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

    Neuronpedia

    INDEX
    Explanations

    instances of possession or control over characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    possessed, possessing, spirits, demonic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     صوتيه
    -0.58
    Aiheesta
    -0.54
    ModelSerializer
    -0.51
    MessageOf
    -0.50
     plegable
    -0.50
     فريبيس
    -0.49
     terecht
    -0.49
     Wikiseite
    -0.48
     ExecuteAsync
    -0.47
    -------------</
    -0.47
    POSITIVE LOGITS
     exor
    0.74
     Exor
    0.64
     demon
    0.46
     voodoo
    0.41
     curar
    0.39
     cure
    0.39
     witchcraft
    0.39
    Lu
    0.39
     Demon
    0.38
     Hippo
    0.38
    Act Density 0.160%

    No Known Activations