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

    Neuronpedia

    INDEX
    Explanations

    phrases related to the lack of information or visibility

    oai_token-act-pair · gpt-3.5-turbo

    verbs and phrases related to documentation and publication

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ugi
    -0.63
    rather
    -0.59
    ggles
    -0.58
    itton
    -0.56
     assorted
    -0.55
     LOT
    -0.53
     coolest
    -0.52
    yeah
    -0.51
     goodness
    -0.51
     humble
    -0.51
    POSITIVE LOGITS
     anymore
    1.75
     nor
    1.66
    yet
    1.15
     yet
    1.09
     anywhere
    1.08
    nor
    1.04
     until
    1.01
     unless
    1.01
     whatsoever
    1.01
     necessarily
    0.98
    Act Density 0.169%

    No Known Activations