© 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. 22539
    Prev
    Next
    INDEX
    Explanations

    gradient and transparency

    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
    éļ¾åħ³
    -0.10
     Laud
    -0.09
    å½ĵãģŁãĤĬ
    -0.09
    èĴ¿
    -0.09
    unal
    -0.09
    /portfolio
    -0.08
    ikal
    -0.08
     audi
    -0.08
    ä»ĺåĩº
    -0.08
    代
    -0.08
    POSITIVE LOGITS
    uttle
    0.10
    é³ı
    0.09
     tủ
    0.09
    樽
    0.09
    åħ¬æĬ¥
    0.09
    etting
    0.09
     fur
    0.09
     Teh
    0.09
     Fur
    0.08
    ä¿¡æģ¯æľįåĬ¡
    0.08
    Activations Density 0.141%

    No Known Activations