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

    Neuronpedia

    INDEX
    Explanations

    references to character dynamics, particularly between heroes and villains

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    onis
    -0.15
     hero
    -0.15
    umont
    -0.15
     çij
    -0.15
    tae
    -0.14
     Sherlock
    -0.14
    Heroes
    -0.14
    _hero
    -0.14
    unread
    -0.14
    704
    -0.14
    POSITIVE LOGITS
     evil
    0.30
     villain
    0.28
     vill
    0.26
     Evil
    0.26
     Vill
    0.24
     villains
    0.23
    evil
    0.23
     superv
    0.21
     evils
    0.21
    boss
    0.20
    Act Density 0.215%

    No Known Activations