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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "previously"

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "previously."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    tin
    -0.60
    mop
    -0.60
     alike
    -0.58
    opl
    -0.57
    rail
    -0.57
    bers
    -0.56
    ________________
    -0.56
    rates
    -0.56
    haw
    -0.55
    bies
    -0.55
    POSITIVE LOGITS
     previously
    3.48
     hitherto
    2.32
     formerly
    2.24
     Previously
    1.92
     previous
    1.87
     originally
    1.79
     recently
    1.67
    Previously
    1.65
     earlier
    1.63
     initially
    1.43
    Act Density 0.013%

    No Known Activations