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

    Neuronpedia

    INDEX
    Explanations

    the word "woefully"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "wo" and its variations in context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Uriel
    -0.77
     Immunity
    -0.74
     Buckley
    -0.65
     dissolve
    -0.65
     Sunshine
    -0.64
     Democracy
    -0.64
     Carbuncle
    -0.63
     Integrity
    -0.63
     Fract
    -0.62
    spr
    -0.62
    POSITIVE LOGITS
    efully
    1.18
     wo
    1.08
    ollen
    1.01
    eful
    0.95
    ofer
    0.88
    wo
    0.88
    jo
    0.87
    oks
    0.85
    ocom
    0.83
    merce
    0.81
    Act Density 0.008%

    No Known Activations