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

    Neuronpedia

    INDEX
    Explanations

    phrases or sentences posing questions

    oai_token-act-pair · gpt-3.5-turbo

    questions that start with "Would."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    values
    -0.66
     Nieto
    -0.64
    story
    -0.63
    Ready
    -0.62
     Practices
    -0.60
     Kag
    -0.60
    Reporting
    -0.60
    Writing
    -0.60
    resources
    -0.60
     Cairo
    -0.59
    POSITIVE LOGITS
     imply
    0.91
     suffice
    0.90
    yip
    0.89
     doubtless
    0.88
     surely
    0.88
     undoubtedly
    0.87
     be
    0.83
     introduce
    0.82
     proble
    0.81
     definitely
    0.81
    Act Density 0.117%

    No Known Activations