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

    Neuronpedia

    INDEX
    Explanations

    negative opinions or critiques, particularly related to political figures

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     maksi
    -1.07
     kosme
    -1.01
     alkoh
    -1.00
     kompati
    -0.94
     akut
    -0.92
     silikon
    -0.92
     radikal
    -0.91
     keramik
    -0.90
     optik
    -0.90
     Kategor
    -0.86
    POSITIVE LOGITS
     serious
    0.89
     actual
    0.85
    actual
    0.82
    serious
    0.81
     genuine
    0.80
     real
    0.75
     ACTUAL
    0.73
     seriousness
    0.72
    Actual
    0.69
     genuinely
    0.68
    Act Density 0.658%

    No Known Activations