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

    Neuronpedia

    INDEX
    Explanations

    statistics and numerical figures

    oai_token-act-pair · gpt-3.5-turbo

    instances of the conjunction "and."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    actionDate
    -0.77
    ESE
    -0.75
    meric
    -0.74
    Hide
    -0.70
    OTA
    -0.69
    NVIDIA
    -0.68
    Gaza
    -0.68
    Unity
    -0.67
    Leaks
    -0.66
    LI
    -0.65
    POSITIVE LOGITS
     consequently
    0.92
     thereby
    0.89
     then
    0.89
     thus
    0.89
     therefore
    0.88
     hence
    0.85
     furthermore
    0.84
     counting
    0.83
     onwards
    0.79
     moreover
    0.77
    Act Density 0.336%

    No Known Activations