© 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. 58796
    Prev
    Next
    INDEX
    Explanations

    fish and fish names

    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
    td
    -0.09
    IZED
    -0.09
    chos
    -0.09
    åıī
    -0.09
    [out
    -0.09
    ÑģÑĢ
    -0.09
    们çļĦ
    -0.09
    artz
    -0.09
    evin
    -0.09
    urus
    -0.08
    POSITIVE LOGITS
    sing
    0.09
    ä¸ĢæĿ¡é¾Ļ
    0.09
    ness
    0.09
    yon
    0.09
    ëŁī
    0.09
    bourne
    0.09
    缮çļĦæĺ¯
    0.09
    å¼ĢèįĴ
    0.09
    borne
    0.09
    HING
    0.08
    Activations Density 0.010%

    No Known Activations