| { | |
| "epoch": 1.0, | |
| "eval_dpo_losses": 0.6851017475128174, | |
| "eval_logits/chosen": -2.796370506286621, | |
| "eval_logits/rejected": -2.7574775218963623, | |
| "eval_logps/chosen": -278.58013916015625, | |
| "eval_logps/rejected": -254.24514770507812, | |
| "eval_loss": 0.6902781128883362, | |
| "eval_positive_losses": 0.036926720291376114, | |
| "eval_rewards/accuracies": 0.609000027179718, | |
| "eval_rewards/chosen": 0.060132984071969986, | |
| "eval_rewards/margins": 0.016795208677649498, | |
| "eval_rewards/margins_max": 0.09327653795480728, | |
| "eval_rewards/margins_min": -0.05095779150724411, | |
| "eval_rewards/margins_std": 0.0482446551322937, | |
| "eval_rewards/rejected": 0.04333777725696564, | |
| "eval_runtime": 428.2812, | |
| "eval_samples": 2000, | |
| "eval_samples_per_second": 4.67, | |
| "eval_steps_per_second": 0.292, | |
| "train_loss": 0.6833745032026056, | |
| "train_runtime": 9125.7542, | |
| "train_samples": 10941, | |
| "train_samples_per_second": 1.199, | |
| "train_steps_per_second": 0.075 | |
| } |