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

    Neuronpedia

    INDEX
    Explanations

    forms of the verb "to be" combined with a range of follow-up phrases or subjects

    oai_token-act-pair · gpt-3.5-turbo

    statements of personal identity or emotional states

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Gap
    -0.72
     entails
    -0.68
     occurs
    -0.67
     externalToEVAOnly
    -0.66
     lasts
    -0.66
     meets
    -0.65
     Colleges
    -0.64
     Utilities
    -0.63
     looms
    -0.63
    bies
    -0.63
    POSITIVE LOGITS
     glad
    1.24
     thankful
    1.19
     grateful
    1.12
     proud
    1.06
     fortunate
    1.04
     convinced
    1.03
     aware
    1.02
     hoping
    0.99
     honored
    0.99
     amazed
    0.99
    Act Density 0.119%

    No Known Activations