© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130285
    Prev
    Next
    INDEX
    Explanations

    This neuron detects parentheses that enclose numeric expressions (e.g. points where a function is applied to a number).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    mathematical expressions and questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     металли
    -0.93
    zzie
    -0.91
     interactiva
    -0.91
    ovico
    -0.91
    unno
    -0.90
    цуз
    -0.89
     berikutnya
    -0.89
    ävän
    -0.88
     SearchView
    -0.86
    nappropriate
    -0.85
    POSITIVE LOGITS
     when
    0.98
    值
    0.94
     algas
    0.94
     then
    0.92
     Trial
    0.91
    Hors
    0.90
     Hors
    0.89
    当时
    0.88
     tarto
    0.88
     عند
    0.87
    Activations Density 0.031%

    No Known Activations