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

    Neuronpedia

    INDEX
    Explanations

    phrases related to statements or opinions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "saying."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    visible
    -0.79
    estern
    -0.77
    peg
    -0.71
    200000
    -0.71
    isible
    -0.69
    wn
    -0.68
    ocument
    -0.68
    ammy
    -0.68
    transfer
    -0.65
    èª
    -0.64
    POSITIVE LOGITS
     Pitch
    0.62
     they
    0.62
     VK
    0.60
    omers
    0.60
     apart
    0.60
     Af
    0.57
     therein
    0.57
     hello
    0.56
     Brand
    0.56
     ISPs
    0.56
    Act Density 0.099%

    No Known Activations