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

    Neuronpedia

    INDEX
    Explanations

    quotations within written texts

    oai_token-act-pair · gpt-3.5-turbo

    quotation marks and dialogue in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     developmental
    -0.76
     metic
    -0.75
     derby
    -0.71
     killer
    -0.71
     shell
    -0.70
     accomp
    -0.69
     sympath
    -0.68
     carrier
    -0.68
     cancell
    -0.67
     destro
    -0.66
    POSITIVE LOGITS
    there
    1.16
    we
    1.12
    trust
    1.09
    straight
    1.08
    Jews
    1.08
    I
    1.08
    double
    1.04
    every
    1.03
    because
    1.03
    nothing
    1.01
    Act Density 0.101%

    No Known Activations