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

    Neuronpedia

    INDEX
    Explanations

    personal pronouns and verbs related to action

    oai_token-act-pair · gpt-3.5-turbo

    pronouns indicating personal and collective identity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Gad
    -0.56
    earch
    -0.56
     Millennium
    -0.56
     Rockefeller
    -0.55
    ¿½
    -0.54
    isher
    -0.54
     Concord
    -0.53
     Amen
    -0.53
     Twelve
    -0.52
     Harm
    -0.51
    POSITIVE LOGITS
    've
    1.32
    'll
    1.28
    're
    1.24
    'd
    1.11
     wanna
    1.06
    'm
    1.04
     haven
    0.98
     gotta
    0.93
     dunno
    0.92
     don
    0.92
    Act Density 0.563%

    No Known Activations