Views
No views yet
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.0516 | 2.67 | 100 | 1.3486 | -7.4806 | -8.2143 | 0.5600 | 0.7337 | -327.5847 | -306.1782 | -0.9599 | -0.9344 |
| 0.0009 | 5.33 | 200 | 1.4038 | -6.2184 | -6.9730 | 0.6000 | 0.7545 | -315.1709 | -293.5561 | -1.1603 | -1.1661 |
| 0.0001 | 8.0 | 300 | 1.4280 | -6.9990 | -7.9051 | 0.5800 | 0.9061 | -324.4926 | -301.3619 | -1.1942 | -1.2090 |
| 0.0001 | 10.67 | 400 | 1.5040 | -7.4341 | -8.3860 | 0.5900 | 0.9519 | -329.3012 | -305.7131 | -1.1850 | -1.2017 |
| 0.0001 | 13.33 | 500 | 1.5476 | -7.6461 | -8.6008 | 0.6000 | 0.9548 | -331.4495 | -307.8324 | -1.1804 | -1.1981 |
| 0.0001 | 16.0 | 600 | 1.5649 | -7.7575 | -8.7226 | 0.5900 | 0.9651 | -332.6667 | -308.9463 | -1.1773 | -1.1949 |
| 0.0001 | 18.67 | 700 | 1.5789 | -7.8106 | -8.7769 | 0.6000 | 0.9663 | -333.2098 | -309.4775 | -1.1762 | -1.1945 |
| 0.0001 | 21.33 | 800 | 1.5773 | -7.8302 | -8.8009 | 0.5900 | 0.9706 | -333.4498 | -309.6743 | -1.1754 | -1.1931 |
| 0.0001 | 24.0 | 900 | 1.5810 | -7.8322 | -8.8002 | 0.5900 | 0.9680 | -333.4431 | -309.6937 | -1.1751 | -1.1931 |
| 0.0001 | 26.67 | 1000 | 1.5757 | -7.8281 | -8.8033 | 0.5900 | 0.9752 | -333.4745 | -309.6530 | -1.1752 | -1.1933 |