© 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. 22-GEMMASCOPE-RES-131K
    7. 14978
    Prev
    Next
    INDEX
    Explanations

    mathematical notation and definitions

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

    This neuron detects definitional language in math text—phrases that introduce or explain a symbol, e.g. “is the coordinate derivative” in a definition.

    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
    ження
    -0.99
    ods
    -0.95
     заключается
    -0.93
    ssysteme
    -0.92
    €™
    -0.91
     NotImplemented
    -0.91
    ;-
    -0.91
     said
    -0.90
    زم
    -0.89
    ң
    -0.88
    POSITIVE LOGITS
    Алексей
    1.07
    Новый
    1.06
    初始
    1.05
     constater
    1.02
    Евгений
    1.00
    Ведущий
    0.94
    Подготовка
    0.90
     craindre
    0.87
     fråga
    0.86
    Уважа
    0.85
    Activations Density 0.021%

    No Known Activations