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

    Neuronpedia

    INDEX
    Explanations

    phrases related to military and defense activities

    oai_token-act-pair · gpt-3.5-turbo

    references to defense-related topics and officials

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hell
    -0.78
    olics
    -0.70
    bern
    -0.68
    pel
    -0.66
    hest
    -0.65
    otin
    -0.65
    à
    -0.65
    upper
    -0.64
    heet
    -0.63
    colo
    -0.62
    POSITIVE LOGITS
     contractor
    1.02
     contractors
    0.89
     Secretary
    0.84
     Department
    0.83
     Ministry
    0.82
     ministry
    0.82
     Memor
    0.81
     Dept
    0.80
     Defense
    0.79
     Forces
    0.79
    Act Density 0.022%

    No Known Activations