© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 89730
    Prev
    Next
    INDEX
    Explanations

    Microsoft compiler directives or "original message"

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

    The neuron detects numeric tokens (e.g. version numbers, years, numeric constants) in the code and comments.

    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
     pw
    -0.67
    Norton
    -0.67
    ENABLED
    -0.64
     Turin
    -0.64
     mắt
    -0.63
     Rectangle
    -0.63
     એ
    -0.62
    њ
    -0.62
    Expr
    -0.62
     SIGNIFIC
    -0.61
    POSITIVE LOGITS
    postMessage
    0.67
    znej
    0.66
     орден
    0.65
    ѧ
    0.63
     Lighthouse
    0.63
    GREES
    0.63
     paste
    0.63
    совой
    0.63
     Gil
    0.59
     ря
    0.59
    Activations Density 0.072%

    No Known Activations