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

    Neuronpedia

    INDEX
    Explanations

    sentences that address the reader directly with "you"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    onom
    -0.15
     Uncomment
    -0.14
    ãĤ¤ãĥ³ãĥĪ
    -0.14
    Å¡en
    -0.14
    avel
    -0.13
    åºĦ
    -0.13
    odÃŃ
    -0.13
     вдÑĢÑĥг
    -0.13
    åĩºæĿ¥
    -0.13
    undry
    -0.13
    POSITIVE LOGITS
     forgot
    0.20
     said
    0.20
     mileage
    0.18
     sir
    0.17
     stated
    0.17
     could
    0.17
     haven
    0.17
     mention
    0.16
     mean
    0.16
    forgot
    0.15
    Act Density 0.061%

    No Known Activations