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

    Neuronpedia

    INDEX
    Explanations

    information related to numerical data, such as percentages, figures, and statistics

    oai_token-act-pair · gpt-3.5-turbo

    coordinating conjunctions and specific percentage values

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Leaks
    -0.78
    OTA
    -0.75
    irie
    -0.69
    imil
    -0.65
    Times
    -0.64
    Planet
    -0.64
    meet
    -0.63
    ħĭ
    -0.62
    lock
    -0.61
    OWS
    -0.61
    POSITIVE LOGITS
     thus
    0.88
     consequently
    0.86
     then
    0.84
     hence
    0.81
     thereby
    0.80
     thence
    0.75
     furthermore
    0.74
     culmin
    0.74
     therefore
    0.74
     averaged
    0.73
    Act Density 0.234%

    No Known Activations