© 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. 34-GEMMASCOPE-RES-131K
    7. 90887
    Prev
    Next
    INDEX
    Explanations

    programming code and specifications

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

    The neuron fires on tokens that mark high-level structural or metadata elements in source code (e.g. preprocessor directives like “endif,” license/header words like “file” and “notice,” “Suite” names, and declaration keywords like “public”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     دارید
    -0.84
    磁盘
    -0.83
    alık
    -0.79
    levy
    -0.77
    ensively
    -0.77
    を見せて
    -0.77
     politische
    -0.76
     disdain
    -0.76
     Roanoke
    -0.76
     http
    -0.76
    POSITIVE LOGITS
    Ỳ
    0.76
     heterosexual
    0.75
     podrá
    0.75
    Paolo
    0.72
    eleri
    0.71
    ỏi
    0.70
    انج
    0.69
    umos
    0.69
     finalidad
    0.68
    цеп
    0.67
    Activations Density 0.000%

    No Known Activations