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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating critical judgment or evaluation

    oai_token-act-pair · gpt-3.5-turbo

    phrases conveying dissatisfaction or negative evaluations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rather
    -0.65
    alm
    -0.61
    olutely
    -0.60
    azar
    -0.59
    ription
    -0.59
    usra
    -0.57
    esses
    -0.57
    cientious
    -0.56
    etermined
    -0.56
    eely
    -0.56
    POSITIVE LOGITS
     anymore
    1.72
     nor
    1.33
     either
    1.02
     whatsoever
    0.90
    nor
    0.86
     yet
    0.86
     anywhere
    0.85
    yet
    0.85
     Enough
    0.78
     necessarily
    0.78
    Act Density 0.293%

    No Known Activations