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

    Neuronpedia

    INDEX
    Explanations

    physical actions and interactions in a story

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    onite
    -1.00
    uristic
    -0.91
     Pros
    -0.89
    IJ
    -0.87
     hallmark
    -0.86
     Impl
    -0.85
     alike
    -0.81
     precursor
    -0.81
    pson
    -0.79
     transpl
    -0.79
    POSITIVE LOGITS
     safely
    1.02
     Sabha
    1.01
    stage
    0.99
    robe
    0.95
     bed
    0.93
    plane
    0.93
     peacefully
    0.91
     tripod
    0.90
     stairs
    0.89
     unst
    0.89
    Act Density 1.461%

    No Known Activations