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

    Neuronpedia

    INDEX
    Explanations

    verbs related to perception like "looked", "sounded", and "seemed"

    oai_token-act-pair · gpt-3.5-turbo

    expressions of perception or opinion related to situations or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    por
    -0.85
    cedented
    -0.83
    opus
    -0.77
    veland
    -0.75
    enta
    -0.74
    eston
    -0.70
    yet
    -0.70
    arta
    -0.69
    enegger
    -0.66
    currently
    -0.66
    POSITIVE LOGITS
    DEV
    0.80
     initially
    0.73
     originally
    0.71
     beforehand
    0.66
     Soviet
    0.66
    wolves
    0.65
     earlier
    0.65
     tremend
    0.62
     unsuccessfully
    0.61
     behavi
    0.60
    Act Density 0.697%

    No Known Activations