© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 103254
    Prev
    Next
    INDEX
    Explanations

    free software redistribute Copyright

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

    This neuron strongly activates on numeric tokens (strings of digits) 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_34/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
    лах
    -0.74
     Pendidikan
    -0.73
    ňte
    -0.65
    谑
    -0.63
    _;
    -0.62
    zell
    -0.61
    horizontalLayout
    -0.60
    Derby
    -0.60
     家具
    -0.59
     най
    -0.59
    POSITIVE LOGITS
     Lupus
    0.81
     kí
    0.79
    arching
    0.78
    msk
    0.77
    ckets
    0.72
     STEP
    0.71
    areas
    0.71
    IDENTITY
    0.70
     LOU
    0.69
     Frog
    0.69
    Activations Density 0.048%

    No Known Activations