| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.230769230769231, | |
| "eval_steps": 220, | |
| "global_step": 440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00018906605922551254, | |
| "loss": 1.1294, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00017767653758542142, | |
| "loss": 1.0356, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0001662870159453303, | |
| "loss": 1.0107, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.0001548974943052392, | |
| "loss": 1.1191, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00014350797266514807, | |
| "loss": 0.7423, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00013211845102505695, | |
| "loss": 0.6658, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.00012072892938496582, | |
| "loss": 0.5811, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00010933940774487472, | |
| "loss": 0.5556, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_loss": 0.3080795109272003, | |
| "eval_runtime": 199.9175, | |
| "eval_samples_per_second": 1.035, | |
| "eval_steps_per_second": 0.13, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 9.79498861047836e-05, | |
| "loss": 0.3572, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 8.656036446469249e-05, | |
| "loss": 0.2907, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 7.517084282460137e-05, | |
| "loss": 0.3575, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 6.378132118451025e-05, | |
| "loss": 0.3076, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 5.239179954441914e-05, | |
| "loss": 0.2052, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 4.100227790432802e-05, | |
| "loss": 0.137, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 2.96127562642369e-05, | |
| "loss": 0.1223, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 1.8223234624145787e-05, | |
| "loss": 0.1175, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 6.83371298405467e-06, | |
| "loss": 0.1137, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "eval_loss": 0.052269067615270615, | |
| "eval_runtime": 199.3757, | |
| "eval_samples_per_second": 1.038, | |
| "eval_steps_per_second": 0.13, | |
| "step": 440 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 220, | |
| "total_flos": 1.5491320119086285e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |