Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Output-based feature description faithfulness on Llama MLP SAE 3.1

55.5Faithfulness Score

EnsembleC (All)

45.41248.03150.6553.269Jan 14, 2025
Updated 4d ago

Evaluation Results

MethodLinks
2025.01
55.5
2025.01
53.1
2025.01
51.6
2025.01
50.7
2025.01
49.6
2025.01
48.9
2025.01
48.2
2025.01
45.8