© 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. 103341
    Prev
    Next
    INDEX
    Explanations

    math and programming libraries

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

    the word "Math" in code and technical contexts, particularly in library names, class names, and namespaces.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires on the literal token “Math,” i.e. occurrences of the identifier or library/class name “Math” in code or technical text.

    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
    灕
    -3.13
    ⽂
    -2.69
    線が
    -2.69
    vantaged
    -2.66
     mobilya
    -2.61
    cuál
    -2.56
     dekoratif
    -2.56
    幼稚
    -2.56
    犛
    -2.56
    飧
    -2.55
    POSITIVE LOGITS
    2
    4.47
    3
    4.34
    4
    4.09
    1
    3.98
    5
    3.97
    7
    3.94
    6
    3.72
    In
    3.52
    a
    3.47
    0
    3.16
    Activations Density 0.003%

    No Known Activations