| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0081662651586297, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00016332530317259402, | |
| "eval_loss": 2.5971131324768066, | |
| "eval_runtime": 99.1924, | |
| "eval_samples_per_second": 25.99, | |
| "eval_steps_per_second": 12.995, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016332530317259401, | |
| "grad_norm": 1.7391607761383057, | |
| "learning_rate": 0.0002, | |
| "loss": 9.9966, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0021232289412437223, | |
| "eval_loss": 2.5043554306030273, | |
| "eval_runtime": 99.143, | |
| "eval_samples_per_second": 26.003, | |
| "eval_steps_per_second": 13.001, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0032665060634518802, | |
| "grad_norm": 3.42970871925354, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 9.7219, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.004246457882487445, | |
| "eval_loss": 2.3963499069213867, | |
| "eval_runtime": 99.1073, | |
| "eval_samples_per_second": 26.012, | |
| "eval_steps_per_second": 13.006, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0048997590951778205, | |
| "grad_norm": 2.8426146507263184, | |
| "learning_rate": 0.0001, | |
| "loss": 9.7187, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.006369686823731167, | |
| "eval_loss": 2.354743003845215, | |
| "eval_runtime": 99.1756, | |
| "eval_samples_per_second": 25.994, | |
| "eval_steps_per_second": 12.997, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0065330121269037604, | |
| "grad_norm": 3.3102850914001465, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 9.8154, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0081662651586297, | |
| "grad_norm": 5.638280391693115, | |
| "learning_rate": 0.0, | |
| "loss": 9.3362, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9080277024374784.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |