© 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. 84500
    Prev
    Next
    INDEX
    Explanations

    function keyword in code

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

    This neuron fires on occurrences of the programming keyword “function” (and closely related JavaScript function identifiers).

    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
    蹋
    -2.70
    鶿
    -2.61
    はっきり
    -2.56
    秾
    -2.48
    たまに
    -2.39
    謌
    -2.38
     terrific
    -2.38
     ferocious
    -2.34
    栳
    -2.33
    蹰
    -2.28
    POSITIVE LOGITS
    ing
    4.19
    a
    3.70
    the
    2.95
    ة
    2.70
     provides
    2.66
    
    2.52
    郿
    2.52
     strives
    2.47
     ilumina
    2.47
    ה
    2.45
    Activations Density 0.008%

    No Known Activations