© 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. 22-GEMMASCOPE-RES-131K
    7. 129954
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on numeric tokens (digits or numbers) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    key role

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     interesses
    -1.22
     telefonu
    -1.13
    に到着
    -1.05
     habitaciones
    -1.04
     diberikan
    -1.03
     menyediakan
    -1.02
     функция
    -1.02
     filozof
    -1.01
    Сла
    -1.00
     habile
    -0.99
    POSITIVE LOGITS
     previous
    1.02
     was
    0.95
     tested
    0.92
    ithub
    0.90
     research
    0.89
    „
    0.89
    зу
    0.88
     previously
    0.88
     study
    0.87
    Booth
    0.87
    Activations Density 0.010%

    No Known Activations