| { | |
| "epoch": 2.986666666666667, | |
| "eval_logits/chosen": -0.4098573923110962, | |
| "eval_logits/rejected": -0.48421889543533325, | |
| "eval_logps/chosen": -0.8694608807563782, | |
| "eval_logps/rejected": -1.295823335647583, | |
| "eval_loss": 0.9151002764701843, | |
| "eval_odds_ratio_loss": 8.100966453552246, | |
| "eval_rewards/accuracies": 0.8399999737739563, | |
| "eval_rewards/chosen": -0.08694608509540558, | |
| "eval_rewards/margins": 0.04263625293970108, | |
| "eval_rewards/rejected": -0.12958233058452606, | |
| "eval_runtime": 2.0333, | |
| "eval_samples_per_second": 24.59, | |
| "eval_sft_loss": 0.10500368475914001, | |
| "eval_steps_per_second": 12.295, | |
| "total_flos": 1.516498186272768e+16, | |
| "train_loss": 1.23334006468455, | |
| "train_runtime": 194.0073, | |
| "train_samples_per_second": 6.959, | |
| "train_steps_per_second": 0.433 | |
| } |