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

    Neuronpedia

    INDEX
    Explanations

    phrases or specific actions preceded by the word "that"

    oai_token-act-pair · gpt-3.5-turbo

    blocks of text or paragraphs devoid of specific content

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    bledon
    -0.61
     Examiner
    -0.60
     Seym
    -0.59
     Borders
    -0.57
    erenn
    -0.56
     Frie
    -0.56
     Depot
    -0.55
    Ire
    -0.54
    Adams
    -0.53
     Sahara
    -0.53
    POSITIVE LOGITS
     violates
    0.83
     lasted
    0.77
     translates
    0.75
     consists
    0.75
     doesnt
    0.74
     includes
    0.74
     entails
    0.72
     resembles
    0.72
     utilizes
    0.72
     consisted
    0.71
    Act Density 0.054%

    No Known Activations