{ "epoch": 2.981366459627329, "eval_logits/chosen": -2.1261515617370605, "eval_logits/rejected": -2.0965681076049805, "eval_logps/chosen": -241.2080078125, "eval_logps/rejected": -252.73114013671875, "eval_loss": 0.5531922578811646, "eval_rewards/accuracies": 0.8055555820465088, "eval_rewards/chosen": 1.6544404029846191, "eval_rewards/margins": 2.9077939987182617, "eval_rewards/rejected": -1.253353476524353, "eval_runtime": 76.4448, "eval_samples_per_second": 14.913, "eval_steps_per_second": 0.235 }