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

    Neuronpedia

    INDEX
    Explanations

    specific names related to literature and fictional characters, particularly those related to horror and fantasy genres

    oai_token-act-pair · gpt-3.5-turbo

    references to classic horror or fantasy literature and characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    reon
    -0.92
    cript
    -0.82
    cling
    -0.77
    ral
    -0.75
    ples
    -0.73
    inence
    -0.73
    race
    -0.73
    ming
    -0.73
    istant
    -0.72
     stewards
    -0.72
    POSITIVE LOGITS
    enstein
    1.16
     Frankenstein
    1.09
     Masquerade
    1.00
     Madness
    0.96
     Dracula
    0.95
     Mansion
    0.89
     Werewolf
    0.87
     Horror
    0.83
    ãĥ«
    0.81
    ãĥ¼ãĤ¯
    0.80
    Act Density 0.023%

    No Known Activations