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

    Neuronpedia

    INDEX
    Explanations

    personal pronouns and reflexive pronouns

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of self or personal identity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Amend
    -0.67
     Sierra
    -0.62
     airspace
    -0.61
     Clover
    -0.61
     racket
    -0.59
     Vil
    -0.59
    akings
    -0.59
     Noon
    -0.58
     Scrib
    -0.58
    allas
    -0.58
    POSITIVE LOGITS
    selves
    1.20
    destruct
    0.92
    ortium
    0.91
    pecially
    0.88
    ridges
    0.87
    self
    0.86
    theless
    0.78
    terday
    0.77
     explanatory
    0.76
    contained
    0.76
    Act Density 0.019%

    No Known Activations