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

    Neuronpedia

    INDEX
    Explanations

    phrases related to negative consequences or problems

    oai_token-act-pair · gpt-3.5-turbo

    mentions of public health and environmental concerns

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    answered
    -0.68
     ********************************
    -0.61
     Thu
    -0.58
    arial
    -0.58
     Lit
    -0.57
    interstitial
    -0.56
    \/
    -0.55
     Identification
    -0.55
    =\"
    -0.54
     Quote
    -0.54
    POSITIVE LOGITS
     but
    1.31
    but
    1.30
    But
    0.96
     BUT
    0.94
     But
    0.91
     nor
    0.79
    }{
    0.76
    BUT
    0.74
    })
    0.73
    though
    0.73
    Act Density 0.307%

    No Known Activations