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

    Neuronpedia

    INDEX
    Explanations

    concluding statements, often starting with "After all" and followed by a related statement

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the phrase "after all."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    pec
    -0.77
    MpServer
    -0.72
    agnetic
    -0.67
    ificantly
    -0.66
    Near
    -0.65
    alysed
    -0.65
    xon
    -0.63
    icia
    -0.63
    ocated
    -0.63
    iHUD
    -0.62
    POSITIVE LOGITS
     why
    0.97
     nobody
    0.94
     imagine
    0.87
     suppose
    0.86
     wouldn
    0.83
    nob
    0.80
     hey
    0.78
     isn
    0.78
    why
    0.77
     if
    0.77
    Act Density 0.174%

    No Known Activations