© 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. 128265
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric tokens (including integers and decimals), effectively detecting numbers in text.

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

    Volunteers planted at

    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
     mín
    -0.90
     potencial
    -0.90
     pozi
    -0.88
     simbol
    -0.86
     terrasse
    -0.85
     Já
    -0.84
     bater
    -0.84
     andar
    -0.83
     negros
    -0.83
     mami
    -0.81
    POSITIVE LOGITS
     gestionar
    0.90
     limpeza
    0.90
     بتوان
    0.88
     recarg
    0.87
     schuld
    0.87
    onais
    0.87
    chait
    0.86
     sellado
    0.84
     incredibly
    0.83
    timbangkan
    0.83
    Activations Density 0.024%

    No Known Activations