| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06320907617504051, | |
| "eval_steps": 13, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0016207455429497568, | |
| "eval_loss": 3.8959412574768066, | |
| "eval_runtime": 5.4702, | |
| "eval_samples_per_second": 47.53, | |
| "eval_steps_per_second": 23.765, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01620745542949757, | |
| "grad_norm": 0.8758059144020081, | |
| "learning_rate": 0.0002, | |
| "loss": 3.4861, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02106969205834684, | |
| "eval_loss": 3.616358757019043, | |
| "eval_runtime": 4.4298, | |
| "eval_samples_per_second": 58.693, | |
| "eval_steps_per_second": 29.347, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.03241491085899514, | |
| "grad_norm": 1.264078140258789, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 3.47, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04213938411669368, | |
| "eval_loss": 3.4697425365448, | |
| "eval_runtime": 4.4531, | |
| "eval_samples_per_second": 58.387, | |
| "eval_steps_per_second": 29.193, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.04862236628849271, | |
| "grad_norm": 0.9814926981925964, | |
| "learning_rate": 0.0001, | |
| "loss": 3.2828, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06320907617504051, | |
| "eval_loss": 3.4281005859375, | |
| "eval_runtime": 4.4556, | |
| "eval_samples_per_second": 58.353, | |
| "eval_steps_per_second": 29.177, | |
| "step": 39 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 352929491976192.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |