© 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. 27378
    Prev
    Next
    INDEX
    Explanations

    dirt, dust, debris

    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
     Pit
    -0.09
    åĽº
    -0.09
    _where
    -0.09
     æĽ
    -0.09
    /pdf
    -0.09
    ãģŀ
    -0.09
    主导
    -0.08
    ÑĪиб
    -0.08
    isas
    -0.08
    .mvp
    -0.08
    POSITIVE LOGITS
    åľ¨éĩĮéĿ¢
    0.11
    ä¸į便
    0.09
    eger
    0.09
    ä¸įæĺĵ
    0.08
    ads
    0.08
    iteur
    0.08
    趸
    0.08
    åģĮ
    0.08
     ÑĢа
    0.08
    _capture
    0.08
    Activations Density 0.062%

    No Known Activations