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

    Neuronpedia

    INDEX
    Explanations

    repeated patterns of characters like "ib" or "ph"

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    bilt
    -0.91
    ISTER
    -0.80
     Ceres
    -0.79
    isters
    -0.78
     Russ
    -0.77
    ister
    -0.77
     Blizzard
    -0.77
     Bender
    -0.74
     marsh
    -0.74
     Prosper
    -0.74
    POSITIVE LOGITS
    odies
    1.37
    ilib
    1.37
    raltar
    1.34
    rahim
    1.30
    ulous
    1.30
    ibl
    1.26
    bole
    1.22
    acter
    1.21
    ody
    1.20
    ration
    1.18
    Act Density 9.763%

    No Known Activations