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

    Neuronpedia

    INDEX
    Explanations

    phrases that convey conditional statements or exceptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    326
    -0.16
    uma
    -0.16
    809
    -0.15
    ajar
    -0.14
    826
    -0.14
    ummies
    -0.14
    dera
    -0.14
     Lair
    -0.14
    urdy
    -0.14
    quets
    -0.14
    POSITIVE LOGITS
     otherwise
    0.24
     specifically
    0.23
    otherwise
    0.22
     explicitly
    0.21
     OTHERWISE
    0.19
     expressly
    0.18
     Otherwise
    0.18
    specific
    0.18
    explicit
    0.18
     especÃŃf
    0.18
    Act Density 0.041%

    No Known Activations