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

    Neuronpedia

    INDEX
    Explanations

    words related to states of confusion or uncertainty

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     bye
    -0.80
     deviations
    -0.77
     Policies
    -0.72
     Equity
    -0.72
     equity
    -0.70
     Order
    -0.68
     guidelines
    -0.67
     Guidelines
    -0.67
     reconciliation
    -0.63
     roles
    -0.63
    POSITIVE LOGITS
    ingly
    1.40
    ciating
    1.05
    ament
    1.02
    stru
    1.00
    iously
    1.00
    ient
    0.92
    ulous
    0.92
    ruciating
    0.91
    vic
    0.91
    amental
    0.90
    Act Density 0.121%

    No Known Activations