LLaMA 7B + Inference Time Intervention (ITI)
Reported on 3 benchmarks across 1 task
Note: results are matched by exact model name. Different papers may use the same name for different model variants.
Natural Language Processing3 results
- % info93.8best: 97.7 (Alpaca 7B + Inference Time Intervention (ITI))
- % true45.1best: 88.6 (Vicuna 7B + Inference Time Intervention (ITI))
- 0.288best: 0.59 (GPT-4 (RLHF))