{ "epoch": 1.0, "eval_logits/chosen": -2.22967529296875, "eval_logits/rejected": -2.2248213291168213, "eval_logps/chosen": -34.096519470214844, "eval_logps/rejected": -37.601966857910156, "eval_loss": 0.6981170177459717, "eval_rewards/accuracies": 0.545265793800354, "eval_rewards/chosen": -0.0557732917368412, "eval_rewards/margins": 0.021040165796875954, "eval_rewards/rejected": -0.07681345194578171, "eval_runtime": 145.6397, "eval_samples": 343, "eval_samples_per_second": 2.355, "eval_steps_per_second": 0.295 }