© 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. 10-GEMMASCOPE-RES-131K
    7. 11316
    Prev
    Next
    INDEX
    Explanations

    code blocks and syntax

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

    function or method definitions, specifically the closing parenthesis and colon before the function body begins.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron activates on numeric literals in code, highlighting tokens that represent numbers.

    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
     라이
    -1.79
     hôte
    -1.66
     appris
    -1.64
     incroyable
    -1.63
     intensif
    -1.61
    ngths
    -1.59
     Economies
    -1.59
     legions
    -1.58
    meanor
    -1.56
     Deleting
    -1.56
    POSITIVE LOGITS
     when
    2.72
     or
    2.45
    ↵↵
    2.08
     they
    1.85
     the
    1.82
    when
    1.82
     we
    1.78
     which
    1.77
     if
    1.73
     these
    1.72
    Activations Density 0.003%

    No Known Activations