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

    Neuronpedia

    INDEX
    Explanations

    Cyrillic text with specific character combinations

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of a specific character or letter in a text, likely relevant to particular languages or scripts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    zzo
    -0.75
    atche
    -0.73
     Starr
    -0.72
    ttes
    -0.69
    Äĩ
    -0.69
     Pearce
    -0.68
    olkien
    -0.68
    otle
    -0.68
     Gutenberg
    -0.67
    ño
    -0.67
    POSITIVE LOGITS
    ÑĤ
    1.08
    к
    1.05
    dimension
    0.88
    и
    0.85
    Ñģ
    0.83
    Ĭ
    0.83
    оÐ
    0.82
    laden
    0.82
    л
    0.81
    ¿
    0.80
    Act Density 0.008%

    No Known Activations