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

    Neuronpedia

    INDEX
    Explanations

    phrases related to quoting someone's speech

    oai_token-act-pair · gpt-3.5-turbo

    direct quotes or reported speech in conversations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Measure
    -0.60
    senal
    -0.59
     silhou
    -0.59
     lovers
    -0.57
     traject
    -0.57
    Model
    -0.56
     vanity
    -0.55
     genital
    -0.55
     manifesto
    -0.54
    ãĥ¼ãĥĨãĤ£
    -0.54
    POSITIVE LOGITS
     told
    0.81
    _.
    0.71
    said
    0.69
    ese
    0.68
     adding
    0.68
    arkin
    0.68
     said
    0.68
     cited
    0.67
     quoted
    0.67
    >]
    0.67
    Act Density 0.178%

    No Known Activations