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

    Neuronpedia

    INDEX
    Explanations

    percentage values

    oai_token-act-pair · gpt-3.5-turbo

    percentages or rates expressed per hundred

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    phis
    -0.61
    apesh
    -0.61
     Tau
    -0.59
     suspensions
    -0.59
    upon
    -0.59
    RAW
    -0.58
    aspers
    -0.57
    !/
    -0.56
     Dungeons
    -0.55
     testimonies
    -0.55
    POSITIVE LOGITS
     cent
    1.08
     cubic
    0.81
     square
    0.80
     dollar
    0.79
     capita
    0.79
     ounce
    0.79
    imeters
    0.78
    billion
    0.78
     pound
    0.77
    malink
    0.74
    Act Density 0.028%

    No Known Activations