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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating contrast or distinguishing facts

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating negation or dismissal

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Presence
    -0.71
     redesign
    -0.66
    akening
    -0.63
    arity
    -0.63
    pedia
    -0.63
    ulas
    -0.59
    overe
    -0.58
    Preview
    -0.58
    gur
    -0.58
     reintrodu
    -0.57
    POSITIVE LOGITS
     whatsoever
    0.92
    THING
    0.85
    affles
    0.73
    ij士
    0.65
    ahu
    0.65
     sudden
    0.65
     us
    0.63
     answers
    0.62
     batted
    0.62
     JUSTICE
    0.61
    Act Density 0.040%

    No Known Activations