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

    Neuronpedia

    INDEX
    Explanations

    sentences related to tragic events and personal or family-related statements

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    thal
    -0.66
     Keynes
    -0.64
     caffe
    -0.64
    izational
    -0.62
    indal
    -0.62
    acebook
    -0.62
     Era
    -0.61
     stimulating
    -0.60
    sis
    -0.59
    rd
    -0.59
    POSITIVE LOGITS
     allege
    0.84
    rieve
    0.79
    rieved
    0.77
    rieving
    0.73
    hetical
    0.73
     plead
    0.72
     grieving
    0.72
     mourn
    0.72
    arten
    0.69
     objected
    0.69
    Act Density 13.259%

    No Known Activations