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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.0472 | 2.67 | 100 | 0.7233 | -3.9671 | -5.5265 | 0.7100 | 1.5594 | -169.6246 | -147.2690 | -1.3141 | -1.2833 |
| 0.0011 | 5.33 | 200 | 1.0202 | -9.9421 | -12.8319 | 0.7800 | 2.8897 | -242.6781 | -207.0196 | -1.1319 | -1.1748 |
| 0.0001 | 8.0 | 300 | 1.0110 | -11.1287 | -14.0752 | 0.7700 | 2.9464 | -255.1111 | -218.8855 | -0.9542 | -1.0067 |
| 0.0001 | 10.67 | 400 | 1.0180 | -11.3089 | -14.2918 | 0.7700 | 2.9828 | -257.2771 | -220.6876 | -0.9396 | -0.9943 |
| 0.0001 | 13.33 | 500 | 1.0310 | -11.4360 | -14.4349 | 0.7600 | 2.9989 | -258.7085 | -221.9584 | -0.9299 | -0.9860 |
| 0.0001 | 16.0 | 600 | 1.0351 | -11.5046 | -14.5147 | 0.7600 | 3.0101 | -259.5066 | -222.6438 | -0.9250 | -0.9813 |
| 0.0001 | 18.67 | 700 | 1.0333 | -11.5258 | -14.5522 | 0.7600 | 3.0265 | -259.8819 | -222.8561 | -0.9224 | -0.9792 |
| 0.0001 | 21.33 | 800 | 1.0377 | -11.5332 | -14.5528 | 0.7600 | 3.0196 | -259.8878 | -222.9305 | -0.9219 | -0.9784 |
| 0.0001 | 24.0 | 900 | 1.0336 | -11.5345 | -14.5657 | 0.7600 | 3.0312 | -260.0164 | -222.9427 | -0.9215 | -0.9782 |
| 0.0001 | 26.67 | 1000 | 1.0392 | -11.5341 | -14.5591 | 0.7600 | 3.0250 | -259.9501 | -222.9392 | -0.9217 | -0.9783 |