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

    Neuronpedia

    INDEX
    Explanations

    phrases related to interactions or events occurring in a narrative or story

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    assies
    -0.88
    hops
    -0.79
    events
    -0.77
    itars
    -0.74
    ources
    -0.72
    rooms
    -0.72
    inances
    -0.71
    ernels
    -0.70
    ummies
    -0.70
    lees
    -0.69
    POSITIVE LOGITS
     standpoint
    0.88
     version
    0.86
     perspective
    0.85
     portion
    0.82
     sized
    0.81
     subset
    0.79
     piece
    0.79
     swath
    0.77
     shell
    0.77
     chunk
    0.77
    Act Density 3.530%

    No Known Activations