© 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. 22-GEMMASCOPE-RES-131K
    7. 129092
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric tokens (standalone digits or multi-digit numbers) in the text.

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

    HTTP status codes

    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
    Lunes
    -1.10
     Gebir
    -1.09
     અને
    -1.09
    btnCancel
    -1.08
    bodied
    -1.07
    ensitive
    -1.07
     bambou
    -1.04
    Alerta
    -1.02
    thodox
    -1.02
     এবং
    -1.02
    POSITIVE LOGITS
     it
    1.02
     not
    0.98
     răng
    0.97
     nothing
    0.97
    request
    0.94
     textAlign
    0.91
     trying
    0.90
     terribly
    0.89
     profoundly
    0.88
     Fresco
    0.87
    Activations Density 0.003%

    No Known Activations