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

    Neuronpedia

    INDEX
    Explanations

    sentences indicating future events or plans

    oai_token-act-pair · gpt-3.5-turbo

    repeated usage of the phrase "will be."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Bars
    -0.67
    ciating
    -0.61
    arming
    -0.61
    might
    -0.61
     compose
    -0.60
    afia
    -0.60
    plex
    -0.59
    INTON
    -0.59
     proposal
    -0.59
    artments
    -0.59
    POSITIVE LOGITS
     able
    1.02
    fall
    1.02
    heading
    0.99
    AMS
    0.95
     rewarded
    0.94
     judged
    0.93
     remembered
    0.91
     seen
    0.88
     replaced
    0.87
    falls
    0.86
    Act Density 0.172%

    No Known Activations