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

    Neuronpedia

    INDEX
    Explanations

    words containing the letter "m" followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    instances of the letter "m"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    DragonMagazine
    -0.82
    ãĥ´ãĤ¡
    -0.77
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    -0.73
    gerald
    -0.71
    é¾įåĸļ士
    -0.71
     Continued
    -0.70
     Accountability
    -0.68
    NESS
    -0.66
    éĹĺ
    -0.66
     Pars
    -0.65
    POSITIVE LOGITS
    iddling
    1.31
    oses
    1.19
    umps
    1.19
    idd
    1.18
    oths
    1.16
    ugg
    1.15
    ixed
    1.13
    azes
    1.12
    uddled
    1.12
    akings
    1.12
    Act Density 0.021%

    No Known Activations