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

    Neuronpedia

    INDEX
    Explanations

    references to specific entities or concepts starting with "Im" or "im"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Im" or its variations, indicating a focus on self-reference or personal statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    NetMessage
    -1.02
    ttes
    -0.77
     grapes
    -0.70
     wool
    -0.70
     stakes
    -0.69
     shroud
    -0.68
     braces
    -0.68
     Nazis
    -0.67
     Skydragon
    -0.67
     Chrys
    -0.66
    POSITIVE LOGITS
    mediate
    1.34
    mediately
    1.31
    perial
    1.29
    ply
    1.25
    perfect
    1.22
    ple
    1.22
    unity
    1.22
    posing
    1.19
    minent
    1.16
    medi
    1.15
    Act Density 0.027%

    No Known Activations