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

    Neuronpedia

    INDEX
    Explanations

    phrases related to hypothetical or potential situations

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating potential outcomes or hypotheticals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Lawrence
    -0.65
     Ste
    -0.64
     Down
    -0.62
     Goose
    -0.62
     Contrast
    -0.61
     Feld
    -0.61
     reception
    -0.61
     contrasting
    -0.61
     Drawn
    -0.60
     Lod
    -0.60
    POSITIVE LOGITS
    be
    1.30
    issue
    1.00
    say
    0.96
    thinking
    0.95
    minded
    0.93
    always
    0.92
    bes
    0.90
    dos
    0.90
    bodied
    0.90
    formation
    0.90
    Act Density 0.024%

    No Known Activations