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

    Neuronpedia

    INDEX
    Explanations

    conditional statements expressing hypothetical situations

    oai_token-act-pair · gpt-3.5-turbo

    conditional phrases starting with "If."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     across
    -0.73
     sustained
    -0.71
     largely
    -0.70
     nearly
    -0.68
     around
    -0.67
     finished
    -0.67
     rose
    -0.67
     present
    -0.67
     mostly
    -0.65
     aboard
    -0.65
    POSITIVE LOGITS
    If
    2.85
    Unless
    2.09
    Otherwise
    2.04
    Whether
    1.99
     If
    1.87
    When
    1.81
    Either
    1.80
    Should
    1.77
    Alternatively
    1.74
    Once
    1.74
    Act Density 0.028%

    No Known Activations