© 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. 28344
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions with functions and variables

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

    This neuron detects variable names and symbols in mathematical expressions (e.g. LaTeX‐style math tokens like E, X, f, x, brackets, parentheses).

    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
     marchand
    -1.09
     django
    -1.05
    vereignty
    -1.04
    izate
    -1.04
     merve
    -1.02
     sélectionné
    -0.98
     ۱۰
    -0.98
     магазин
    -0.97
     ۱۷
    -0.97
     \"
    -0.97
    POSITIVE LOGITS
     there
    1.77
    There
    1.50
     There
    1.34
     at
    1.26
     because
    1.23
     our
    1.23
     it
    1.23
    there
    1.16
     when
    1.15
     with
    1.12
    Activations Density 0.003%

    No Known Activations