© 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. 10-GEMMASCOPE-RES-131K
    7. 99663
    Prev
    Next
    INDEX
    Explanations

    TypeScript code definitions

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

    The neuron fires on opening curly‐brace tokens that introduce new code blocks (e.g. function bodies, getters/setters).

    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
     or
    -2.16
    ;
    -1.84
     our
    -1.80
     when
    -1.70
    K
    -1.69
     which
    -1.68
    $
    -1.63
     having
    -1.63
     making
    -1.61
     on
    -1.59
    POSITIVE LOGITS
     apesar
    1.88
     extremamente
    1.74
    sandalia
    1.70
    ʺ
    1.56
    oth
    1.56
    をよく
    1.55
     transparan
    1.55
     abenço
    1.55
    edì
    1.53
     atualmente
    1.52
    Activations Density 0.008%

    No Known Activations