| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1822429906542056, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004672897196261682, | |
| "eval_loss": 2.2308504581451416, | |
| "eval_runtime": 7.5828, | |
| "eval_samples_per_second": 12.001, | |
| "eval_steps_per_second": 6.066, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02336448598130841, | |
| "grad_norm": 1.5134570598602295, | |
| "learning_rate": 5e-05, | |
| "loss": 2.2188, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04672897196261682, | |
| "grad_norm": 2.049241542816162, | |
| "learning_rate": 0.0001, | |
| "loss": 2.1443, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04672897196261682, | |
| "eval_loss": 1.7166945934295654, | |
| "eval_runtime": 7.8328, | |
| "eval_samples_per_second": 11.618, | |
| "eval_steps_per_second": 5.873, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07009345794392523, | |
| "grad_norm": 2.114860773086548, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.5615, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09345794392523364, | |
| "grad_norm": 1.852180004119873, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.4648, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09345794392523364, | |
| "eval_loss": 1.411849021911621, | |
| "eval_runtime": 7.9752, | |
| "eval_samples_per_second": 11.41, | |
| "eval_steps_per_second": 5.768, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11682242990654206, | |
| "grad_norm": 1.8431528806686401, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.3921, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.14018691588785046, | |
| "grad_norm": 1.7438124418258667, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3751, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.14018691588785046, | |
| "eval_loss": 1.3371673822402954, | |
| "eval_runtime": 8.1135, | |
| "eval_samples_per_second": 11.216, | |
| "eval_steps_per_second": 5.67, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16355140186915887, | |
| "grad_norm": 1.98276686668396, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.3656, | |
| "step": 35 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7213300567769088.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |