© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 1749
    Prev
    Next
    INDEX
    Explanations

    be followed by article/adjective

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

    The neuron selectively activates on numeric tokens—years, counts, measurements, or other numbers—wherever they appear 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_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
     and
    -2.16
     elig
    -1.50
     seleccionar
    -1.49
     folgendes
    -1.49
     varandra
    -1.47
    lese
    -1.46
    quirer
    -1.44
     acri
    -1.43
     Kiev
    -1.42
     Gebruik
    -1.41
    POSITIVE LOGITS
     a
    4.28
     an
    2.56
     their
    2.53
     some
    2.38
     the
    2.11
     about
    2.05
     konkrét
    1.95
     that
    1.91
     our
    1.89
     márm
    1.77
    Activations Density 0.042%

    No Known Activations