| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2694300518134715, | |
| "eval_steps": 10, | |
| "global_step": 26, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010362694300518135, | |
| "eval_loss": 3.649900197982788, | |
| "eval_runtime": 0.7545, | |
| "eval_samples_per_second": 54.343, | |
| "eval_steps_per_second": 27.834, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05181347150259067, | |
| "grad_norm": 2.8373703956604004, | |
| "learning_rate": 5e-05, | |
| "loss": 3.5652, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.10362694300518134, | |
| "grad_norm": 2.706880569458008, | |
| "learning_rate": 0.0001, | |
| "loss": 3.4354, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10362694300518134, | |
| "eval_loss": 3.3804771900177, | |
| "eval_runtime": 0.7498, | |
| "eval_samples_per_second": 54.684, | |
| "eval_steps_per_second": 28.009, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15544041450777202, | |
| "grad_norm": 1.4844454526901245, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 3.2829, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.20725388601036268, | |
| "grad_norm": 1.2455134391784668, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 3.0929, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.20725388601036268, | |
| "eval_loss": 3.0835258960723877, | |
| "eval_runtime": 0.7639, | |
| "eval_samples_per_second": 53.673, | |
| "eval_steps_per_second": 27.491, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.25906735751295334, | |
| "grad_norm": 0.8904907703399658, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 3.0248, | |
| "step": 25 | |
| } | |
| ], | |
| "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": 89160891236352.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |