© 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. 15291
    Prev
    Next
    INDEX
    Explanations

    number | betting | research

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

    The neuron is primarily detecting numeric expressions (including numbers, decimals, odds, and related punctuation) 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
     usta
    -1.32
    時点
    -1.30
     französische
    -1.28
     decorativo
    -1.24
    /...
    -1.21
     daz
    -1.20
     affects
    -1.20
    ...:
    -1.20
     umo
    -1.19
     haup
    -1.19
    POSITIVE LOGITS
     in
    1.41
     now
    1.17
    numberOfLines
    1.15
     škola
    1.13
    glises
    1.13
     momentan
    1.10
     gillar
    1.09
    setResizable
    1.08
     Menschheit
    1.08
     Siempre
    1.07
    Activations Density 0.010%

    No Known Activations