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

    Neuronpedia

    INDEX
    Explanations

    instances where the text discusses a topic in relation to a specific area or aspect

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to specific contexts or conditions affecting policy

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    also
    -0.83
     ALSO
    -0.77
     likewise
    -0.75
     gladly
    -0.73
    yeah
    -0.73
    only
    -0.72
     accordingly
    -0.71
    successfully
    -0.69
    whatever
    -0.68
    ciation
    -0.67
    POSITIVE LOGITS
     regards
    1.15
     regard
    1.05
     relation
    0.92
     areas
    0.88
     cases
    0.86
     terms
    0.85
     respect
    0.80
     rural
    0.77
     light
    0.77
     younger
    0.76
    Act Density 0.186%

    No Known Activations