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

    Neuronpedia

    INDEX
    Explanations

    ASCII art

    oai_token-act-pair · gpt-3.5-turbo

    patterns or symbols relating to numerical data or mathematical expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     mathemat
    -0.68
     credential
    -0.67
     subcontract
    -0.67
     agre
    -0.65
     Cosponsors
    -0.64
     harness
    -0.63
     hiber
    -0.63
     reconc
    -0.63
    cffff
    -0.62
     princ
    -0.61
    POSITIVE LOGITS
     Carbuncle
    0.72
    wise
    0.72
     attRot
    0.72
     istg
    0.71
    ercise
    0.68
     Jinn
    0.67
    opl
    0.65
    enko
    0.64
     Zhao
    0.63
     xx
    0.63
    Act Density 0.063%

    No Known Activations