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

    Neuronpedia

    INDEX
    Explanations

    references to boxes or box-related concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    box, boxes, boxed, boxy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    CppMethod
    -0.49
    styleType
    -0.43
    HttpException
    -0.41
     Heritage
    -0.40
    setHorizontal
    -0.40
    stateProvider
    -0.40
    NewUrlParser
    -0.39
    OGND
    -0.39
     Thackeray
    -0.39
     Learned
    -0.38
    POSITIVE LOGITS
     box
    2.06
     boxes
    1.88
     Box
    1.87
     BOX
    1.84
     Boxes
    1.78
    Box
    1.77
    box
    1.70
    BOX
    1.63
    boxes
    1.59
    Boxes
    1.59
    Act Density 0.027%

    No Known Activations