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

    Neuronpedia

    INDEX
    Explanations

    contractions of "can not" and "can"

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     Américas
    -0.62
     bakso
    -0.56
     ruinas
    -0.56
     Résultats
    -0.54
     Catedral
    -0.54
     parlamento
    -0.53
     churras
    -0.53
    Formazione
    -0.52
     Aé
    -0.52
     levure
    -0.52
    POSITIVE LOGITS
     disreg
    0.93
     affor
    0.82
    
    0.78
     reluct
    0.75
     impra
    0.73
     resear
    0.73
     increa
    0.71
     suscep
    0.70
     pollut
    0.69
     excru
    0.69
    Act Density 0.202%

    No Known Activations