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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.1181 | 2.67 | 100 | 2.5145 | -3.2811 | -2.1770 | 0.3500 | -1.1041 | -203.8431 | -267.3410 | -0.8870 | -0.8359 |
| 0.0064 | 5.33 | 200 | 2.8001 | -4.5020 | -4.5078 | 0.4800 | 0.0058 | -206.4329 | -268.6977 | -0.6007 | -0.6015 |
| 0.0024 | 8.0 | 300 | 4.0781 | -10.0150 | -8.7704 | 0.4500 | -1.2446 | -211.1690 | -274.8232 | -0.5634 | -0.5539 |
| 0.0 | 10.67 | 400 | 2.9500 | -4.5899 | -4.2752 | 0.4600 | -0.3147 | -206.1744 | -268.7953 | -0.7237 | -0.7333 |
| 0.0 | 13.33 | 500 | 2.9586 | -4.5119 | -4.1757 | 0.4700 | -0.3362 | -206.0639 | -268.7087 | -0.7220 | -0.7324 |
| 0.0 | 16.0 | 600 | 2.9732 | -4.5309 | -4.1470 | 0.4600 | -0.3838 | -206.0320 | -268.7297 | -0.7234 | -0.7333 |
| 0.0 | 18.67 | 700 | 2.9960 | -4.5757 | -4.1770 | 0.4600 | -0.3987 | -206.0654 | -268.7795 | -0.7215 | -0.7319 |
| 0.0 | 21.33 | 800 | 3.0039 | -4.5481 | -4.1465 | 0.4600 | -0.4015 | -206.0314 | -268.7488 | -0.7222 | -0.7330 |
| 0.0 | 24.0 | 900 | 2.9905 | -4.5855 | -4.1741 | 0.4600 | -0.4114 | -206.0621 | -268.7904 | -0.7219 | -0.7324 |
| 0.0 | 26.67 | 1000 | 2.9825 | -4.5568 | -4.1702 | 0.4400 | -0.3866 | -206.0578 | -268.7585 | -0.7217 | -0.7321 |