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

    Neuronpedia

    INDEX
    Explanations

    nonsense words and symbols

    oai_token-act-pair · gpt-3.5-turbo

    specific non-standard characters or symbols

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    arton
    -0.77
     flour
    -0.77
    leground
    -0.73
     bonded
    -0.73
    wagen
    -0.72
     betting
    -0.70
     presidency
    -0.69
    olithic
    -0.68
    erate
    -0.68
     agre
    -0.68
    POSITIVE LOGITS
    ãĤ
    1.77
    ãģ
    1.76
    ãģª
    1.67
    ãģĦ
    1.56
    ãģ¾
    1.55
    ãĥı
    1.53
    ãĢģ
    1.52
    ãĥ¼ãĥ
    1.52
    ãĥĩ
    1.52
    ãĤĭ
    1.51
    Act Density 0.015%

    No Known Activations