Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

NLP settings

Benchmarks

Task NameDataset NameSOTA ResultTrend
Concept Erasure39 NLP settings (13 LLMs x 3 concepts: sycophancy, gender, safety)
Mean Residual Leakage |DS|0
40
Showing 1 of 1 rows