© 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. 128979
    Prev
    Next
    INDEX
    Explanations

    This neuron looks for uses of the “.” member‐access operator (i.e. object property or method calls via a dot) in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    code execution and object manipulation

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     by
    -2.42
     their
    -2.36
     first
    -2.33
     all
    -2.28
     not
    -2.27
     just
    -2.14
     on
    -2.11
     people
    -2.09
     had
    -2.08
     an
    -1.99
    POSITIVE LOGITS
     trám
    2.36
     verhind
    1.91
    ᛬
    1.79
     REALLY
    1.78
     ส์
    1.77
     aprobó
    1.77
     weihnachten
    1.74
     Dinas
    1.73
     绿色
    1.73
     relâche
    1.72
    Activations Density 0.007%

    No Known Activations