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

    Neuronpedia

    INDEX
    Explanations

    words related to challenges, difficulty, or intensity

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of difficulty or challenge

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    uality
    -0.69
    atern
    -0.66
    ership
    -0.65
     Burton
    -0.65
    rompt
    -0.65
     Spect
    -0.64
    itas
    -0.63
     TAG
    -0.62
    ulet
    -0.62
     Griffith
    -0.62
    POSITIVE LOGITS
     hardest
    1.21
    iest
    0.88
     destro
    0.86
     imaginable
    0.84
     toughest
    0.83
     hitter
    0.82
     harder
    0.81
    entimes
    0.75
     easiest
    0.73
     darkest
    0.72
    Act Density 0.003%

    No Known Activations