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

    Neuronpedia

    INDEX
    Explanations

    High activation for the Turkish letter "ı" with variations

    oai_token-act-pair · gpt-3.5-turbo

    specific character patterns or symbols that appear repeatedly

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Appalach
    -0.80
     guiActiveUnfocused
    -0.71
    arsity
    -0.71
     Spartan
    -0.69
     decomp
    -0.67
    maxwell
    -0.66
     Wilmington
    -0.65
    Chel
    -0.65
     Sussex
    -0.64
    IFIED
    -0.64
    POSITIVE LOGITS
    1.06
    ·
    1.02
    ÅŁ
    0.98
    ¾
    0.95
    ı
    0.93
    ĥ
    0.92
    Ì
    0.92
    oÄŁ
    0.89
    Ĩ
    0.89
    ¸
    0.88
    Act Density 0.010%

    No Known Activations