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

    Neuronpedia

    INDEX
    Explanations

    words related to improvement, progress, or motivation

    oai_token-act-pair · gpt-3.5-turbo

    special characters, particularly the character "Â" which appears prominently in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    enburg
    -0.78
    enegger
    -0.77
    ãģ®éŃĶ
    -0.68
     Athena
    -0.65
    iary
    -0.65
    iant
    -0.62
     Atom
    -0.62
    iard
    -0.61
    otrop
    -0.60
     Druid
    -0.60
    POSITIVE LOGITS
    Â
    1.38
    ¹
    1.30
    ¬
    1.15
    ¼
    1.15
    ¥
    1.14
    ¤
    1.11
    ª
    1.09
    ¿
    1.09
    ¸
    1.04
    ³
    1.04
    Act Density 0.010%

    No Known Activations