Language Modeling on WikiText-103 20w x 2048
9.603Perplexity (PPL)
fp16
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.05 | 9.603 | 0 | |
2026.05 | 9.621 | 0.019 | |
2026.05 | 9.701 | 0.098 | |
2026.05 | 10.698 | 1.095 | |
2026.05 | 11.427 | 1.824 | |
2026.05 | 44.564 | 34.96 | |
2026.05 | 118.977 | 109.4 | |
2026.05 | 121.715 | 112.1 | |
2026.05 | 588.222 | 578.6 | |
2026.05 | 1,529.4 | 1,520 |