© 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. 72167
    Prev
    Next
    INDEX
    Explanations

    code class names and programming constructs

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

    This neuron detects code tokens that represent class or type names (e.g. identifiers in CamelCase used for types or constructors).

    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
    mbda
    -1.17
    aucune
    -1.11
     either
    -1.09
     publicidade
    -1.09
    équipements
    -1.09
     осіб
    -1.04
     some
    -1.03
    tvguide
    -1.02
    umblr
    -0.99
     Gehäuse
    -0.99
    POSITIVE LOGITS
     anormal
    1.13
     ваше
    1.05
    LayoutStyle
    0.98
     Loon
    0.97
    蕎麦
    0.97
    Marie
    0.95
     plafon
    0.95
     रु
    0.94
     placer
    0.94
    Му
    0.93
    Activations Density 0.005%

    No Known Activations