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

    Neuronpedia

    INDEX
    Explanations

    phrases related to discussing a specific topic or subject

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ©¶æ¥µ
    -0.87
    ô
    -0.78
    eps
    -0.78
    phia
    -0.77
    Ò
    -0.76
    ĸļ
    -0.74
    cffff
    -0.73
    marine
    -0.72
    tap
    -0.71
    ``
    -0.71
    POSITIVE LOGITS
     specifics
    0.85
     questions
    0.82
     fairness
    0.76
     example
    0.74
     excuses
    0.73
     comparisons
    0.71
     resolving
    0.71
     other
    0.71
     why
    0.70
     reviewing
    0.70
    Act Density 0.021%

    No Known Activations