Character-level language modeling on Enwik8 (val)
1.04BPC
Adaptive Transformer
Updated 1mo ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2019.11 | 1.04 | - | - | - | |
2019.11 | 1.096 | - | - | - | |
2019.11 | 1.1 | - | - | - | |
2019.11 | 1.135 | - | - | - | |
2025.11 | 1.136 | - | - | - | |
2025.11 | 1.141 | - | - | - | |
2025.11 | 1.143 | - | - | - | |
2025.11 | 1.145 | - | - | - | |
2025.11 | 1.147 | - | - | - | |
2025.11 | 1.154 | - | - | - | |
2019.11 | 1.182 | - | - | - | |
2017.09 | 1.19 | - | - | - | |
2017.09 | 1.21 | - | - | - | |
2017.09 | 1.25 | - | - | - | |
2017.09 | 1.26 | - | - | - | |
2017.09 | 1.28 | - | - | - | |
2017.09 | 1.29 | - | - | - | |
2019.11 | 1.312 | - | - | - | |
2017.09 | 1.35 | - | - | - | |
2017.09 | 1.36 | - | - | - | |
2017.09 | 1.37 | - | - | - | |
2026.04 | 1.8429 | 1.2774 | - | - | |
2026.04 | 1.8504 | 1.2826 | - | - | |
2026.05 | - | 6.571 | 1.4 | - | |
2026.05 | - | 6.616 | - | - | |
2026.05 | - | 6.662 | - | - | |
2026.05 | - | 6.426 | 3.5 | 2.1 | |
2026.05 | - | 6.532 | - | - | |
2026.05 | - | 6.657 | - | - | |
2026.05 | - | 6.687 | 5.1 | 1.6 | |
2026.05 | - | 6.74 | - | - | |
2026.05 | - | 7.045 | - | - | |
2026.05 | - | 6.333 | 6.7 | 1.6 | |
2026.05 | - | 6.422 | - | - | |
2026.05 | - | 6.787 | - | - | |
2026.05 | - | 6.196 | 9.9 | 3.2 | |
2026.05 | - | 6.286 | - | - | |
2026.05 | - | 6.873 | - | - | |
2026.05 | - | 6.433 | 11.4 | 1.5 | |
2026.05 | - | 6.549 | - | - | |
2026.05 | - | 7.259 | - | - | |
2026.05 | - | 6.169 | - | - |