Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    a specific block of text about unique items

    oai_token-act-pair · gemini-2.0-flash

    words indicating uniqueness

    np_token-act-pair-logits · gpt-4o-mini

    unique

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    RenderAtEndOf
    -0.81
    AddTagHelper
    -0.79
    contentLoaded
    -0.76
    -0.75
     AssemblyCompany
    -0.75
     autorytatywna
    -0.72
    parsedMessage
    -0.71
     Paglinawan
    -0.71
    AndEndTag
    -0.69
    mergeFrom
    -0.69
    POSITIVE LOGITS
     uni
    0.93
     once
    0.93
     Uni
    0.88
     einmal
    0.82
     unique
    0.80
    once
    0.78
     única
    0.78
     único
    0.78
     singular
    0.78
     One
    0.77
    Act Density 1.060%

    No Known Activations