© 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. 22-GEMMASCOPE-RES-131K
    7. 39547
    Prev
    Next
    INDEX
    Explanations

    what you are

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

    The neuron activates on numeric tokens (whole or decimal numbers) 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_22/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
     célè
    -1.82
     bénéfic
    -1.72
     empê
    -1.67
     炖
    -1.67
    corrhi
    -1.63
     trajets
    -1.63
     éta
    -1.62
    Ꮴ
    -1.61
     éton
    -1.60
    おそ
    -1.60
    POSITIVE LOGITS
     even
    1.92
     just
    1.59
     all
    1.58
     everything
    1.52
    .....
    1.45
    ....
    1.44
     generally
    1.44
     ....
    1.43
     etc
    1.42
     T
    1.41
    Activations Density 0.024%

    No Known Activations