© 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. 57380
    Prev
    Next
    INDEX
    Explanations

    dodgeball, dodge city, avoiding danger

    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
    毫ä¸įçĬ¹è±«
    -0.09
    åĿļå®ļ
    -0.08
    è·ı
    -0.08
    tlement
    -0.08
    究
    -0.08
    forced
    -0.07
    æķĻè®Ń
    -0.07
    uetype
    -0.07
    åĨ¤
    -0.07
    æİİ
    -0.07
    POSITIVE LOGITS
    åĽŀéģ¿
    0.11
    躲
    0.10
    æİīäºĨ
    0.10
    èĩªå¦Ĥ
    0.09
    éĤ¨
    0.09
    ä¸ĢåıĮ
    0.09
    æİī
    0.09
    ç¾ļ
    0.09
    pard
    0.09
    éģ¿å¼Ģ
    0.09
    Activations Density 0.007%

    No Known Activations