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

    Neuronpedia

    INDEX
    Explanations

    numerical patterns related to hexadecimal values

    oai_token-act-pair · gpt-3.5-turbo

    numerical data and code snippets involving character representations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ĸļ
    -0.67
     omission
    -0.64
    execute
    -0.57
    isance
    -0.54
    issance
    -0.54
    dule
    -0.53
    xual
    -0.53
    soDeliveryDate
    -0.53
    200000
    -0.51
    atural
    -0.50
    POSITIVE LOGITS
    reau
    0.66
    nir
    0.57
    agall
    0.55
    atten
    0.54
     Ferry
    0.52
     vegetables
    0.49
     FG
    0.49
    stan
    0.48
    agles
    0.48
     Islands
    0.48
    Act Density 1.009%

    No Known Activations