| { |
| "checkpoint": "checkpoint-00000704", |
| "ema": { |
| "elapsed_seconds": 54.355851609958336, |
| "logps_chosen": -601.3653792912514, |
| "logps_rejected": -652.0577934662236, |
| "loss": 0.8922630808546166, |
| "nll_chosen_per_token": 0.5581774069041227, |
| "pairs": 903, |
| "response_tokens": 2363563, |
| "rewards_accuracy": 0.7895902547065338, |
| "rewards_chosen": -1.6784689598041251, |
| "rewards_margin": 1.3302929308987403, |
| "rewards_rejected": -3.008761755519795 |
| }, |
| "normal": { |
| "elapsed_seconds": 53.85813916684128, |
| "logps_chosen": -601.1099114064231, |
| "logps_rejected": -651.828419158361, |
| "loss": 0.8913962870077173, |
| "nll_chosen_per_token": 0.5579402858352829, |
| "pairs": 903, |
| "response_tokens": 2363563, |
| "rewards_accuracy": 0.7918050941306756, |
| "rewards_chosen": -1.6656929430111433, |
| "rewards_margin": 1.331598123442691, |
| "rewards_rejected": -2.997290931270764 |
| }, |
| "optimizer_step": 704, |
| "type": "preference_validation" |
| } |
|
|
