| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.7906976744186047, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 9363.349507841189, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 30.9126, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 556.2728379648993, | |
| "learning_rate": 1.973044870579824e-05, | |
| "loss": 21.2307, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 559.6918657048097, | |
| "learning_rate": 1.686241637868734e-05, | |
| "loss": 25.9559, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 375.2686916285916, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 18.4454, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 170.37327814160207, | |
| "learning_rate": 6.039202339608432e-06, | |
| "loss": 16.4221, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 161.47537049884457, | |
| "learning_rate": 1.6451218858706374e-06, | |
| "loss": 15.3885, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 138.06585339598334, | |
| "learning_rate": 0.0, | |
| "loss": 14.5343, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "step": 30, | |
| "total_flos": 8193589641216.0, | |
| "train_loss": 18.98553034464518, | |
| "train_runtime": 179.3227, | |
| "train_samples_per_second": 22.819, | |
| "train_steps_per_second": 0.167 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 8193589641216.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |