|
{ |
|
"epoch": 8.88888888888889, |
|
"eval_logits/chosen": -0.39696812629699707, |
|
"eval_logits/rejected": -0.5027680397033691, |
|
"eval_logps/chosen": -1.6218750476837158, |
|
"eval_logps/rejected": -2.2122323513031006, |
|
"eval_loss": 1.6737581491470337, |
|
"eval_odds_ratio_loss": 14.46813678741455, |
|
"eval_rewards/accuracies": 0.800000011920929, |
|
"eval_rewards/chosen": -0.16218750178813934, |
|
"eval_rewards/margins": 0.05903574824333191, |
|
"eval_rewards/rejected": -0.22122326493263245, |
|
"eval_runtime": 0.4129, |
|
"eval_samples_per_second": 24.22, |
|
"eval_sft_loss": 0.22694425284862518, |
|
"eval_steps_per_second": 12.11 |
|
} |