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

    Neuronpedia

    INDEX
    Explanations

    past-tense verbs relating to events or situations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the verb "was" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    inav
    -0.80
    enary
    -0.77
     Compare
    -0.64
     Supports
    -0.62
    ï¸ı
    -0.61
     Examples
    -0.61
    inas
    -0.61
     Explore
    -0.61
     Horizons
    -0.61
     Visit
    -0.61
    POSITIVE LOGITS
    hes
    1.17
    wolves
    1.11
    nt
    0.99
     supposed
    0.94
     destined
    0.91
     indeed
    0.90
     never
    0.90
     originally
    0.89
    wolf
    0.87
     deemed
    0.86
    Act Density 0.355%

    No Known Activations