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

    Neuronpedia

    INDEX
    Explanations

    medical terms and research methodology related to clinical trials

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     disagre
    -1.62
     embodi
    -1.58
     emphat
    -1.56
     reluct
    -1.54
     uncin
    -1.52
     affor
    -1.50
     squa
    -1.50
     increa
    -1.50
     perfon
    -1.49
     suscep
    -1.48
    POSITIVE LOGITS
     Then
    1.02
     They
    0.92
     When
    0.91
     But
    0.90
     While
    0.89
     However
    0.88
     It
    0.87
     That
    0.87
     ***!
    0.86
     After
    0.85
    Act Density 0.476%

    No Known Activations