© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38945
    Prev
    Next
    INDEX
    Explanations

    mathematical notation and symbols

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

    The neuron strongly activates on LaTeX math operator symbols—especially the minus‐sign/hyphen token (and to a lesser extent the caret “^” for superscripts).

    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
    s
    -1.17
    sy
    -1.06
    ates
    -1.04
     南瓜
    -1.02
    ann
    -0.99
     美人
    -0.99
     detalladas
    -0.99
     пользователей
    -0.98
     financieras
    -0.98
    $(
    -0.97
    POSITIVE LOGITS
     generan
    1.11
    crème
    1.08
     the
    1.06
    enang
    1.05
     conoci
    1.04
     depicting
    1.02
    ΢
    1.00
    upuk
    0.99
    想像
    0.99
     forming
    0.97
    Activations Density 0.003%

    No Known Activations