© 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. 4157
    Prev
    Next
    INDEX
    Explanations

    under/over statements

    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
    åĽŀ头çľĭ
    -0.09
    åĢļ
    -0.09
    åij¨è½¬
    -0.09
    åĽŀé¦ĸ
    -0.09
     drift
    -0.09
    ĶåĽŀ
    -0.08
    ä¸įåĪ©
    -0.08
    agne
    -0.08
    à¸Ľà¸¥
    -0.08
    éķ¿çŁŃ
    -0.08
    POSITIVE LOGITS
    elts
    0.09
     Slee
    0.09
    uchen
    0.09
    isoft
    0.09
    åľŁè±ª
    0.09
    BEL
    0.08
    _MORE
    0.08
    ypy
    0.08
    ität
    0.08
     Zuk
    0.08
    Activations Density 0.043%

    No Known Activations