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

    Neuronpedia

    INDEX
    Explanations

    comparisons between different items or entities

    oai_token-act-pair · gpt-3.5-turbo

    comparative phrases expressing contrast or opposition

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    collar
    -0.81
    ERN
    -0.80
    overed
    -0.76
    eur
    -0.76
    abet
    -0.75
    red
    -0.75
    mberg
    -0.75
    olog
    -0.74
    lene
    -0.72
    yright
    -0.72
    POSITIVE LOGITS
     mindset
    0.66
     scissors
    0.65
     apples
    0.65
    nesday
    0.64
     nil
    0.63
     bandits
    0.63
     averages
    0.63
     cannabin
    0.62
     monkeys
    0.62
     situational
    0.61
    Act Density 0.014%

    No Known Activations