© 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. 22-GEMMASCOPE-RES-131K
    4. 199
    Prev
    Next
    INDEX
    Explanations

    function definitions

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

    This neuron activates on single-letter identifiers immediately followed by an open parenthesis, i.e. function names or calls in the mathematical notation.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    معرفی
    -0.98
    moderne
    -0.96
     farol
    -0.96
    akteur
    -0.92
    魅力
    -0.92
    adventures
    -0.90
    észet
    -0.90
    ideos
    -0.89
    توضیحات
    -0.88
    描述
    -0.86
    POSITIVE LOGITS
     that
    1.07
     hopes
    0.96
    羁
    0.94
    気になった
    0.93
    に行く
    0.90
    ั
    0.89
    $__
    0.88
    辿
    0.88
    րա
    0.86
    了一会儿
    0.86
    Activations Density 0.007%

    No Known Activations