| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0009079269002444419, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.3280176929344664e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 80.0382, | |
| "eval_samples_per_second": 225.98, | |
| "eval_steps_per_second": 112.996, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00011640088464672332, | |
| "grad_norm": 1.1555752754211426, | |
| "learning_rate": 5e-05, | |
| "loss": 81.231, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00023280176929344664, | |
| "grad_norm": 1.4119420051574707, | |
| "learning_rate": 0.0001, | |
| "loss": 70.7834, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00023280176929344664, | |
| "eval_loss": NaN, | |
| "eval_runtime": 78.0869, | |
| "eval_samples_per_second": 231.626, | |
| "eval_steps_per_second": 115.82, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0003492026539401699, | |
| "grad_norm": 1.923086166381836, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 65.3133, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00046560353858689327, | |
| "grad_norm": 1.532097339630127, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 75.6257, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00046560353858689327, | |
| "eval_loss": NaN, | |
| "eval_runtime": 78.9205, | |
| "eval_samples_per_second": 229.18, | |
| "eval_steps_per_second": 114.596, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0005820044232336166, | |
| "grad_norm": 1.3649100065231323, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 62.8437, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0006984053078803399, | |
| "grad_norm": 1.821262001991272, | |
| "learning_rate": 5e-05, | |
| "loss": 84.6001, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0006984053078803399, | |
| "eval_loss": NaN, | |
| "eval_runtime": 79.1527, | |
| "eval_samples_per_second": 228.508, | |
| "eval_steps_per_second": 114.26, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0008148061925270633, | |
| "grad_norm": 1.472610592842102, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 52.4185, | |
| "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": 30257971200.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |