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

    Neuronpedia

    INDEX
    Explanations

    references to breaks or interruptions in activity or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĤįãģĨ
    -0.16
    tae
    -0.14
    reh
    -0.14
    achel
    -0.14
    oom
    -0.14
    Undo
    -0.14
    loser
    -0.14
     late
    -0.14
    isha
    -0.13
     Undo
    -0.13
    POSITIVE LOGITS
     hiatus
    0.35
    ä¼ij
    0.30
    gap
    0.24
     absence
    0.23
     gap
    0.21
     odst
    0.21
    -gap
    0.20
     INTERRUPTION
    0.20
    Gap
    0.20
     suspension
    0.20
    Act Density 0.197%

    No Known Activations