© 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. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 59724
    Prev
    Next
    INDEX
    Explanations

    mathematical formulas and notation

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects commas used as argument‐separators (e.g. “i,\;j” or “k,\;n–k”) inside LaTeX/math expressions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    した
    -2.48
    是想
    -2.31
     своему
    -2.17
    珖
    -2.16
    鲀
    -2.16
    Ꮉ
    -2.16
    谞
    -2.14
    珰
    -2.13
    %.
    -2.11
    你可以
    -2.09
    POSITIVE LOGITS
    }$
    2.94
    [
    2.91
    =\
    2.27
    窳
    2.23
    after
    2.20
    ,\
    2.20
    b
    2.20
    a
    2.16
    矇
    2.14
     bén
    2.13
    Activations Density 0.003%

    No Known Activations