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

    Neuronpedia

    INDEX
    Explanations

    text related to statistics and percentages

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Celt
    -0.64
     cour
    -0.64
    lish
    -0.63
    ified
    -0.61
     Gur
    -0.61
    itiz
    -0.61
     tenant
    -0.59
     treasury
    -0.59
     flakes
    -0.58
     cogn
    -0.55
    POSITIVE LOGITS
    pmwiki
    0.97
     [+
    0.95
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    0.89
    BILITIES
    0.83
     Unloaded
    0.76
    ĸļ
    0.75
     Parables
    0.75
    endif
    0.74
    âķ
    0.74
     Characters
    0.74
    Act Density 0.087%

    No Known Activations