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

    Neuronpedia

    INDEX
    Explanations

    sentences containing the word "So" followed by a pronoun and some action or decision

    oai_token-act-pair · gpt-3.5-turbo

    instances of personal pronouns and collective references

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    aining
    -0.68
    lif
    -0.62
     Notice
    -0.62
    iscovery
    -0.61
    YS
    -0.61
     sensing
    -0.59
    ilty
    -0.59
     witnessing
    -0.59
    Relations
    -0.59
    Vict
    -0.59
    POSITIVE LOGITS
     resorted
    1.32
     opted
    1.25
     decided
    1.24
     devised
    1.19
     enlisted
    1.09
     teamed
    1.02
     undertook
    1.02
     instituted
    1.01
     embarked
    1.01
     ventured
    0.97
    Act Density 0.382%

    No Known Activations