© 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. 22-GEMMASCOPE-RES-131K
    7. 26320
    Prev
    Next
    INDEX
    Explanations

    Android UI initialization and events

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

    This neuron fires strongly on the “(” token in code, i.e. it detects open‐parenthesis characters (typically in method signatures or calls).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     now
    -0.98
    并在
    -0.96
    setLevel
    -0.92
     agree
    -0.88
     меры
    -0.86
     there
    -0.86
    れません
    -0.85
    рою
    -0.85
    now
    -0.84
    begin
    -0.83
    POSITIVE LOGITS
    Compared
    1.00
     compared
    0.98
    лича
    0.96
    fotografía
    0.96
    tamentos
    0.96
     Compared
    0.95
     podéis
    0.95
    Ajay
    0.94
    加拿
    0.93
    を探す
    0.91
    Activations Density 0.004%

    No Known Activations