| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 18, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 7.8279015056079535, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1218, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 11.864680166140012, | |
| "learning_rate": 1.8314696123025456e-05, | |
| "loss": 1.3958, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.6723018884658813, | |
| "eval_runtime": 4.1855, | |
| "eval_samples_per_second": 10.512, | |
| "eval_steps_per_second": 0.717, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 5.32505230615824, | |
| "learning_rate": 1e-05, | |
| "loss": 1.0573, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.5254077911376953, | |
| "eval_runtime": 3.9837, | |
| "eval_samples_per_second": 11.045, | |
| "eval_steps_per_second": 0.753, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 3.989613773868231, | |
| "learning_rate": 1.6853038769745466e-06, | |
| "loss": 0.8462, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.5098522901535034, | |
| "eval_runtime": 3.9707, | |
| "eval_samples_per_second": 11.081, | |
| "eval_steps_per_second": 0.756, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 18, | |
| "total_flos": 3716488888320.0, | |
| "train_loss": 1.0306008060773213, | |
| "train_runtime": 321.5114, | |
| "train_samples_per_second": 3.546, | |
| "train_steps_per_second": 0.056 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 18, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 3716488888320.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |