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

    Neuronpedia

    INDEX
    Explanations

    phrases related to economic status and statistics

    oai_token-act-pair · gpt-3.5-turbo

    references to emotional states and social conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    tains
    -0.86
     involves
    -0.86
     Contains
    -0.86
     Adds
    -0.85
     is
    -0.82
     meets
    -0.81
     leads
    -0.80
     serves
    -0.79
    needs
    -0.79
     expires
    -0.79
    POSITIVE LOGITS
     were
    1.66
     weren
    1.56
    were
    1.47
     tended
    1.28
     wore
    1.18
    Were
    1.15
     lacked
    1.14
     behaved
    1.12
     Were
    1.10
     undertook
    1.09
    Act Density 0.774%

    No Known Activations