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

    Neuronpedia

    INDEX
    Explanations

    quotes from political figures

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks that often indicate dialogue or speech

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     tremend
    -0.74
    urat
    -0.71
    ridor
    -0.68
    izo
    -0.68
    æ©
    -0.68
     equivalents
    -0.66
    abytes
    -0.65
     fres
    -0.65
    izons
    -0.63
    avorite
    -0.63
    POSITIVE LOGITS
     '[
    0.78
     Countdown
    0.73
     Cancel
    0.72
     Actor
    0.71
     '
    0.70
     Appears
    0.70
     Bringing
    0.70
     Saying
    0.70
     Doesn
    0.69
    Fact
    0.68
    Act Density 0.057%

    No Known Activations