{ "epoch": 1.0, "eval_logits/chosen": -1.3928931951522827, "eval_logits/rejected": -1.3641632795333862, "eval_logps/chosen": -0.5413306355476379, "eval_logps/rejected": -0.7055191397666931, "eval_loss": 1.1963940858840942, "eval_nll_loss": 0.0, "eval_rewards/accuracies": 0.6251680254936218, "eval_rewards/chosen": -1.0826612710952759, "eval_rewards/margins": 0.32764163613319397, "eval_rewards/rejected": -1.4110382795333862, "eval_runtime": 26.5134, "eval_samples": 1961, "eval_samples_per_second": 73.963, "eval_steps_per_second": 2.338 }