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

    Neuronpedia

    INDEX
    Explanations

    terms that emphasize a comparison relationship

    oai_token-act-pair · gpt-3.5-turbo

    phrases that discuss various kinds of relationships or connections

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    UFF
    -0.87
    oker
    -0.85
    ECK
    -0.83
    ucket
    -0.82
    owler
    -0.79
    baum
    -0.75
    fly
    -0.74
    utters
    -0.74
    itters
    -0.74
    iolet
    -0.73
    POSITIVE LOGITS
    hips
    0.84
     relations
    0.78
    ability
    0.78
     relation
    0.74
     thereto
    0.74
    borough
    0.73
    eering
    0.72
    gradient
    0.70
    ally
    0.69
    ually
    0.68
    Act Density 0.025%

    No Known Activations