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

    Neuronpedia

    INDEX
    Explanations

    phrases of praise or criticism

    oai_token-act-pair · gpt-3.5-turbo

    phrases of high praise or critical assessment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    externalActionCode
    -0.85
    isen
    -0.68
     Repeat
    -0.62
    tails
    -0.62
     Exxon
    -0.61
    iland
    -0.59
    interrupted
    -0.59
    rief
    -0.58
    OTUS
    -0.58
    lyak
    -0.57
    POSITIVE LOGITS
     virtues
    0.81
     unbeat
    0.75
    uilt
    0.74
     defic
    0.70
     gobl
    0.68
     greatness
    0.67
     evils
    0.66
     toughness
    0.63
     superiority
    0.62
     cumbers
    0.61
    Act Density 0.525%

    No Known Activations