© 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. 34631
    Prev
    Next
    INDEX
    Explanations

    conjunctions and explanatory terms

    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
    parsed
    -0.10
    æ½®
    -0.09
    èĢģé¼ł
    -0.09
    atem
    -0.08
    æľĿ
    -0.08
    è¿ĺ
    -0.08
    itest
    -0.08
    usta
    -0.08
    itas
    -0.08
    oment
    -0.08
    POSITIVE LOGITS
    oola
    0.09
    éĺģ
    0.09
    æ¯ķ
    0.08
    èĢģ人家
    0.08
    Ùħؤ
    0.08
    æľĢ好æĺ¯
    0.08
    ð¥
    0.08
    èĭ¥æĤ¨
    0.08
     prer
    0.08
    ADER
    0.08
    Activations Density 0.092%

    No Known Activations