Multiple Choice Classification on ARC-C, HellaSwag, Lambada, PIQA, and WinoGrande (Zero-Shot)
56.31ARC-C Accuracy
FP16
Updated 29d ago
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
2026.06 | 56.31 | 77.72 | 64.86 | 80.58 | 70.88 | 70.07 | |
2026.06 | 54.52 | 76.76 | 62.91 | 78.35 | 69.46 | 68.4 | |
2026.06 | 53.92 | 77.02 | 64.41 | 79.76 | 70.72 | 69.17 | |
2026.06 | 53.5 | 78.96 | 75.33 | 81.23 | 73.48 | 72.5 | |
2026.06 | 52.73 | 78.08 | 70 | 79.43 | 71.67 | 70.38 | |
2026.06 | 51.96 | 78.4 | 74.67 | 80.2 | 72.53 | 71.55 | |
2026.06 | 51.19 | 77.55 | 70.37 | 78.73 | 69.3 | 69.43 | |
2026.06 | 51.01 | 78.94 | 71.92 | 79.92 | 72.93 | 70.94 | |
2026.06 | 49.91 | 77.41 | 74.02 | 79.6 | 70.64 | 70.32 |