© 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. 18601
    Prev
    Next
    INDEX
    Explanations

    liner

    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
    ening
    -0.10
    alias
    -0.09
    大æ°Ķ
    -0.09
    éļľ
    -0.09
    ãģªãĤĭ
    -0.09
    a
    -0.09
    æݪ
    -0.09
    ä¸ĢåIJį
    -0.08
    ened
    -0.08
    naire
    -0.08
    POSITIVE LOGITS
    è¿Ļå°±æĺ¯
    0.09
    ordes
    0.09
    æĹłæ¯Ķ
    0.09
    /frontend
    0.08
    UED
    0.08
    illos
    0.08
    pline
    0.08
    izia
    0.08
    .precision
    0.08
    å®Įåħ¨æ²¡æľī
    0.08
    Activations Density 0.003%

    No Known Activations