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

    Neuronpedia

    INDEX
    Explanations

    sentences with contrasting or opposing viewpoints

    oai_token-act-pair · gpt-3.5-turbo

    indicators of urgent societal issues and problems

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     mos
    -0.66
    MpServer
    -0.64
    Legend
    -0.57
     Started
    -0.56
     english
    -0.56
    igraph
    -0.54
     Chennai
    -0.54
     skins
    -0.54
     haha
    -0.53
     vocals
    -0.53
    POSITIVE LOGITS
     insofar
    0.86
     nonetheless
    0.79
     moreover
    0.77
     undermines
    0.73
     ought
    0.70
     anyway
    0.68
     undermining
    0.68
     undermined
    0.67
     surely
    0.67
     deterrence
    0.67
    Act Density 1.088%

    No Known Activations