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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating improvement or change

    oai_token-act-pair · gpt-3.5-turbo

    terms related to improvements or positive changes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     compuls
    -0.66
    itu
    -0.63
    agate
    -0.62
    ictionary
    -0.60
    rive
    -0.59
     contracting
    -0.59
    amina
    -0.58
    ictions
    -0.58
    angling
    -0.57
    Sources
    -0.56
    POSITIVE LOGITS
     compared
    0.97
    iator
    0.80
     leaps
    0.79
    over
    0.76
    atform
    0.76
    undown
    0.72
     drastically
    0.71
     leap
    0.67
     from
    0.67
    inion
    0.66
    Act Density 0.125%

    No Known Activations