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

    Neuronpedia

    INDEX
    Explanations

    adjectives describing something as strange, unusual, and out of the ordinary

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "bizarre."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    onder
    -0.73
    abet
    -0.72
    aper
    -0.72
    ophers
    -0.71
    tes
    -0.70
    claimer
    -0.70
    bern
    -0.70
    payer
    -0.69
    ILA
    -0.69
    getting
    -0.68
    POSITIVE LOGITS
    ly
    1.18
    ishly
    0.96
     bizarre
    0.92
     antics
    0.92
     twists
    0.84
    ulously
    0.82
     anomalies
    0.81
     occurrences
    0.76
     nonsensical
    0.75
    ingly
    0.75
    Act Density 0.011%

    No Known Activations