© 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. 15787
    Prev
    Next
    INDEX
    Explanations

    class definitions

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

    This neuron fires on code-definition keywords, especially the "class" (and occasionally "def") tokens that introduce definitions in source 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
     évé
    -1.52
    ttää
    -1.52
    蚰
    -1.49
     struktur
    -1.45
     capot
    -1.45
     Američ
    -1.41
    Jaunes
    -1.40
     bluz
    -1.38
    贋
    -1.38
     pneus
    -1.38
    POSITIVE LOGITS
    L
    1.95
    h
    1.91
    N
    1.90
    K
    1.89
    D
    1.89
    k
    1.87
    V
    1.86
    m
    1.82
    g
    1.81
    w
    1.81
    Activations Density 0.003%

    No Known Activations