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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a specific point in time

    oai_token-act-pair · gpt-3.5-turbo

    the repeated usage of the word "this."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    isms
    -0.71
    aws
    -0.68
    cakes
    -0.67
    mist
    -0.66
    iably
    -0.66
    pots
    -0.65
     Remain
    -0.64
     actionGroup
    -0.62
    mans
    -0.62
    gas
    -0.62
    POSITIVE LOGITS
     article
    0.79
     scenario
    0.79
     regard
    0.77
     particular
    0.77
     week
    0.74
     latter
    0.73
     predicament
    0.73
     newfound
    0.72
     episode
    0.71
     circumstance
    0.69
    Act Density 0.126%

    No Known Activations