You Do Not Fully Utilize Transformer's Representation Capacity Paper • 2502.09245 • Published Feb 13 • 35 • 3
You Do Not Fully Utilize Transformer's Representation Capacity Paper • 2502.09245 • Published Feb 13 • 35
You Do Not Fully Utilize Transformer's Representation Capacity Paper • 2502.09245 • Published Feb 13 • 35
Analyze Feature Flow to Enhance Interpretation and Steering in Language Models Paper • 2502.03032 • Published Feb 5 • 60
The Differences Between Direct Alignment Algorithms are a Blur Paper • 2502.01237 • Published Feb 3 • 115
elephantmipt/sae_Qwen_Qwen2.5-7B_resid_pre_layer_24_size_16384_batchtopk_reg_coeff_0.0018 Updated Dec 29, 2024
elephantmipt/sae_Qwen_Qwen2.5-7B_resid_pre_layer_18_size_16384_batchtopk_reg_coeff_0.0018 Updated Dec 29, 2024
elephantmipt/sae_Qwen_Qwen2.5-7B_resid_pre_layer_12_size_16384_batchtopk_reg_coeff_0.0018 Updated Dec 29, 2024
elephantmipt/sae_Qwen_Qwen2.5-7B_resid_pre_layer_12_size_16384_batchtopk_reg_coeff_0.0018 Updated Dec 29, 2024
elephantmipt/sae_Qwen_Qwen2.5-7B_resid_pre_layer_24_size_16384_batchtopk_reg_coeff_0.0018 Updated Dec 29, 2024
elephantmipt/sae_Qwen_Qwen2.5-7B_resid_pre_layer_18_size_16384_batchtopk_reg_coeff_0.0018 Updated Dec 29, 2024
elephantmipt/sae_Qwen_Qwen2.5-7B_resid_pre_layer_12_size_16384_batchtopk_reg_coeff_0.0018 Updated Dec 29, 2024
elephantmipt/sae_Qwen_Qwen2.5-7B_resid_pre_layer_24_size_16384_batchtopk_reg_coeff_0.0018 Updated Dec 29, 2024
elephantmipt/sae_Qwen_Qwen2.5-7B_resid_pre_layer_18_size_16384_batchtopk_reg_coeff_0.0018 Updated Dec 29, 2024