© 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. 5789
    Prev
    Next
    INDEX
    Explanations

    Frewater, Computers, nubs

    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
    æ·±åħ¥äººå¿ĥ
    -0.09
    注æĺİ
    -0.09
    çļĦåį°è±¡
    -0.09
    æľĽ
    -0.09
     tink
    -0.09
    æħķ
    -0.08
    æľīä¸Ģ
    -0.08
    欢åĸľ
    -0.08
    绩
    -0.08
    çļĦ认åı¯
    -0.08
    POSITIVE LOGITS
     dial
    0.09
    гаÑĢ
    0.09
    æĴ¥
    0.09
    ZO
    0.09
    adena
    0.09
    ollo
    0.09
     Pun
    0.09
    wipe
    0.09
    ÑĦик
    0.09
    CDC
    0.08
    Activations Density 3.238%

    No Known Activations