Common Sense Reasoning on Common Sense Reasoning
47.9ARC AccuracyHyLo-Qwen-14MLA14GDN
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| HyLo-Qwen-14MLA14GDNTeacher=8B, KV cache=7.8%, Training Context Length=64K2026.04 | 47.9 | 74.6 | 61.9 | 38.2 | 75 | 36.8 | 62.3 | 56.7 | |
| HyLo-Qwen-14MLA14M2Teacher=8B, KV cache=7.8%, Training Context Length=8K2026.04 | 45.9 | 73.2 | 61.2 | 39.2 | 73.8 | 36.5 | 63.1 | 56.1 | |
| HyLo-Qwen-14MLA14GDNTeacher=8B, KV cache=7.8%, Training Context Length=8K2026.04 | 45.9 | 73.3 | 62.1 | 38.4 | 74.8 | 37.4 | 63.4 | 56.5 | |
| HyLo-Qwen-14MLA14M2Teacher=8B, KV cache=7.8%, Training Context Length=64K2026.04 | 45.4 | 73.3 | 61.1 | 37.8 | 73.7 | 36.6 | 61.8 | 55.7 | |
| HyLo-Qwen-7MLA21GDNTeacher=8B, KV cache=3.9%, Training Context Length=8K2026.04 | 45.3 | 72.7 | 61.5 | 39.4 | 73.2 | 36.3 | 64.6 | 56.1 | |
| HyLo-Qwen-7MLA21M2Teacher=8B, KV cache=3.9%, Training Context Length=8K2026.04 | 44.3 | 71.4 | 60.5 | 39.4 | 73.3 | 36.3 | 61.5 | 55.2 | |
| HyLo-Qwen-7MLA21GDNTeacher=8B, KV cache=3.9%, Training Context Length=64K2026.04 | 44.2 | 71.4 | 61.2 | 37.4 | 73.7 | 36.9 | 63.1 | 55.4 | |
| HyLo-Qwen-7MLA21M2Teacher=8B, KV cache=3.9%, Training Context Length=64K2026.04 | 42.7 | 70 | 60.3 | 38 | 73.5 | 35.7 | 63.8 | 54.9 | |
| Jet Nemotron-2BTeacher=-, KV cache=2.1%2026.04 | 42.5 | 54.5 | 64.4 | 34 | 73.5 | 35.4 | 64.9 | 52.7 | |
| Hype Net (7FA21LA)Teacher=1.7B, KV cache=25%2026.04 | 41.6 | 67.9 | 57.4 | 36.6 | 72.7 | 32.9 | 63.1 | 53.2 |