© 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. 73504
    Prev
    Next
    INDEX
    Explanations

    percentage values and variations

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

    The neuron selectively activates on numeric tokens—i.e. it picks out numbers (especially percentages or other quantitative values) 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
     παιδιά
    -0.86
    TryGetValue
    -0.86
    cão
    -0.82
     logiciel
    -0.81
    回事
    -0.81
     ruido
    -0.76
    ふる
    -0.76
    arão
    -0.75
    꽝
    -0.75
     recibir
    -0.74
    POSITIVE LOGITS
     percent
    1.27
     percentage
    1.14
     Percent
    1.13
    Percent
    0.97
     (%)
    0.95
     प्रतिशत
    0.95
    越大
    0.91
    %，
    0.90
    percent
    0.87
     Percentage
    0.86
    Activations Density 0.043%

    No Known Activations