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

    Neuronpedia

    INDEX
    Explanations

    instances of conditional phrases and hypothetical scenarios

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     very
    -0.16
    stad
    -0.15
    aidu
    -0.15
    kop
    -0.14
     pretty
    -0.14
     quite
    -0.14
     Tran
    -0.14
    alu
    -0.13
     Gat
    -0.13
     deg
    -0.13
    POSITIVE LOGITS
    ông
    0.17
    ỡ
    0.17
     fos
    0.17
    ovaly
    0.16
    سات
    0.16
    instead
    0.15
    Were
    0.15
    iyat
    0.15
    BorderStyle
    0.14
    .habbo
    0.14
    Act Density 0.069%

    No Known Activations