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

    Neuronpedia

    INDEX
    Explanations

    assertive statements containing the word "tell"

    oai_token-act-pair · gpt-3.5-turbo

    expressions of anticipation or expectation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Almighty
    -0.79
    Kings
    -0.66
    catentry
    -0.64
    Magn
    -0.62
    LIN
    -0.62
    enhagen
    -0.61
     Naples
    -0.58
    Mos
    -0.58
    father
    -0.57
    mone
    -0.57
    POSITIVE LOGITS
     anymore
    1.00
     anywhere
    0.81
     nor
    0.75
     any
    0.74
     anybody
    0.73
     whatsoever
    0.72
     enough
    0.71
    ANY
    0.69
     bothered
    0.69
     bother
    0.69
    Act Density 0.361%

    No Known Activations