© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 70908
    Prev
    Next
    INDEX
    Explanations

    programming languages and phrases

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

    The neuron fires on code‐style punctuation that closes or terminates statements or blocks—e.g. closing braces, parentheses, and semicolons.

    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
    TokenNameDOT
    -1.09
     共
    -1.06
     garota
    -1.04
    from
    -1.02
    ße
    -1.01
    were
    -1.00
    but
    -0.99
     微
    -0.99
     STATE
    -0.97
    焼く
    -0.97
    POSITIVE LOGITS
     ему
    1.05
    之时
    1.02
    的结果
    0.98
    スズキ
    0.98
    Конечно
    0.97
     powinien
    0.96
    鐽
    0.96
     that
    0.96
    必须要
    0.95
    ハロウィン
    0.95
    Activations Density 0.000%

    No Known Activations