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

    Neuronpedia

    INDEX
    Explanations

    instances of the word 'while' in the text

    oai_token-act-pair · gpt-3.5-turbo

    instances of contrast or simultaneous conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    aer
    -0.91
    hack
    -0.86
    Bas
    -0.79
    ogen
    -0.76
    atter
    -0.74
    iotic
    -0.74
    urable
    -0.73
    oux
    -0.73
    enter
    -0.73
    bard
    -0.72
    POSITIVE LOGITS
     browsing
    1.02
     touring
    0.93
     researching
    0.90
     attending
    0.88
     acknowledging
    0.88
     maintaining
    0.87
     compiling
    0.83
     commuting
    0.82
     ignoring
    0.82
     mentioning
    0.81
    Act Density 0.057%

    No Known Activations