| { | |
| "epoch": 1.0, | |
| "eval_dpo_losses": 0.6846575736999512, | |
| "eval_logits/chosen": -2.76676082611084, | |
| "eval_logits/rejected": -2.728935718536377, | |
| "eval_logps/chosen": -282.11871337890625, | |
| "eval_logps/rejected": -257.9208679199219, | |
| "eval_loss": 0.7750731110572815, | |
| "eval_positive_losses": 0.8595818281173706, | |
| "eval_rewards/accuracies": 0.5899999737739563, | |
| "eval_rewards/chosen": 0.024747073650360107, | |
| "eval_rewards/margins": 0.018166616559028625, | |
| "eval_rewards/margins_max": 0.12387742102146149, | |
| "eval_rewards/margins_min": -0.0705643892288208, | |
| "eval_rewards/margins_std": 0.06444399803876877, | |
| "eval_rewards/rejected": 0.006580457091331482, | |
| "eval_runtime": 427.7975, | |
| "eval_samples": 2000, | |
| "eval_samples_per_second": 4.675, | |
| "eval_steps_per_second": 0.292, | |
| "train_loss": 0.6494339942932129, | |
| "train_runtime": 4013.1211, | |
| "train_samples": 5678, | |
| "train_samples_per_second": 1.415, | |
| "train_steps_per_second": 0.088 | |
| } |