© 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. 50333
    Prev
    Next
    INDEX
    Explanations

    adhesive sticking taping

    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
    icken
    -0.10
    æ´Ĺè¡£
    -0.09
    ondo
    -0.09
    涮
    -0.08
    ç¼ĺåĪĨ
    -0.08
     closet
    -0.08
     Barnett
    -0.08
    èĸ®
    -0.08
    æĬĬæīĭ
    -0.08
    ä¸Ģç¬Ķ
    -0.08
    POSITIVE LOGITS
    pire
    0.10
     Vict
    0.09
    å¼Ľ
    0.09
    deen
    0.09
    Sad
    0.08
     remot
    0.08
    红æĹĹ
    0.08
    XS
    0.08
    boards
    0.08
    _STD
    0.08
    Activations Density 0.033%

    No Known Activations