© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 71410
    Prev
    Next
    INDEX
    Explanations

    math and technical terms

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

    The neuron selectively activates on numeric expressions—especially decimal numbers and other quantities—wherever they occur in the 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
     on
    -2.44
     for
    -1.91
     every
    -1.67
     one
    -1.61
     most
    -1.59
     made
    -1.53
     number
    -1.53
     all
    -1.53
     like
    -1.47
     several
    -1.41
    POSITIVE LOGITS
    those
    1.63
     THOSE
    1.57
    那些
    1.57
    breviations
    1.54
    urezza
    1.49
    諼
    1.44
    ٧
    1.43
    </strong>
    1.43
     rafraî
    1.42
     kaksi
    1.41
    Activations Density 0.132%

    No Known Activations