© 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. 49747
    Prev
    Next
    INDEX
    Explanations

    good old, amount, number, chance

    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
    urv
    -0.09
    æ±ĩ
    -0.09
    éĢģä¸Ĭ
    -0.08
    è¿Ļä¹Ī说
    -0.08
    lined
    -0.08
    à¸Ļ
    -0.08
    åĨĴçĿĢ
    -0.08
    çŃ¹
    -0.08
    inan
    -0.08
    (nb
    -0.08
    POSITIVE LOGITS
    为客æĪ·
    0.10
    说ä¸įåĩº
    0.10
     ÐĴедÑĮ
    0.10
    身åIJİ
    0.09
    greens
    0.09
    ä¸įéĶĻ
    0.09
    aber
    0.09
    建ç«ĭèµ·
    0.09
    åįģåĩłä¸ª
    0.09
     dozen
    0.09
    Activations Density 0.029%

    No Known Activations