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

    Neuronpedia

    INDEX
    Explanations

    proper nouns representing names of individuals

    oai_token-act-pair · gpt-3.5-turbo

    references to specific individuals, denoted by the word "His" or "Her."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ����
    -0.80
    ̶
    -0.78
    —-
    -0.77
    .–
    -0.74
    oso
    -0.71
    ÙIJ
    -0.70
    ij士
    -0.70
    DN
    -0.70
    xxx
    -0.68
    âľ
    -0.68
    POSITIVE LOGITS
     detractors
    1.06
     biggest
    1.04
     goal
    1.03
     successor
    1.01
    itage
    1.00
     inability
    0.99
     penchant
    0.99
     willingness
    0.98
     predecessors
    0.97
     newfound
    0.97
    Act Density 0.158%

    No Known Activations