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

    Neuronpedia

    INDEX
    Explanations

    terms related to accuracy and measurements of precision

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    PathVariable
    -0.76
     Pines
    -0.72
    neté
    -0.66
    jaman
    -0.66
     Soda
    -0.65
     Bro
    -0.64
    Soda
    -0.64
    odly
    -0.63
    mush
    -0.63
     Bot
    -0.63
    POSITIVE LOGITS
     ACC
    1.08
     Acu
    1.03
     accredit
    1.01
     Acorn
    1.01
     Akk
    0.99
     Acme
    0.98
     ACR
    0.97
     accumulator
    0.95
     ACS
    0.94
     Accu
    0.93
    Act Density 0.197%

    No Known Activations