© 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. 17852
    Prev
    Next
    INDEX
    Explanations

    global variable

    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
    å±Ĥå±Ĥ
    -0.09
    mob
    -0.09
    å·¥ä½ľå²Ĺä½į
    -0.09
    é½IJ
    -0.08
     Petty
    -0.08
    icy
    -0.08
    èı©æıIJ
    -0.08
    acy
    -0.08
    olist
    -0.08
     меÑħаниз
    -0.08
    POSITIVE LOGITS
    (Global
    0.15
    /global
    0.11
    è§Ĩéĩİ
    0.11
    ToLocal
    0.11
    æīĭèĦļ
    0.09
    -local
    0.09
    (global
    0.09
    æıĸ
    0.09
    尺度
    0.09
    éĹ»åIJį
    0.09
    Activations Density 0.093%

    No Known Activations