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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating quantities or amounts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     habit
    -0.16
    anded
    -0.15
    ilda
    -0.15
    duk
    -0.15
    immel
    -0.14
    år
    -0.14
    çħ
    -0.14
    aser
    -0.14
    bitset
    -0.14
    panse
    -0.14
    POSITIVE LOGITS
    ideo
    0.16
    ulet
    0.16
    æĭĶ
    0.15
    ella
    0.14
    ihn
    0.14
     風
    0.14
    оÑıн
    0.14
     Done
    0.14
     else
    0.13
     besides
    0.13
    Act Density 0.075%

    No Known Activations