Mathematical Reasoning on Math Out-of-Domain
67.59AccuracyBase
Evaluation Results
| Method | Links | |
|---|---|---|
| BaseBase Model=Llama3.2-3B2026.05 | 67.59 | |
| BaseBase Model=Qwen2.5-1.5B2026.05 | 64.83 | |
| PionBase Model=Qwen2.5-1.5B2026.05 | 62.16 | |
| AdamWBase Model=Qwen2.5-1.5B2026.05 | 62.13 | |
| MuonBase Model=Qwen2.5-1.5B2026.05 | 61.22 | |
| MuonBase Model=Llama3.2-3B2026.05 | 61.2 | |
| AdamWBase Model=Llama3.2-3B2026.05 | 60.86 | |
| PionBase Model=Llama3.2-3B2026.05 | 60.44 |