© 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. 49569
    Prev
    Next
    INDEX
    Explanations

    combination of

    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
    è¡Ķ
    -0.11
    æĸ§
    -0.10
    æĸ°çļĦä¸Ģ
    -0.09
    èĿĻ
    -0.08
    ä¿Ń
    -0.08
    æİ¥è¿ŀ
    -0.08
    OTHER
    -0.08
    éĹ´çļĦ
    -0.08
    åħ¨å®¶
    -0.08
    itel
    -0.08
    POSITIVE LOGITS
    /all
    0.11
    æĬĢæľ¯åĴĮ
    0.10
    宿
    0.09
    _FWD
    0.09
    aroo
    0.09
    è·Łä¸įä¸Ĭ
    0.08
    éªĮè¯ģ
    0.08
    ÙħصاÙĦØŃ
    0.08
    ark
    0.08
     discipline
    0.08
    Activations Density 0.028%

    No Known Activations