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

    Neuronpedia

    INDEX
    Explanations

    phrases related to controversial incidents or cases

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    aye
    -0.70
     liking
    -0.70
     vowel
    -0.67
    hhhh
    -0.67
     weights
    -0.67
    idity
    -0.65
     vow
    -0.64
    urances
    -0.64
     salaries
    -0.63
     interfaces
    -0.63
    POSITIVE LOGITS
     unsolved
    1.25
     tragic
    1.20
     unfolding
    1.17
     recounted
    1.03
     tragedies
    1.01
     traumat
    0.97
     traumatic
    0.96
     heartbreaking
    0.96
     sympt
    0.96
    Scene
    0.95
    Act Density 0.330%

    No Known Activations