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

    Neuronpedia

    INDEX
    Explanations

    the repeated mention of the word "three."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    instances of the number "three"

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    IAS
    -0.67
    isky
    -0.63
    Hub
    -0.60
    utable
    -0.59
    UGE
    -0.58
    agate
    -0.58
    gain
    -0.57
    buffer
    -0.57
    aban
    -0.55
    osate
    -0.55
    POSITIVE LOGITS
    teenth
    1.20
    rd
    1.16
    teen
    1.14
     quarters
    1.01
     hundred
    1.00
    peat
    0.99
    uously
    0.98
    quel
    0.91
     successive
    0.89
     weeks
    0.89
    Act Density 0.068%

    No Known Activations