© 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. 27430
    Prev
    Next
    INDEX
    Explanations

    how much/many/widespread

    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
    antal
    -0.10
    -Pack
    -0.09
    ç«ŀ
    -0.09
    ä¸į失
    -0.09
    asad
    -0.08
     Roose
    -0.08
    FSIZE
    -0.08
    æĦıåij³çĿĢ
    -0.08
    ä¸įçŁ¥
    -0.08
     Público
    -0.08
    POSITIVE LOGITS
    æĿ¥çļĦ
    0.11
    è¿Ľå±ķ
    0.10
     progress
    0.10
    å®ŀéĻħæĥħåĨµ
    0.09
    ä¸Ĭå®ĺ
    0.08
     ÑģебÑı
    0.08
    croft
    0.08
    大å¥ĸ
    0.08
    ught
    0.08
     Progress
    0.08
    Activations Density 0.052%

    No Known Activations