Mathematical Reasoning on GSM8K (Performance vs Context Size)
74.3Acc (128)EvoToken (10k FT)
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| EvoToken (10k FT)NFE / Gen Len=12026.01 | 74.3 | 83.47 | 84.46 | 80.74 | |
| EvoToken (10k FT)NFE / Gen Len=1/22026.01 | 73.54 | 82.03 | 81.8 | 79.12 | |
| FT-Base (10k FT)NFE / Gen Len=12026.01 | 71.04 | 82.11 | 82.56 | 78.57 | |
| BaselineNFE / Gen Len=12026.01 | 70.2 | 79.3 | 83.47 | 77.66 | |
| BaselineNFE / Gen Len=1/22026.01 | 67.55 | 77.63 | 79.83 | 75 | |
| EvoToken (10k FT)NFE / Gen Len=1/42026.01 | 64.82 | 75.74 | 72.33 | 70.96 | |
| FT-Base (10k FT)NFE / Gen Len=1/22026.01 | 63.91 | 78.62 | 79 | 73.84 | |
| BaselineNFE / Gen Len=1/42026.01 | 59.14 | 68.23 | 66.57 | 64.65 | |
| FT-Base (10k FT)NFE / Gen Len=1/42026.01 | 49.05 | 62.17 | 61.87 | 57.7 |