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

    Neuronpedia

    INDEX
    Explanations

    locations and dates

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks, particularly commas, in complex sentence structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    estone
    -0.57
    alogue
    -0.55
    orest
    -0.54
    atively
    -0.54
    estones
    -0.51
    irection
    -0.50
    itialized
    -0.50
    ishly
    -0.50
    ord
    -0.50
    ocl
    -0.49
    POSITIVE LOGITS
     respectively
    0.79
     namely
    0.67
     including
    0.63
     etc
    0.61
     culminating
    0.58
     according
    0.58
     aka
    0.58
     lest
    0.58
     albeit
    0.57
     provoking
    0.56
    Act Density 0.549%

    No Known Activations