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

    Neuronpedia

    INDEX
    Explanations

    first-person and third-person pronouns in reflective contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Millennium
    -0.54
     Mandatory
    -0.53
     Episode
    -0.51
     Nationwide
    -0.50
     Unlimited
    -0.48
     Round
    -0.47
     Reading
    -0.46
     Measure
    -0.46
    assembly
    -0.46
     Around
    -0.45
    POSITIVE LOGITS
    'll
    1.08
    'd
    1.06
    've
    0.98
     didn
    0.89
     forgot
    0.87
    're
    0.86
     swore
    0.85
     hadn
    0.84
     didnt
    0.83
     knew
    0.83
    Act Density 0.405%

    No Known Activations