| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 500.0, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 66.67, | |
| "eval_loss": 3.64453125, | |
| "eval_runtime": 2.3703, | |
| "eval_samples_per_second": 3.797, | |
| "eval_steps_per_second": 0.844, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 133.33, | |
| "eval_loss": 4.5703125, | |
| "eval_runtime": 2.3618, | |
| "eval_samples_per_second": 3.811, | |
| "eval_steps_per_second": 0.847, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 166.67, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0101, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "eval_loss": 5.2109375, | |
| "eval_runtime": 2.3656, | |
| "eval_samples_per_second": 3.805, | |
| "eval_steps_per_second": 0.845, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 266.67, | |
| "eval_loss": 5.54296875, | |
| "eval_runtime": 2.3648, | |
| "eval_samples_per_second": 3.806, | |
| "eval_steps_per_second": 0.846, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 333.33, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0681, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 333.33, | |
| "eval_loss": 5.72265625, | |
| "eval_runtime": 2.3619, | |
| "eval_samples_per_second": 3.81, | |
| "eval_steps_per_second": 0.847, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "eval_loss": 5.8671875, | |
| "eval_runtime": 2.3493, | |
| "eval_samples_per_second": 3.831, | |
| "eval_steps_per_second": 0.851, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 466.67, | |
| "eval_loss": 5.99609375, | |
| "eval_runtime": 2.3621, | |
| "eval_samples_per_second": 3.81, | |
| "eval_steps_per_second": 0.847, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0193, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "step": 1500, | |
| "total_flos": 3004912665362432.0, | |
| "train_loss": 0.36582159678141274, | |
| "train_runtime": 9345.1631, | |
| "train_samples_per_second": 1.231, | |
| "train_steps_per_second": 0.161 | |
| } | |
| ], | |
| "max_steps": 1500, | |
| "num_train_epochs": 500, | |
| "total_flos": 3004912665362432.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |