| { | |
| "epoch": 1.0, | |
| "eval_dpo_losses": 0.6626402139663696, | |
| "eval_logits/chosen": -2.78564453125, | |
| "eval_logits/rejected": -2.747466802597046, | |
| "eval_logps/chosen": -268.0898132324219, | |
| "eval_logps/rejected": -252.3152313232422, | |
| "eval_loss": 0.6810576319694519, | |
| "eval_positive_losses": 0.1281057745218277, | |
| "eval_rewards/accuracies": 0.726190447807312, | |
| "eval_rewards/chosen": 0.16405092179775238, | |
| "eval_rewards/margins": 0.06570956110954285, | |
| "eval_rewards/margins_max": 0.2505451738834381, | |
| "eval_rewards/margins_min": -0.10226902365684509, | |
| "eval_rewards/margins_std": 0.11677397042512894, | |
| "eval_rewards/rejected": 0.09834136068820953, | |
| "eval_runtime": 388.0843, | |
| "eval_samples": 2000, | |
| "eval_samples_per_second": 5.154, | |
| "eval_steps_per_second": 0.162, | |
| "train_loss": 0.6809778677109991, | |
| "train_runtime": 43303.5615, | |
| "train_samples": 61134, | |
| "train_samples_per_second": 1.412, | |
| "train_steps_per_second": 0.088 | |
| } |