| { |
| "checkpoint": "checkpoint-00005155", |
| "ema": { |
| "elapsed_seconds": 19.04161799896974, |
| "logps_chosen": -1186.886, |
| "logps_rejected": -1057.69725, |
| "loss": 0.6683739013671876, |
| "nll_chosen_per_token": 1.0625191911903595, |
| "pairs": 1000, |
| "response_tokens": 2072791, |
| "rewards_accuracy": 0.811, |
| "rewards_chosen": 0.7128524169921875, |
| "rewards_margin": 2.95132275390625, |
| "rewards_rejected": -2.23847021484375 |
| }, |
| "normal": { |
| "elapsed_seconds": 18.97880223800894, |
| "logps_chosen": -1186.8255, |
| "logps_rejected": -1057.895, |
| "loss": 0.6673851318359375, |
| "nll_chosen_per_token": 1.0624650306298111, |
| "pairs": 1000, |
| "response_tokens": 2072791, |
| "rewards_accuracy": 0.81, |
| "rewards_chosen": 0.71587646484375, |
| "rewards_margin": 2.96424072265625, |
| "rewards_rejected": -2.2483642578125 |
| }, |
| "optimizer_step": 5155, |
| "type": "preference_validation" |
| } |
|
|
