© 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. 26226
    Prev
    Next
    INDEX
    Explanations

    field names and identifiers

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

    This neuron detects qualified names in code (identifiers followed by a “.”), such as table-alias or schema prefixes in SQL (“cl.”, “nsp.”, “tbl1.”, etc.).

    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
    čki
    -0.91
    qiang
    -0.88
    ोड
    -0.88
    ckeye
    -0.87
    arised
    -0.86
    ႆ
    -0.85
    Čo
    -0.84
    valdi
    -0.84
     และ
    -0.84
    other
    -0.84
    POSITIVE LOGITS
     name
    1.16
     pessoas
    1.02
    เดียว
    0.95
     jotka
    0.95
     estudiantes
    0.93
     metų
    0.91
     مرة
    0.90
     glossy
    0.89
     miembros
    0.87
     existe
    0.87
    Activations Density 0.009%

    No Known Activations