Language Modeling on The Pile DM Math (test)
82.14BPB (%)kNN
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| kNNTime Budget (T)=2.0s, Average Selection Size (N*)=172026.05 | 82.14 | — | — | |
| HullFTTime Budget (T)=2.0s, Average Selection Size (N*)=222026.05 | 82.58 | — | -0.44 | |
| kNNTime Budget (T)=1.75s, Average Selection Size (N*)=152026.05 | 83.31 | — | — | |
| SIFTTime Budget (T)=2.0s, Average Selection Size (N*)=132026.05 | 83.97 | — | — | |
| HullFTTime Budget (T)=1.75s, Average Selection Size (N*)=192026.05 | 84.12 | — | -0.81 | |
| SIFTTime Budget (T)=1.75s, Average Selection Size (N*)=112026.05 | 85.29 | — | — | |
| Split ModelModel structure=Split, Total compute budget=720B tokens, Number of parameters=1.3B, ts=340B2026.03 | — | 7.81 | — | |
| Standalone Pretrained ModelModel structure=Standalone, Total compute budget=720B tokens, Number of parameters=1.3B2026.03 | — | 8.14 | — |