": 0.957, | |
| "eval_loss": 0.2673340439796448, | |
| "eval_runtime": 11.4297, | |
| "eval_samples_per_second": 87.492, | |
| "eval_steps_per_second": 2.8, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.4319, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_accuracy": 0.958, | |
| "eval_loss": 0.23862411081790924, | |
| "eval_runtime": 11.48, | |
| "eval_samples_per_second": 87.108, | |
| "eval_steps_per_second": 2.787, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.0, | |
| "loss": 0.4591, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.961, | |
| "eval_loss": 0.22941315174102783, | |
| "eval_runtime": 11.5406, | |
| "eval_samples_per_second": 86.651, | |
| "eval_steps_per_second": 2.773, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "step": 70, | |
| "total_flos": 6.943780113009869e+17, | |
| "train_loss": 0.9216747965131488, | |
| "train_runtime": 351.6178, | |
| "train_samples_per_second": 25.596, | |
| "train_steps_per_second": 0.199 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 70, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 6.943780113009869e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |