© 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. 34336
    Prev
    Next
    INDEX
    Explanations

    numbers, units, or measurements

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

    This neuron activates on numeric tokens—digits or numbers (including measurements, sizes, frequencies, and other quantitative values).

    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
    為に
    -2.11
    mathrm
    -1.79
    に追加
    -1.72
    さぁ
    -1.58
    事が
    -1.58
    VILLE
    -1.50
    期刊
    -1.50
    ほら
    -1.47
    事で
    -1.47
    It
    -1.45
    POSITIVE LOGITS
    があります
    2.03
     in
    1.82
    があり
    1.70
    もある
    1.69
    がある
    1.69
     to
    1.66
    があるので
    1.63
    などは
    1.58
    がありますが
    1.50
    もあって
    1.48
    Activations Density 0.022%

    No Known Activations