© 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. 62584
    Prev
    Next
    INDEX
    Explanations

    replace 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
    Not in Any Lists

    No Comments

    Negative Logits
    仪
    -0.10
    est
    -0.10
    izz
    -0.10
    aland
    -0.10
    ç¬ijè¯Ŀ
    -0.09
    庾
    -0.09
    åĿŀ
    -0.09
    à¹Ħà¸ĭ
    -0.09
    wig
    -0.09
    emia
    -0.09
    POSITIVE LOGITS
     Ñģобой
    0.16
    ä¸İåıijå±ķ
    0.10
    able
    0.10
    æį¢äºĨ
    0.10
    åĴĮå®ĮåĸĦ
    0.10
    ä¸įäºĨ
    0.10
    ably
    0.09
    hips
    0.09
    /Add
    0.09
    oder
    0.09
    Activations Density 0.044%

    No Known Activations