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

    Neuronpedia

    INDEX
    Explanations

    the word "very" used to intensify various descriptors or facts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    orrh
    -0.16
    ãĤ´ãĥª
    -0.16
    stroy
    -0.16
     phong
    -0.15
    pa
    -0.15
    iet
    -0.15
    oris
    -0.15
    oS
    -0.14
    ascript
    -0.14
    strand
    -0.14
    POSITIVE LOGITS
     same
    0.19
    ewan
    0.17
    NS
    0.16
    790
    0.15
     opposite
    0.15
    same
    0.15
     first
    0.15
    411
    0.14
     reason
    0.14
     last
    0.14
    Act Density 0.023%

    No Known Activations