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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a future prediction or expectation

    oai_token-act-pair · gpt-3.5-turbo

    references to future expectations or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    orthodox
    -0.87
    raid
    -0.85
    ricular
    -0.84
    illusion
    -0.79
    tops
    -0.79
    inguished
    -0.78
    icon
    -0.77
    elsius
    -0.76
    orsi
    -0.76
    ²¾
    -0.76
    POSITIVE LOGITS
     undone
    1.11
     forth
    0.92
     WRITE
    0.84
     out
    0.80
     knocking
    0.80
     ashore
    0.79
     up
    0.76
     INTO
    0.74
     flooding
    0.74
     Forth
    0.74
    Act Density 0.060%

    No Known Activations