© 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. 10-GEMMASCOPE-RES-131K
    7. 112511
    Prev
    Next
    INDEX
    Explanations

    code snippets and documentation

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

    HTML/XML comment closing tags, particularly the `-->` sequence.

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

    The neuron activates primarily on code comments and comment delimiters (e.g., “/*…*/” or “<!--…-->”).

    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
     și
    -1.95
     Başkanı
    -1.91
     to
    -1.90
    Ț
    -1.69
     \%(
    -1.64
    meteries
    -1.61
     và
    -1.60
    stata
    -1.59
     that
    -1.58
     AND
    -1.58
    POSITIVE LOGITS
     how
    1.96
    鷽
    1.71
    褙
    1.61
     ,'
    1.52
    8
    1.52
     jezt
    1.51
    ges
    1.50
    do
    1.50
    瞜
    1.48
    does
    1.48
    Activations Density 0.009%

    No Known Activations