© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 58122
    Prev
    Next
    INDEX
    Explanations

    endings like aud, edy, Center

    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
    Not in Any Lists

    No Comments

    Negative Logits
    夬
    -0.10
    /vnd
    -0.09
    /LICENSE
    -0.09
    ockey
    -0.09
    ä¸Ģ人
    -0.08
     rig
    -0.08
    irts
    -0.08
    @s
    -0.08
    å¾ģ
    -0.08
    çijŁ
    -0.08
    POSITIVE LOGITS
     behold
    0.09
    Ä°M
    0.09
    åķ¬
    0.09
     ÑĤоÑĢ
    0.08
    æľ¯
    0.08
    æĸĭ
    0.08
    zá
    0.08
    åĽ¢è´Ń
    0.08
    æİĮ
    0.08
    shield
    0.08
    Activations Density 0.120%

    No Known Activations