© 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. 39956
    Prev
    Next
    INDEX
    Explanations

    ribbons and bows

    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
    ç¢Į
    -0.10
    çŀ»
    -0.09
    çĭłæĬĵ
    -0.09
    èĿ¼
    -0.09
    åı¯ä»¥è®©
    -0.09
    _COLORS
    -0.09
    igure
    -0.08
    ueling
    -0.08
    orce
    -0.08
    oola
    -0.08
    POSITIVE LOGITS
    åĬ¨æľº
    0.10
    解å¼Ģ
    0.09
    icker
    0.08
    åŁİéķĩ
    0.08
    åŁł
    0.08
    æī£
    0.08
    åĩĩ
    0.08
    ç¼ļ
    0.08
    çħ¤æ°Ķ
    0.08
    Extras
    0.08
    Activations Density 0.052%

    No Known Activations