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

    Neuronpedia

    INDEX
    Explanations

    personal pronouns (I, she, he, you, they) followed by verbs

    oai_token-act-pair · gpt-3.5-turbo

    references to personal experiences and relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    uria
    -0.79
     Fuller
    -0.64
     Infinite
    -0.63
    artifacts
    -0.62
     Airbus
    -0.61
    umption
    -0.61
    ãĤº
    -0.61
     Lifetime
    -0.60
     Dawson
    -0.59
     Moreno
    -0.59
    POSITIVE LOGITS
     befriend
    1.15
     met
    1.08
     interacted
    1.05
     interviewed
    0.97
     admire
    0.96
     acquaintance
    0.96
    RL
    0.95
     trusts
    0.94
     trust
    0.93
     trusted
    0.91
    Act Density 0.125%

    No Known Activations