Student performance simulation on NAEP Grade 4
0.7Pearson Correlation (r)Gemma-3-27b
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Gemma-3-27bModel=Gemma-3-27b2026.01 | 0.7 | 0.74 | 0.88 | |
| Gemma-2-9bModel=Gemma-2-9b2026.01 | 0.69 | 0.69 | 0.84 | |
| Gemma-2-27bModel=Gemma-2-27b2026.01 | 0.69 | 0.68 | 0.85 | |
| Qwen2.5-32BModel=Qwen2.5-32B2026.01 | 0.6 | 0.64 | 0.82 | |
| Qwen2-7bModel=Qwen2-7b2026.01 | 0.6 | 0.64 | 0.79 | |
| Llama-3-8bModel=Llama-3-8b2026.01 | 0.42 | 0.4 | 0.67 |