| { | |
| "epoch": 1.0, | |
| "eval_dpo_losses": 0.6691617965698242, | |
| "eval_logits/chosen": -2.305570602416992, | |
| "eval_logits/rejected": -2.1975905895233154, | |
| "eval_logps/chosen": -262.83721923828125, | |
| "eval_logps/rejected": -254.97503662109375, | |
| "eval_loss": 0.6798878312110901, | |
| "eval_positive_losses": 0.08152038604021072, | |
| "eval_rewards/accuracies": 0.7123016119003296, | |
| "eval_rewards/chosen": 0.12936948239803314, | |
| "eval_rewards/margins": 0.05091732367873192, | |
| "eval_rewards/margins_max": 0.19700397551059723, | |
| "eval_rewards/margins_min": -0.07996664941310883, | |
| "eval_rewards/margins_std": 0.09219963103532791, | |
| "eval_rewards/rejected": 0.07845214754343033, | |
| "eval_runtime": 387.9178, | |
| "eval_samples": 2000, | |
| "eval_samples_per_second": 5.156, | |
| "eval_steps_per_second": 0.162, | |
| "train_loss": 0.68390608707419, | |
| "train_runtime": 41762.7734, | |
| "train_samples": 61134, | |
| "train_samples_per_second": 1.464, | |
| "train_steps_per_second": 0.091 | |
| } |