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

    Neuronpedia

    INDEX
    Explanations

    the word "very."

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "very"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    osal
    -0.80
    olor
    -0.75
    adelphia
    -0.74
    iture
    -0.70
    imus
    -0.69
    ensis
    -0.68
    sburgh
    -0.67
    onis
    -0.66
    endum
    -0.65
    ãĥ¼ãĥĨãĤ£
    -0.63
    POSITIVE LOGITS
     informative
    0.78
     different
    0.78
     important
    0.77
     expensive
    0.76
     rarely
    0.76
     rare
    0.75
     seldom
    0.75
     handy
    0.74
     similar
    0.74
     readable
    0.73
    Act Density 0.069%

    No Known Activations