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

    Neuronpedia

    INDEX
    Explanations

    phrases related to emotional or descriptive language

    oai_token-act-pair · gpt-3.5-turbo

    expressions of wonder and creativity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ashington
    -0.78
     Admin
    -0.71
     ObamaCare
    -0.66
    utterstock
    -0.65
     Redmond
    -0.65
    EU
    -0.64
     Welfare
    -0.63
     ®
    -0.63
    alert
    -0.62
    BTC
    -0.62
    POSITIVE LOGITS
     ['
    1.35
    laughs
    1.14
     storytelling
    1.08
     [
    1.07
     ["
    1.03
     creatively
    0.93
     melodies
    0.92
     improvis
    0.92
     lyr
    0.91
    Laughs
    0.91
    Act Density 1.068%

    No Known Activations