| { | |
| "best_metric": 0.3455743193626404, | |
| "best_model_checkpoint": "outputs/t5-v1_1-base-detox/checkpoint-540", | |
| "epoch": 3.9953917050691246, | |
| "global_step": 540, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9e-05, | |
| "loss": 8.9719, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.031940221786499, | |
| "eval_runtime": 22.7196, | |
| "eval_samples_per_second": 136.446, | |
| "eval_steps_per_second": 17.078, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 8e-05, | |
| "loss": 1.9438, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.6424035429954529, | |
| "eval_runtime": 22.8507, | |
| "eval_samples_per_second": 135.663, | |
| "eval_steps_per_second": 16.98, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7e-05, | |
| "loss": 0.8941, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.43842175602912903, | |
| "eval_runtime": 22.8845, | |
| "eval_samples_per_second": 135.463, | |
| "eval_steps_per_second": 16.955, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 6e-05, | |
| "loss": 0.6694, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.3455743193626404, | |
| "eval_runtime": 22.9242, | |
| "eval_samples_per_second": 135.228, | |
| "eval_steps_per_second": 16.925, | |
| "step": 540 | |
| } | |
| ], | |
| "max_steps": 1350, | |
| "num_train_epochs": 10, | |
| "total_flos": 5262887053578240.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |