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

    Neuronpedia

    INDEX
    Explanations

    quotations in text

    oai_token-act-pair · gpt-3.5-turbo

    quotation marks and phrases within them

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Fas
    -0.79
     cite
    -0.78
     Versus
    -0.77
     XL
    -0.76
     cf
    -0.76
     ABE
    -0.75
     tackle
    -0.75
     deductions
    -0.74
     BD
    -0.73
     chant
    -0.72
    POSITIVE LOGITS
    absolutely
    1.88
    extremely
    1.87
    completely
    1.81
    very
    1.75
    pretty
    1.72
    highly
    1.72
    quite
    1.71
    too
    1.67
    really
    1.66
    probably
    1.65
    Act Density 0.083%

    No Known Activations