© 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. 72788
    Prev
    Next
    INDEX
    Explanations

    code, equations, and trademarks

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

    This neuron is tuned to detect programming‐style tokens (e.g. code identifiers, file names/extensions, operators, and numeric literals in code).

    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
    ;
    -1.27
    łego
    -1.20
    :
    -1.09
     voran
    -1.01
    NOWLEDGE
    -1.01
    _
    -1.00
    peru
    -0.98
    '
    -0.98
     benötigt
    -0.98
     anstatt
    -0.97
    POSITIVE LOGITS
    geh
    1.20
    muje
    1.17
    ajes
    1.13
     slår
    1.12
     biela
    1.11
     fera
    1.10
     vertes
    1.09
    Jeudi
    1.09
     editText
    1.09
     oferta
    1.09
    Activations Density 0.038%

    No Known Activations