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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a specific quantity or limitation

    oai_token-act-pair · gpt-3.5-turbo

    the repetitive phrase "only" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     regardless
    -0.72
     anytime
    -0.70
    tails
    -0.67
    insula
    -0.66
    axis
    -0.65
     irrespective
    -0.64
    abol
    -0.62
    ahead
    -0.62
     includ
    -0.62
    align
    -0.60
    POSITIVE LOGITS
     rudimentary
    0.91
     minimal
    0.91
     marginally
    0.89
     minor
    0.88
     ONE
    0.87
     curs
    0.87
     sporadic
    0.86
     modest
    0.85
     superficial
    0.84
     occasional
    0.79
    Act Density 0.094%

    No Known Activations