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

    Neuronpedia

    INDEX
    Explanations

    names of specific locations such as streets and buildings, particularly those related to "Castle"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the word "Castle" or "castle" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    atives
    -0.82
    pace
    -0.79
    ostics
    -0.78
    earcher
    -0.76
    imester
    -0.73
    nesota
    -0.70
    Ö¼
    -0.70
    paces
    -0.68
    iating
    -0.67
    iferation
    -0.67
    POSITIVE LOGITS
     Templar
    0.89
    llan
    0.87
     castle
    0.83
     Castle
    0.81
     Oblivion
    0.74
    holder
    0.73
     castles
    0.72
     Siege
    0.72
    fish
    0.72
     Mansion
    0.71
    Act Density 0.021%

    No Known Activations