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

    Neuronpedia

    INDEX
    Explanations

    the word "mean" used in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    statements that introduce exceptions or clarifications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    utenberg
    -0.78
    isin
    -0.78
    antha
    -0.73
    tex
    -0.71
    pes
    -0.71
     Peb
    -0.68
    bj
    -0.68
    figured
    -0.66
    bah
    -0.66
    figure
    -0.64
    POSITIVE LOGITS
     anymore
    0.98
     anything
    0.94
     nor
    0.86
     anyone
    0.86
    erest
    0.83
     anybody
    0.81
     necessarily
    0.81
     any
    0.77
    eworthy
    0.71
     goodbye
    0.70
    Act Density 0.060%

    No Known Activations