{ "epoch": 0.9965156794425087, "eval_logits/chosen": -2.5121426582336426, "eval_logits/rejected": -2.479111909866333, "eval_logps/chosen": -261.0513610839844, "eval_logps/rejected": -266.893798828125, "eval_loss": 0.671826183795929, "eval_rewards/accuracies": 0.62109375, "eval_rewards/chosen": 0.015785472467541695, "eval_rewards/margins": 0.05810253322124481, "eval_rewards/rejected": -0.04231705889105797, "eval_runtime": 101.5931, "eval_samples": 2000, "eval_samples_per_second": 19.686, "eval_steps_per_second": 0.315 }