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

    Neuronpedia

    INDEX
    Explanations

    emotional aspects or qualities in phrases

    oai_token-act-pair · gpt-3.5-turbo

    references to ambiguous or vague concepts or feelings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    DOS
    -0.80
    enter
    -0.73
    iper
    -0.73
    ortex
    -0.72
    anos
    -0.69
    ude
    -0.69
     Shutterstock
    -0.69
    de
    -0.68
    nor
    -0.68
    ipers
    -0.67
    POSITIVE LOGITS
     else
    1.28
     resembling
    1.07
     akin
    0.99
    Else
    0.96
     awfully
    0.92
     uncanny
    0.86
     intangible
    0.84
     miraculous
    0.84
     unforeseen
    0.84
     awful
    0.84
    Act Density 0.062%

    No Known Activations