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

    Neuronpedia

    INDEX
    Explanations

    phrases related to technology instructions

    oai_token-act-pair · gpt-3.5-turbo

    frequent punctuation or comma usage in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    jit
    -0.72
    Said
    -0.70
    ¬¼
    -0.66
    Amazing
    -0.65
    minster
    -0.64
    aha
    -0.64
    ãĤ´ãĥ³
    -0.63
     Advertisement
    -0.62
     Worse
    -0.60
    tained
    -0.60
    POSITIVE LOGITS
     please
    1.14
     you
    0.99
     choose
    0.90
    please
    0.90
     proceed
    0.89
     however
    0.89
     we
    0.88
     whichever
    0.87
     subtract
    0.85
     specify
    0.83
    Act Density 0.197%

    No Known Activations