© 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. 10-GEMMASCOPE-RES-131K
    7. 112207
    Prev
    Next
    INDEX
    Explanations

    defining mathematical expressions

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

    This neuron detects numerical and arithmetic-expression tokens (digits, fractions, decimals, and math operators).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ſo
    -2.25
    ――――
    -2.23
     formid
    -2.22
     geene
    -2.09
    ◆◆
    -2.08
     แต่
    -2.06
    potentially
    -2.03
     prestig
    -2.02
     incont
    -1.99
     Eſ
    -1.98
    POSITIVE LOGITS
     with
    2.69
     when
    2.44
     five
    2.42
     because
    2.41
     if
    2.36
     also
    2.34
     which
    2.33
     on
    2.28
     give
    2.25
     before
    2.22
    Activations Density 0.359%

    No Known Activations