| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02087246921310791, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0004174493842621582, | |
| "eval_loss": 3.6245229244232178, | |
| "eval_runtime": 18.065, | |
| "eval_samples_per_second": 55.854, | |
| "eval_steps_per_second": 27.955, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004174493842621582, | |
| "grad_norm": 1.1807235479354858, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 3.129, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.005426841995408057, | |
| "eval_loss": 3.0532212257385254, | |
| "eval_runtime": 16.8979, | |
| "eval_samples_per_second": 59.712, | |
| "eval_steps_per_second": 29.885, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.008348987685243165, | |
| "grad_norm": 1.8960473537445068, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 2.8603, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.010853683990816114, | |
| "eval_loss": 2.9474551677703857, | |
| "eval_runtime": 16.936, | |
| "eval_samples_per_second": 59.577, | |
| "eval_steps_per_second": 29.818, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.012523481527864746, | |
| "grad_norm": 1.0281912088394165, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 2.8738, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01628052598622417, | |
| "eval_loss": 2.912461757659912, | |
| "eval_runtime": 16.7721, | |
| "eval_samples_per_second": 60.159, | |
| "eval_steps_per_second": 30.11, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.01669797537048633, | |
| "grad_norm": 1.1397675275802612, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 2.8672, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02087246921310791, | |
| "grad_norm": 1.4951775074005127, | |
| "learning_rate": 0.0, | |
| "loss": 2.8104, | |
| "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": 467450359971840.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |