Multitask Language Understanding on MMLU (Accuracy, Exit Position)
68MMLU AccuracyBackbone
Evaluation results
| Method | Links | ||
|---|---|---|---|
| BackboneBackbone Model=Llama3.1 8B (32 layers)2026.04 | 68 | — | |
| River-LLMBackbone Model=Llama3.1 8B (32 layers), Threshold (τ)=0.72026.04 | 67.4 | 14.01 | |
| River-LLMBackbone Model=Llama3.1 8B (32 layers), Threshold (τ)=0.52026.04 | 66.1 | 3.04 | |
| BackboneBackbone Model=Llama3.2 1B (16 layers)2026.04 | 46.1 | — | |
| River-LLMBackbone Model=Llama3.2 1B (16 layers), Threshold (τ)=0.72026.04 | 46 | 12.7 | |
| River-LLMBackbone Model=Llama3.2 1B (16 layers), Threshold (τ)=0.52026.04 | 44.3 | 4.05 |