© 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. 65796
    Prev
    Next
    INDEX
    Explanations

    one is

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

    The neuron detects numeric tokens (including integers and decimals) representing measurements, quantities, or other numeric data.

    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
    There
    -1.84
    If
    -1.78
    Everyone
    -1.66
    When
    -1.65
    Sometimes
    -1.64
    Those
    -1.62
    Everything
    -1.62
    スコット
    -1.61
    Your
    -1.61
     there
    -1.60
    POSITIVE LOGITS
     a
    2.59
     to
    1.81
     byste
    1.55
    PDOException
    1.50
    .«
    1.49
    ządu
    1.48
    után
    1.47
     that
    1.46
     their
    1.45
     fenti
    1.42
    Activations Density 0.019%

    No Known Activations