| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.9981670994348555, | |
| "global_step": 3272, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.246638141809291e-05, | |
| "loss": 0.706, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.7039228677749634, | |
| "eval_runtime": 30.7626, | |
| "eval_samples_per_second": 177.586, | |
| "eval_steps_per_second": 5.559, | |
| "step": 818 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 3.4825794621026896e-05, | |
| "loss": 0.7042, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 2.718520782396088e-05, | |
| "loss": 0.7032, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6934332847595215, | |
| "eval_runtime": 30.7656, | |
| "eval_samples_per_second": 177.569, | |
| "eval_steps_per_second": 5.558, | |
| "step": 1636 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 1.9544621026894865e-05, | |
| "loss": 0.7031, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6931120753288269, | |
| "eval_runtime": 30.7565, | |
| "eval_samples_per_second": 177.621, | |
| "eval_steps_per_second": 5.56, | |
| "step": 2455 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 1.1904034229828852e-05, | |
| "loss": 0.7027, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 4.263447432762836e-06, | |
| "loss": 0.7021, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931692361831665, | |
| "eval_runtime": 30.7781, | |
| "eval_samples_per_second": 177.497, | |
| "eval_steps_per_second": 5.556, | |
| "step": 3272 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 3272, | |
| "total_flos": 3.902826037167698e+17, | |
| "train_loss": 0.7034443619781718, | |
| "train_runtime": 7284.2364, | |
| "train_samples_per_second": 57.518, | |
| "train_steps_per_second": 0.449 | |
| } | |
| ], | |
| "max_steps": 3272, | |
| "num_train_epochs": 4, | |
| "total_flos": 3.902826037167698e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |