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

    Neuronpedia

    INDEX
    Explanations

    the word "sometimes"

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "sometimes."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    shi
    -0.85
    uid
    -0.78
    yet
    -0.77
    ye
    -0.74
    eki
    -0.72
    zag
    -0.71
    eering
    -0.70
    gur
    -0.70
     Kardashian
    -0.70
    isites
    -0.70
    POSITIVE LOGITS
    entimes
    1.16
    interstitial
    0.88
    theless
    0.88
    etheless
    0.86
     overlooked
    0.85
    asionally
    0.80
     unsus
    0.77
     glimps
    0.77
    igham
    0.74
     mistaken
    0.73
    Act Density 0.014%

    No Known Activations