© 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. 30594
    Prev
    Next
    INDEX
    Explanations

    wih /

    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
     Bard
    -0.08
    ype
    -0.08
    3
    -0.08
    phet
    -0.08
    .sourceforge
    -0.08
    顺
    -0.08
    thon
    -0.08
    åı«åģļ
    -0.08
    0
    -0.08
    åİŁæĿ¥æĺ¯
    -0.08
    POSITIVE LOGITS
     regards
    0.10
    weeney
    0.10
    amp
    0.09
    èĻIJå¾ħ
    0.09
     regard
    0.09
    æŁ¬
    0.09
    ä¸ªå·¥ä½ľ
    0.08
    éĵĢ
    0.08
    è°Ħ
    0.08
    ç³Ĺ
    0.08
    Activations Density 0.020%

    No Known Activations