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

    Neuronpedia

    INDEX
    Explanations

    text related to legal issues and criminal activities

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    士
    -1.07
    isible
    -0.83
    âĺĨ
    -0.83
    tis
    -0.81
    inas
    -0.79
     Auditor
    -0.77
    livion
    -0.77
     Stadium
    -0.77
    ordan
    -0.76
     univers
    -0.76
    POSITIVE LOGITS
     inappropriately
    1.00
     allegedly
    0.95
     activities
    0.90
     sexually
    0.87
     aliases
    0.85
     advising
    0.83
     Manafort
    0.82
     Breitbart
    0.81
     Zy
    0.80
     pornographic
    0.80
    Act Density 5.132%

    No Known Activations