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

    Neuronpedia

    INDEX
    Explanations

    sentences ending with periods

    oai_token-act-pair · gpt-3.5-turbo

    sentences or statements that convey overall assessments or conclusions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     sustained
    -0.79
     contested
    -0.76
    escription
    -0.76
     moderates
    -0.73
     affili
    -0.72
     alleged
    -0.72
     incon
    -0.70
     projected
    -0.70
    iannopoulos
    -0.70
     sanctions
    -0.69
    POSITIVE LOGITS
     Luckily
    1.61
     Thankfully
    1.50
     Fortunately
    1.46
     Especially
    1.36
     Depending
    1.33
     Unless
    1.29
     Unfortunately
    1.26
     Ideally
    1.25
     Sometimes
    1.23
     Usually
    1.23
    Act Density 0.344%

    No Known Activations