© 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. 10-GEMMASCOPE-RES-131K
    4. 86846
    Prev
    Next
    INDEX
    Explanations

    database ids and relationship names

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

    The neuron strongly activates on underscore‐laden tokens (e.g. snake_case table or column names), i.e. programming identifiers containing “_.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     the
    -1.91
     not
    -1.85
     those
    -1.71
     both
    -1.59
    ..”
    -1.41
     Polity
    -1.39
     kwi
    -1.38
     quoc
    -1.35
    vidas
    -1.34
     etui
    -1.33
    POSITIVE LOGITS
    ABCDEF
    1.55
    了过来
    1.47
     его
    1.45
    넋
    1.43
    ן
    1.39
    geslacht
    1.38
    .}(
    1.38
    mjs
    1.37
     แต่
    1.36
     jį
    1.33
    Activations Density 0.029%

    No Known Activations