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

    Neuronpedia

    INDEX
    Explanations

    phrases related to sorting or sifting through information or objects

    oai_token-act-pair · gpt-3.5-turbo

    variations of the word "shift"

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

    the word stem "ift" regardless of the word it is part of

    oai_token-act-pair · gpt-4-turbo
    New Auto-Interp
    Negative Logits
    Downloadha
    -0.84
    SOURCE
    -0.73
    nces
    -0.66
    thia
    -0.66
     Incarn
    -0.65
    vid
    -0.63
     CAP
    -0.63
    Winged
    -0.61
    URRENT
    -0.61
     Surviv
    -0.60
    POSITIVE LOGITS
    ift
    1.25
    lift
    1.01
    ifted
    1.00
    ifting
    0.95
    ifts
    0.95
    ifter
    0.90
    ifty
    0.84
    imore
    0.82
    uese
    0.82
    ovember
    0.81
    Act Density 0.016%

    No Known Activations