© 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. 128151
    Prev
    Next
    INDEX
    Explanations

    opening parenthesis followed by word

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

    The neuron fires on the appearance of parentheses (i.e. "(" or ")" tokens), especially indicating a function call or parenthetical phrase.

    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
    菝
    -2.61
     tocó
    -2.41
    涐
    -2.11
    ֢
    -2.06
    ڧ
    -2.05
     schöner
    -1.99
    我們
    -1.99
    硖
    -1.98
    !</
    -1.98
    </sup>
    -1.94
    POSITIVE LOGITS
    ’
    3.14
     essas
    2.30
    ,,,
    2.17
     \
    2.08
     ส์
    2.05
    ꜆
    2.05
    翃
    1.96
    ,.
    1.95
    啬
    1.91
     mauva
    1.91
    Activations Density 0.003%

    No Known Activations