| { | |
| "best_metric": 2.4525136947631836, | |
| "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": 2.6424834728240967, | |
| "eval_runtime": 206.3704, | |
| "eval_samples_per_second": 30.402, | |
| "eval_steps_per_second": 3.804, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6053, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.592982053756714, | |
| "eval_runtime": 206.2608, | |
| "eval_samples_per_second": 30.418, | |
| "eval_steps_per_second": 3.806, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5637, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.558687448501587, | |
| "eval_runtime": 206.6302, | |
| "eval_samples_per_second": 30.363, | |
| "eval_steps_per_second": 3.799, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5384, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 2.5261261463165283, | |
| "eval_runtime": 206.3988, | |
| "eval_samples_per_second": 30.397, | |
| "eval_steps_per_second": 3.803, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4965, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.5020904541015625, | |
| "eval_runtime": 273.7006, | |
| "eval_samples_per_second": 22.923, | |
| "eval_steps_per_second": 2.868, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4706, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 2.474956512451172, | |
| "eval_runtime": 206.4725, | |
| "eval_samples_per_second": 30.387, | |
| "eval_steps_per_second": 3.802, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4542, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.4525136947631836, | |
| "eval_runtime": 307.0185, | |
| "eval_samples_per_second": 20.435, | |
| "eval_steps_per_second": 2.557, | |
| "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 | |
| } | |