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

    Neuronpedia

    INDEX
    Explanations

    references to threats against individuals, families, and national security

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

    threatening someone with blackmail

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    windowFixed
    -0.57
    principalColumn
    -0.56
    exitRule
    -0.43
     ligiloj
    -0.42
    ToScroll
    -0.41
     BoxDecoration
    -0.40
    AsUp
    -0.40
     cherchés
    -0.39
     iprot
    -0.39
    iastes
    -0.38
    POSITIVE LOGITS
    threat
    1.32
     threat
    1.32
     threats
    1.30
     threatened
    1.27
    Threat
    1.26
     threaten
    1.24
     threatening
    1.22
     threatens
    1.19
     Threat
    1.13
    threatening
    1.13
    Act Density 0.034%

    No Known Activations