© 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. 34648
    Prev
    Next
    INDEX
    Explanations

    people and places

    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
    вин
    -0.10
    æѤåĪ»
    -0.10
    rvine
    -0.09
    uset
    -0.09
     ÑĢаÑģ
    -0.09
    /problem
    -0.09
    ä½Ĩä¸įéĻIJäºİ
    -0.09
    pected
    -0.09
    WISE
    -0.08
    (separator
    -0.08
    POSITIVE LOGITS
    º
    0.10
    reme
    0.10
    tle
    0.10
    een
    0.09
    ç¾ŀ
    0.09
    åģ¿
    0.09
    mania
    0.09
    اء
    0.09
    cott
    0.09
    aceous
    0.09
    Activations Density 0.028%

    No Known Activations