© 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. 63447
    Prev
    Next
    INDEX
    Explanations

    can take, kill, withhold, could be

    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
    æĢ¥éľĢ
    -0.09
    åĿĬ
    -0.09
    mits
    -0.09
    èµİåĽŀ
    -0.09
    紧迫
    -0.09
     urgency
    -0.08
    ä¿ĥè¿ĽäºĨ
    -0.08
    abee
    -0.08
    è¢ľ
    -0.08
    主åĬ¨æĢ§
    -0.08
    POSITIVE LOGITS
    éļıæŶ
    0.12
    éļ¨æĻĤ
    0.10
    çŀ½
    0.09
    ewe
    0.09
    åıĸè¯ģ
    0.09
    æī§ä¸ļ
    0.08
    亲èĩª
    0.08
    æ³ķå¾ĭ责任
    0.08
    .public
    0.08
    inati
    0.08
    Activations Density 0.037%

    No Known Activations