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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.0273 | 4.0 | 100 | 2.6475 | 9.4236 | 9.7300 | 0.5600 | -0.3064 | -231.1558 | -235.4583 | -0.6116 | -0.6128 |
| 0.003 | 8.0 | 200 | 2.7277 | -2.3653 | -3.6848 | 0.5700 | 1.3195 | -246.0612 | -248.5571 | -0.6651 | -0.6643 |
| 0.0159 | 12.0 | 300 | 3.7598 | 6.7420 | 5.0769 | 0.5700 | 1.6652 | -236.3260 | -238.4378 | -0.6714 | -0.6918 |
| 0.0 | 16.0 | 400 | 4.3021 | -4.0714 | -4.2226 | 0.4800 | 0.1512 | -246.6587 | -250.4527 | -0.5222 | -0.5521 |
| 0.0 | 20.0 | 500 | 4.2861 | -4.1790 | -4.3335 | 0.4700 | 0.1544 | -246.7819 | -250.5723 | -0.5219 | -0.5517 |
| 0.0 | 24.0 | 600 | 4.3048 | -4.1842 | -4.3136 | 0.4700 | 0.1294 | -246.7599 | -250.5781 | -0.5211 | -0.5507 |
| 0.0 | 28.0 | 700 | 4.2866 | -4.2055 | -4.3353 | 0.4700 | 0.1297 | -246.7839 | -250.6018 | -0.5215 | -0.5511 |
| 0.0 | 32.0 | 800 | 4.3425 | -4.2344 | -4.2951 | 0.4500 | 0.0606 | -246.7392 | -250.6339 | -0.5212 | -0.5512 |
| 0.0 | 36.0 | 900 | 4.3225 | -4.1986 | -4.2873 | 0.4800 | 0.0888 | -246.7307 | -250.5940 | -0.5215 | -0.5512 |
| 0.0 | 40.0 | 1000 | 4.3180 | -4.2235 | -4.3178 | 0.4700 | 0.0943 | -246.7645 | -250.6217 | -0.5209 | -0.5511 |