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

    Neuronpedia

    INDEX
    Explanations

    numeric sequences, specifically with the number 67 repeated within them

    oai_token-act-pair · gpt-3.5-turbo

    numerical values, especially those related to statistics or data points

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     ticking
    -0.67
     Drew
    -0.67
     craft
    -0.63
     shake
    -0.63
     multiple
    -0.63
     shout
    -0.59
     scr
    -0.59
     Counter
    -0.59
     Scrib
    -0.58
    nesday
    -0.58
    POSITIVE LOGITS
    66
    2.94
    67
    2.92
    68
    2.80
    69
    2.68
    71
    2.38
    65
    2.38
    63
    2.19
    70
    2.18
    72
    2.14
    64
    2.13
    Act Density 0.042%

    No Known Activations