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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "dive" and related terms

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "dive" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     ACTIONS
    -0.73
    Percent
    -0.70
    minus
    -0.64
    weather
    -0.64
    ika
    -0.63
     Eliot
    -0.61
    VALUE
    -0.61
    numbered
    -0.61
     Mechdragon
    -0.61
    Transfer
    -0.61
    POSITIVE LOGITS
     Dive
    1.02
     diving
    0.99
     dive
    0.97
    ulic
    0.91
    tails
    0.91
    tail
    0.84
     deeper
    0.83
    river
    0.82
     dives
    0.81
     underwater
    0.77
    Act Density 0.025%

    No Known Activations