{ "epoch": 1.0, "eval_logits/chosen": -2.369189977645874, "eval_logits/rejected": -2.323690414428711, "eval_logps/chosen": -143.66659545898438, "eval_logps/rejected": -171.08460998535156, "eval_loss": 0.14922186732292175, "eval_rewards/accuracies": 0.8359375, "eval_rewards/chosen": -1.5451841354370117, "eval_rewards/margins": 5.666274070739746, "eval_rewards/rejected": -7.2114577293396, "eval_runtime": 155.3805, "eval_samples": 506, "eval_samples_per_second": 3.257, "eval_steps_per_second": 0.103 }