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

    Neuronpedia

    INDEX
    Explanations

    years

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    alias
    -0.67
    fecture
    -0.65
    Exit
    -0.63
    atives
    -0.63
    agate
    -0.61
    inqu
    -0.61
    aug
    -0.61
    ife
    -0.60
    atche
    -0.59
     Shack
    -0.59
    POSITIVE LOGITS
     ago
    1.03
    entimes
    0.95
    lasting
    0.77
    ankind
    0.75
     overdue
    0.75
     overlooked
    0.71
     ranging
    0.70
     past
    0.67
     millennia
    0.66
     forgotten
    0.65
    Act Density 10.892%

    No Known Activations