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

    Neuronpedia

    INDEX
    Explanations

    words related to dissatisfaction or frustration

    oai_token-act-pair · gpt-3.5-turbo

    references to emotional reactions and social interactions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     GOODMAN
    -0.70
    itionally
    -0.66
    roximately
    -0.66
    ocument
    -0.65
     certification
    -0.64
    encia
    -0.60
    rowth
    -0.58
     certified
    -0.57
    aceae
    -0.56
    hess
    -0.56
    POSITIVE LOGITS
     decency
    0.80
     clich
    0.77
     lest
    0.76
     coward
    0.76
     cynicism
    0.72
     uttered
    0.70
     nihil
    0.70
     insults
    0.69
     inco
    0.69
     modesty
    0.68
    Act Density 1.756%

    No Known Activations