| { | |
| "best_metric": 2.451620578765869, | |
| "best_model_checkpoint": "./outputs/checkpoint-700", | |
| "epoch": 0.5100182149362478, | |
| "eval_steps": 100, | |
| "global_step": 700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.7401, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.6424810886383057, | |
| "eval_runtime": 206.4108, | |
| "eval_samples_per_second": 30.396, | |
| "eval_steps_per_second": 3.803, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6061, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.5929574966430664, | |
| "eval_runtime": 206.2177, | |
| "eval_samples_per_second": 30.424, | |
| "eval_steps_per_second": 3.807, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5643, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.5578970909118652, | |
| "eval_runtime": 206.509, | |
| "eval_samples_per_second": 30.381, | |
| "eval_steps_per_second": 3.801, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5383, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 2.5257716178894043, | |
| "eval_runtime": 206.1349, | |
| "eval_samples_per_second": 30.436, | |
| "eval_steps_per_second": 3.808, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4959, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.5014102458953857, | |
| "eval_runtime": 282.6461, | |
| "eval_samples_per_second": 22.197, | |
| "eval_steps_per_second": 2.777, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4693, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 2.475057601928711, | |
| "eval_runtime": 205.7759, | |
| "eval_samples_per_second": 30.489, | |
| "eval_steps_per_second": 3.815, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4523, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.451620578765869, | |
| "eval_runtime": 322.0554, | |
| "eval_samples_per_second": 19.481, | |
| "eval_steps_per_second": 2.437, | |
| "step": 700 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 2.042513200078848e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |