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

    Neuronpedia

    INDEX
    Explanations

    phrases related to analyzing, processing, and interpreting information

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate emotional processing or contemplation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    thood
    -0.73
    riched
    -0.72
    ¥ŀ
    -0.66
    ãĤ£
    -0.63
    ô
    -0.61
    uci
    -0.61
    izont
    -0.61
    ids
    -0.58
    aldi
    -0.58
     (%)
    -0.58
    POSITIVE LOGITS
     comments
    0.86
     explanations
    0.83
     comment
    0.82
     unanswered
    0.79
     clarification
    0.79
     explanation
    0.78
     hypocrisy
    0.77
     speculation
    0.77
     confirmation
    0.75
     accusations
    0.72
    Act Density 1.037%

    No Known Activations