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

    Neuronpedia

    INDEX
    Explanations

    adverbs describing the intensity of a condition or emotion

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "very."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    adelphia
    -0.84
    olor
    -0.81
    onis
    -0.75
    osal
    -0.72
    ourses
    -0.71
     Cheong
    -0.70
    amel
    -0.69
    ensis
    -0.69
    sburgh
    -0.68
    iture
    -0.68
    POSITIVE LOGITS
     important
    0.96
     informative
    0.93
     difficult
    0.93
     unlikely
    0.92
     exciting
    0.90
     rare
    0.89
     busy
    0.89
     expensive
    0.88
     interesting
    0.88
     readable
    0.87
    Act Density 0.069%

    No Known Activations