| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.526315789473684, | |
| "eval_steps": 500, | |
| "global_step": 12, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 7.687494915493854, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.4116, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 1.418346881866455, | |
| "eval_runtime": 5.3559, | |
| "eval_samples_per_second": 11.949, | |
| "eval_steps_per_second": 0.747, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 4.445203660946351, | |
| "learning_rate": 3.969463130731183e-06, | |
| "loss": 1.3604, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 1.2682572603225708, | |
| "eval_runtime": 5.3303, | |
| "eval_samples_per_second": 12.007, | |
| "eval_steps_per_second": 0.75, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 2.4792286423157206, | |
| "learning_rate": 4.774575140626317e-07, | |
| "loss": 1.1532, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 1.2627166509628296, | |
| "eval_runtime": 5.5111, | |
| "eval_samples_per_second": 11.613, | |
| "eval_steps_per_second": 0.726, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "step": 12, | |
| "total_flos": 2460210954240.0, | |
| "train_loss": 1.2326995531717937, | |
| "train_runtime": 212.4422, | |
| "train_samples_per_second": 4.166, | |
| "train_steps_per_second": 0.056 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 12, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2460210954240.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |