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

    Neuronpedia

    INDEX
    Explanations

    relational words expressing opinions or emotions

    oai_token-act-pair · gpt-3.5-turbo

    expressions of personal feelings and emotions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Camer
    -0.71
    adish
    -0.68
    atile
    -0.67
     Agricultural
    -0.66
     Display
    -0.65
     Shutterstock
    -0.65
     Extras
    -0.64
    licensed
    -0.64
    photos
    -0.63
    esan
    -0.62
    POSITIVE LOGITS
     comfortable
    0.96
     compelled
    0.90
     uneasy
    0.87
     passionately
    0.87
     vibrations
    0.81
     strongly
    0.80
     uncomfortable
    0.79
    igue
    0.78
    ulz
    0.76
    rils
    0.76
    Act Density 0.036%

    No Known Activations