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

    Neuronpedia

    INDEX
    Explanations

    quotations and punctuations like commas and periods

    oai_token-act-pair · gpt-3.5-turbo

    punctuation and periods signaling the end of statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     quir
    -0.91
     flo
    -0.90
     glim
    -0.90
     tack
    -0.87
     tossing
    -0.79
     hopped
    -0.78
     grun
    -0.78
     jarring
    -0.77
     uncom
    -0.77
     bourbon
    -0.75
    POSITIVE LOGITS
     Therefore
    1.80
     Hence
    1.67
    Therefore
    1.64
     Moreover
    1.64
     Besides
    1.57
     Afterwards
    1.51
     Otherwise
    1.46
     Furthermore
    1.45
    Moreover
    1.41
     Secondly
    1.41
    Act Density 0.464%

    No Known Activations