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

    Neuronpedia

    INDEX
    Explanations

    personal expressions of opinions or emotions

    oai_token-act-pair · gpt-3.5-turbo

    expressions of personal opinions and emotions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Triumph
    -0.72
     Hutch
    -0.67
     Glover
    -0.67
     overturned
    -0.65
     Blooming
    -0.65
     Goodwin
    -0.64
     staffed
    -0.62
     Wiggins
    -0.62
     Malaysia
    -0.60
     constructed
    -0.59
    POSITIVE LOGITS
     displeasure
    0.88
    urances
    0.77
    rompt
    0.77
     condolences
    0.76
    ociation
    0.76
    Letter
    0.75
     actionGroup
    0.74
     superiority
    0.73
     disapproval
    0.73
     preference
    0.71
    Act Density 0.140%

    No Known Activations