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

    Neuronpedia

    INDEX
    Explanations

    chemical compounds and elements

    oai_token-act-pair · gpt-3.5-turbo

    parentheses or other grouping symbols

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     timely
    -0.77
     college
    -0.74
     waiver
    -0.73
     consolation
    -0.73
     reservation
    -0.72
     reserves
    -0.71
     colleges
    -0.71
     professors
    -0.71
     managerial
    -0.70
     interviews
    -0.70
    POSITIVE LOGITS
    shown
    1.67
    pictured
    1.62
    sometimes
    1.62
    also
    1.61
    pron
    1.57
    sic
    1.57
    which
    1.53
    often
    1.52
    literally
    1.52
    aka
    1.50
    Act Density 0.134%

    No Known Activations