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

    Neuronpedia

    INDEX
    Explanations

    keywords related to violence and criminal acts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    SourceChecksum
    -0.48
     dienst
    -0.42
     vermogen
    -0.40
     unicórnio
    -0.40
     Tienes
    -0.39
     gekomen
    -0.38
     preciosa
    -0.38
     horaires
    -0.37
     gră
    -0.37
    ConverterFactory
    -0.37
    POSITIVE LOGITS
     attack
    0.68
     attacks
    0.65
     vandal
    0.59
    attacks
    0.59
    attack
    0.58
     <<<<<<<<<<<<<<
    0.58
     attackers
    0.58
     Attacks
    0.57
     sabotage
    0.56
     attacked
    0.56
    Act Density 0.480%

    No Known Activations