MINT Workshop @ NeurIPS2024 retweetledi

Thread on the best paper award at our workshop on model interventions @MINT_workshop
Ekdeep Singh Lubana@EkdeepL
Paper alert––*Awarded best paper* at NeurIPS workshop on Foundation Model Interventions! 🧵👇 We analyze the (in)abilities of SAEs by relating them to the field of disentangled rep. learning, where limitations of AE based interpretability protocols have been well established!🤯
English







