© 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. 34515
    Prev
    Next
    INDEX
    Explanations

    whip, puffed, flip, wrought, Whitney

    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
    à¹ĥà¸ļ
    -0.09
    ook
    -0.09
    .Dependency
    -0.09
    azes
    -0.09
    æ²¹èħ»
    -0.09
    à¸ŀวà¸ģ
    -0.08
    vak
    -0.08
    们
    -0.08
    æ¼ł
    -0.08
    åŁĨ
    -0.08
    POSITIVE LOGITS
    bole
    0.11
    åıįæĺłåĩº
    0.10
    壤
    0.10
    pest
    0.09
    躬
    0.09
    _Arg
    0.09
    èĸĽ
    0.09
    ottom
    0.08
    orte
    0.08
    _STMT
    0.08
    Activations Density 0.001%

    No Known Activations