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

    Neuronpedia

    INDEX
    Explanations

    phrases related to vertical positioning or alignment

    oai_token-act-pair · gpt-3.5-turbo

    instances related to vertical structure or orientation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    lude
    -0.74
    geries
    -0.73
    deen
    -0.71
    gate
    -0.70
    Lib
    -0.69
    taker
    -0.69
    ifer
    -0.69
    cia
    -0.68
    etta
    -0.66
    bane
    -0.66
    POSITIVE LOGITS
     vertical
    3.45
     horizontal
    2.65
     Vertical
    2.54
     vertically
    2.38
    Vert
    1.92
     horizontally
    1.91
     horizont
    1.77
     diagonal
    1.59
    izontal
    1.52
     perpendicular
    1.44
    Act Density 0.012%

    No Known Activations