© 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. 71522
    Prev
    Next
    INDEX
    Explanations

    mathematical notation and variables

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

    This neuron activates on tokens that are parts of inline mathematical expressions—especially variable names with bracketed indices or subscripts (e.g. “a[”, “n]”, “y[n]”), arithmetic operators, and related symbols.

    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
    niki
    -1.09
    															
    -1.03
     argumenta
    -1.03
    纰
    -1.01
    startGame
    -1.00
     Piscina
    -1.00
    idé
    -0.99
     seda
    -0.98
     növény
    -0.96
    ské
    -0.96
    POSITIVE LOGITS
     nhất
    1.02
    =""></
    0.96
    🫵
    0.95
     повин
    0.95
     based
    0.95
     confided
    0.95
     seguente
    0.94
     our
    0.91
     fortalec
    0.91
     previous
    0.91
    Activations Density 0.004%

    No Known Activations