{ "epoch": 2.9762532981530345, "eval_logits/chosen": -2.0633089542388916, "eval_logits/rejected": -2.0397870540618896, "eval_logps/chosen": -288.552001953125, "eval_logps/rejected": -237.90699768066406, "eval_loss": 0.455725759267807, "eval_rewards/accuracies": 0.819767415523529, "eval_rewards/chosen": 1.5898916721343994, "eval_rewards/margins": 2.9930419921875, "eval_rewards/rejected": -1.403150200843811, "eval_runtime": 180.1111, "eval_samples_per_second": 14.93, "eval_steps_per_second": 0.239 }