{ "epoch": 0.9994767137624281, "eval_logits/chosen": -2.5900135040283203, "eval_logits/rejected": -2.5359914302825928, "eval_logps/chosen": -282.1169128417969, "eval_logps/rejected": -272.686279296875, "eval_loss": 0.5283266305923462, "eval_rewards/accuracies": 0.773809552192688, "eval_rewards/chosen": -0.016292406246066093, "eval_rewards/margins": 1.2304461002349854, "eval_rewards/rejected": -1.2467385530471802, "eval_runtime": 205.7802, "eval_samples": 2000, "eval_samples_per_second": 9.719, "eval_steps_per_second": 0.306 }