© 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. 110968
    Prev
    Next
    INDEX
    Explanations

    text or code snippets

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

    line breaks or newlines in chat logs, code, and other structured text formats.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron detects standalone exclamatory interjections or emphasis markers (e.g. “dang,” “BINGO,” “Holy Cow,” ellipses, 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_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
     by
    -1.74
     his
    -1.57
    detek
    -1.45
     each
    -1.36
     your
    -1.34
    Dentist
    -1.30
     the
    -1.28
    Perú
    -1.27
     through
    -1.24
     should
    -1.23
    POSITIVE LOGITS
    ="
    1.98
    ("
    1.62
    $
    1.57
    5
    1.43
    ='
    1.43
    就是要
    1.39
    [
    1.38
    4
    1.34
     begge
    1.32
    һ
    1.32
    Activations Density 0.078%

    No Known Activations