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

    Neuronpedia

    INDEX
    Explanations

    programming code elements and functions like loops, conditional statements, and method calls

    oai_token-act-pair · gpt-3.5-turbo

    numeric characters and their arrangement in data structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Satanic
    -0.68
     bunk
    -0.67
     McGee
    -0.64
     underage
    -0.64
    ihadi
    -0.62
     Osc
    -0.62
     Boll
    -0.61
     incorpor
    -0.60
     trolling
    -0.59
     screenings
    -0.59
    POSITIVE LOGITS
    {
    2.16
     {
    2.15
    {{
    2.01
    {\
    1.84
     {{
    1.73
     {"
    1.56
     {\
    1.55
     {}
    1.54
    ={
    1.51
     {*
    1.50
    Act Density 0.034%

    No Known Activations