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

    Neuronpedia

    INDEX
    Explanations

    numerical values preceded by a dollar sign

    oai_token-act-pair · gpt-3.5-turbo

    monetary values, particularly those in a specific numeric format

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     50
    -0.80
     100
    -0.75
     72
    -0.74
     70
    -0.71
     60
    -0.71
     73
    -0.71
     300
    -0.71
     200
    -0.71
     74
    -0.70
     51
    -0.69
    POSITIVE LOGITS
    15
    2.72
    16
    2.50
    14
    2.46
    17
    2.41
    13
    2.30
    12
    2.15
    18
    2.14
    11
    2.14
    10
    2.03
    19
    2.00
    Act Density 0.038%

    No Known Activations