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

    Neuronpedia

    INDEX
    Explanations

    instances of conjunctions and transition phrases

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     Kategor
    -0.84
     lele
    -0.83
     teras
    -0.81
     antik
    -0.79
     kram
    -0.79
     gend
    -0.77
     meras
    -0.76
     panik
    -0.75
     kac
    -0.75
     optik
    -0.75
    POSITIVE LOGITS
     deserves
    0.81
     prolly
    0.80
     certainly
    0.70
     reflects
    0.67
     probably
    0.67
     therefor
    0.67
     contributes
    0.66
     shouldn
    0.66
     shouldnt
    0.65
     reminds
    0.65
    Act Density 0.371%

    No Known Activations