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

    Neuronpedia

    INDEX
    Explanations

    references to UI components, specifically boxes and lists

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

    box symbol or code word

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    zarchiwizowane
    -0.45
     Difficult
    -0.45
     Bergmann
    -0.44
    difficult
    -0.43
    Difficult
    -0.41
    ffffffff
    -0.40
     pinulongan
    -0.40
     Heil
    -0.40
    ktm
    -0.40
     <<<<<<<<<<<<<<
    -0.40
    POSITIVE LOGITS
    Box
    2.31
     Box
    1.88
    BOX
    1.54
     BOX
    1.49
     Boxer
    1.15
    box
    1.14
     Boxes
    1.01
    Boxes
    0.99
    obox
    0.95
     box
    0.93
    Act Density 0.004%

    No Known Activations