© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 92532
    Prev
    Next
    INDEX
    Explanations

    code and scientific writing

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

    The neuron strongly activates on programming‐style tokens—identifiers, method names, and punctuation (e.g. parentheses, dots) found 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_34/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
    Spin
    -0.71
    lexer
    -0.70
    spin
    -0.68
    colors
    -0.68
    orange
    -0.66
    OrderService
    -0.64
     Delgado
    -0.64
     schwarz
    -0.63
    řed
    -0.63
    適
    -0.63
    POSITIVE LOGITS
     isLoading
    0.72
    哚
    0.72
    hasMoreElements
    0.68
    mitas
    0.68
    xas
    0.67
    곽
    0.67
     flanks
    0.66
    extérieur
    0.66
    freund
    0.66
    vate
    0.65
    Activations Density 0.085%

    No Known Activations