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

    Neuronpedia

    INDEX
    Explanations

    instances of parentheses with various content inside

    oai_token-act-pair · gpt-3.5-turbo

    the use of parentheses in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     comparisons
    -0.72
     joints
    -0.69
     abroad
    -0.68
     maker
    -0.67
     galleries
    -0.66
     firewall
    -0.66
     favour
    -0.66
     snapping
    -0.65
     bows
    -0.65
     undertaking
    -0.65
    POSITIVE LOGITS
    mostly
    1.74
    possibly
    1.63
    usually
    1.57
    almost
    1.55
    sic
    1.55
    albeit
    1.49
    literally
    1.48
    sometimes
    1.47
    formerly
    1.43
    often
    1.43
    Act Density 0.070%

    No Known Activations