© 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. 10-GEMMASCOPE-RES-131K
    7. 129004
    Prev
    Next
    INDEX
    Explanations

    numbers containing 5

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

    The neuron fires on numeric tokens beginning with “35” (e.g. “35-year-old,” “35 years,” “35th anniversary,” or numbers like “350”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     diversos
    -2.42
     curtail
    -2.20
     determinados
    -2.11
     Incidentally
    -2.06
     Envío
    -2.05
     Publicar
    -1.97
     Seorang
    -1.96
     seleccionamos
    -1.95
     avowed
    -1.95
     Something
    -1.95
    POSITIVE LOGITS
    m
    2.58
    ’
    2.23
    .”
    2.19
    ،
    2.05
    w
    1.99
    ()
    1.95
    ;
    1.94
     ‘
    1.93
     excelente
    1.90
    <>
    1.89
    Activations Density 0.023%

    No Known Activations