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

    Neuronpedia

    INDEX
    Explanations

    personal pronouns followed by high-context personal or subjective statements

    oai_token-act-pair · gpt-3.5-turbo

    instances of self-referential statements or personal pronouns

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Gale
    -0.71
     Nora
    -0.69
     Net
    -0.65
     Uriel
    -0.62
     Aberdeen
    -0.61
     Cost
    -0.60
     Violet
    -0.60
     Mats
    -0.60
     Jarrett
    -0.59
     Impact
    -0.59
    POSITIVE LOGITS
    'm
    1.30
    've
    1.27
    'd
    1.08
    'll
    1.06
    RL
    1.01
     suppose
    0.97
     presume
    0.91
     adore
    0.90
    OTA
    0.90
    ANA
    0.87
    Act Density 0.140%

    No Known Activations