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

    Neuronpedia

    INDEX
    Explanations

    the word "despite" followed by a description or contrast

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate contrasting situations or circumstances

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    oided
    -0.76
    unless
    -0.74
    itored
    -0.70
    aires
    -0.69
    bernatorial
    -0.66
    earchers
    -0.65
    .-
    -0.64
    elfth
    -0.64
    opus
    -0.64
    ribe
    -0.64
    POSITIVE LOGITS
     these
    1.07
     foregoing
    0.98
     this
    0.88
     growing
    0.82
     hindsight
    0.81
     setbacks
    0.81
     widespread
    0.80
     declining
    0.80
     advancements
    0.79
     advances
    0.79
    Act Density 0.133%

    No Known Activations