| { | |
| "epoch": 2.9865871833084947, | |
| "eval_log_odds_chosen": 0.2937372922897339, | |
| "eval_log_odds_ratio": -0.6945178508758545, | |
| "eval_logits/chosen": 300.6891174316406, | |
| "eval_logits/rejected": 271.8756103515625, | |
| "eval_logps/chosen": -1.0802680253982544, | |
| "eval_logps/rejected": -1.2502641677856445, | |
| "eval_loss": 1.539820671081543, | |
| "eval_nll_loss": 1.4724125862121582, | |
| "eval_rewards/accuracies": 0.5395683646202087, | |
| "eval_rewards/chosen": -0.05401340499520302, | |
| "eval_rewards/margins": 0.00849980115890503, | |
| "eval_rewards/rejected": -0.06251321732997894, | |
| "eval_runtime": 112.2968, | |
| "eval_samples": 553, | |
| "eval_samples_per_second": 4.924, | |
| "eval_steps_per_second": 1.238, | |
| "total_flos": 0.0, | |
| "train_loss": 1.4570662823027956, | |
| "train_runtime": 13599.7579, | |
| "train_samples": 5364, | |
| "train_samples_per_second": 1.183, | |
| "train_steps_per_second": 0.037 | |
| } |