{ "epoch": 1.0, "eval_logits/chosen": 98.71852111816406, "eval_logits/rejected": 98.69664001464844, "eval_logps/chosen": -32.537837982177734, "eval_logps/rejected": -36.13877868652344, "eval_loss": 1.023661732673645, "eval_rewards/accuracies": 0.531146228313446, "eval_rewards/chosen": -0.06626088172197342, "eval_rewards/margins": 0.054315000772476196, "eval_rewards/rejected": -0.12057589739561081, "eval_runtime": 103.7609, "eval_samples": 343, "eval_samples_per_second": 3.306, "eval_steps_per_second": 0.414 }