© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26
    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 22502
    Prev
    Next
    INDEX
    Explanations

    one or two

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

    The neuron detects approximate small-quantity expressions, particularly the phrase “one or two.”

    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
    {
    -2.42
    A
    -2.05
    E
    -2.03
    So
    -1.99
    [
    -1.92
    -1.91
    But
    -1.90
     But
    -1.88
    1
    -1.88
    The
    -1.85
    POSITIVE LOGITS
     of
    2.55
    tación
    1.99
    ámicas
    1.93
     zodat
    1.81
    maakt
    1.73
    liefert
    1.73
    handeling
    1.73
    zookeeper
    1.73
    AutoresizingMask
    1.73
     significados
    1.73
    Activations Density 0.006%

    No Known Activations