| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 3750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 4.658291339874268, | |
| "learning_rate": 2.8703181864639013e-06, | |
| "loss": 0.64, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 4.34930944442749, | |
| "learning_rate": 2.5036959095382875e-06, | |
| "loss": 0.6413, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6420477032661438, | |
| "eval_runtime": 20.5136, | |
| "eval_samples_per_second": 97.497, | |
| "eval_steps_per_second": 12.187, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 5.983999252319336, | |
| "learning_rate": 1.963525491562421e-06, | |
| "loss": 0.4847, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 5.177482604980469, | |
| "learning_rate": 1.3432073050985201e-06, | |
| "loss": 0.3344, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.585903167724609, | |
| "learning_rate": 7.500000000000003e-07, | |
| "loss": 0.3318, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7324458360671997, | |
| "eval_runtime": 20.5637, | |
| "eval_samples_per_second": 97.259, | |
| "eval_steps_per_second": 12.157, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 5.989898681640625, | |
| "learning_rate": 2.86474508437579e-07, | |
| "loss": 0.1551, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 5.329368591308594, | |
| "learning_rate": 3.277859889929147e-08, | |
| "loss": 0.1518, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.994577169418335, | |
| "eval_runtime": 20.6046, | |
| "eval_samples_per_second": 97.066, | |
| "eval_steps_per_second": 12.133, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 3750, | |
| "total_flos": 2.3924465493816115e+17, | |
| "train_loss": 0.37510518595377607, | |
| "train_runtime": 2372.5928, | |
| "train_samples_per_second": 12.644, | |
| "train_steps_per_second": 1.581 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3750, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.3924465493816115e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |