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

    Neuronpedia

    INDEX
    Explanations

    references to physical bearing, weight, or characteristics

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "bear" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    anwhile
    -0.78
    lectic
    -0.72
    inx
    -0.71
    arta
    -0.71
    enta
    -0.70
    committee
    -0.69
    arters
    -0.66
    phis
    -0.66
    ugu
    -0.66
    ablishment
    -0.66
    POSITIVE LOGITS
    bear
    1.02
     cub
    0.88
     claws
    0.87
     bear
    0.87
     Bears
    0.87
     Grizz
    0.83
     bears
    0.82
     paws
    0.82
    beit
    0.80
     paw
    0.79
    Act Density 0.011%

    No Known Activations