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

    Neuronpedia

    INDEX
    Explanations

    every instance of the word "exactly"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "exactly" to emphasize precision or specificity in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ker
    -0.80
    rug
    -0.77
    rift
    -0.76
    olyn
    -0.73
     sacrific
    -0.65
    asta
    -0.64
    kers
    -0.64
    cler
    -0.63
    gers
    -0.63
    isson
    -0.63
    POSITIVE LOGITS
     opposite
    0.74
    ãĤ¨
    0.73
     wrong
    0.72
    ãĥ´ãĤ¡
    0.67
     aligned
    0.67
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    0.65
    æ©Ł
    0.65
     suited
    0.63
    ÃĽ
    0.63
     tuned
    0.62
    Act Density 0.014%

    No Known Activations