© 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. 103978
    Prev
    Next
    INDEX
    Explanations

    quotes and brackets

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

    the opening quotation mark of a field or column name in database schemas and structured data definitions.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron activates on numeric tokens containing a decimal point—i.e. floating-point literals.

    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
     siè
    -1.87
    rfan
    -1.74
    definitely
    -1.72
     rigue
    -1.68
    wonderful
    -1.68
    extremely
    -1.65
    certainly
    -1.64
    hever
    -1.62
     JOHN
    -1.61
    absolutely
    -1.61
    POSITIVE LOGITS
     which
    2.41
     that
    2.22
     there
    1.83
     their
    1.82
     how
    1.76
     unique
    1.73
     for
    1.70
     has
    1.69
     it
    1.67
     time
    1.67
    Activations Density 0.041%

    No Known Activations