© 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. 110645
    Prev
    Next
    INDEX
    Explanations

    Suppose mathematical questions

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

    The neuron is detecting JSON (or similar) object openings—i.e. the “{” that begins an object.

    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
     começar
    -1.61
    dits
    -1.53
    丝毫
    -1.45
    饜
    -1.41
     Propriété
    -1.38
     garantía
    -1.36
     quatrième
    -1.34
     véhicules
    -1.34
     décennies
    -1.34
     Très
    -1.31
    POSITIVE LOGITS
     are
    1.71
     it
    1.55
     which
    1.49
     the
    1.47
     is
    1.42
     what
    1.41
     by
    1.38
    '].
    1.37
     in
    1.36
     to
    1.32
    Activations Density 0.020%

    No Known Activations