© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 50168
    Prev
    Next
    INDEX
    Explanations

    not certified

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

    The neuron fires on the punctuation-heavy delimiters and separators you see in code (e.g. “.{”, commas between imported symbols, braces) that typically mark structured import lists and boilerplate sections.

    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
    blank
    -0.73
    constructor
    -0.73
    ready
    -0.71
     slim
    -0.69
    éder
    -0.66
    sup
    -0.66
     disappears
    -0.66
     बना
    -0.65
    грузки
    -0.64
    WEBPACK
    -0.63
    POSITIVE LOGITS
    reeg
    0.72
     Kild
    0.71
    口腔
    0.68
     obyvateľov
    0.66
     Ministério
    0.65
    outgoing
    0.65
     Garanti
    0.65
     Aver
    0.65
     Mennon
    0.64
    歳の
    0.63
    Activations Density 0.076%

    No Known Activations