© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 16530
    Prev
    Next
    INDEX
    Explanations

    end up with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    ä¹ĭå¤Ħ
    -0.10
    ä¸ĢåŃĹ
    -0.10
    zeÅĽ
    -0.10
    _ENTER
    -0.09
    å®Įæ¯ķ
    -0.09
    人å±ħ
    -0.09
    аÑĪ
    -0.09
    俾
    -0.09
    ienne
    -0.09
    enge
    -0.09
    POSITIVE LOGITS
    满满çļĦ
    0.10
    pole
    0.08
     cuk
    0.08
    çĥ«
    0.08
     responsibility
    0.08
    å¾Īéĩįè¦ģ
    0.08
    ä»Ģä¹Īæĺ¯
    0.07
    æŀģ大çļĦ
    0.07
    itude
    0.07
     fewer
    0.07
    Activations Density 0.012%

    No Known Activations