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

    Neuronpedia

    INDEX
    Explanations

    phrases related to explosive devices or bomb-related activities

    oai_token-act-pair · gpt-3.5-turbo

    references to bombs or explosive devices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    laus
    -0.90
    SEA
    -0.80
     Vale
    -0.71
     Legends
    -0.70
     Dise
    -0.68
    Past
    -0.66
    éĸ
    -0.66
    Sea
    -0.64
    nee
    -0.64
     Absent
    -0.64
    POSITIVE LOGITS
    arding
    1.23
    arded
    1.16
     detonated
    1.02
     deton
    0.93
     bomb
    0.89
    astic
    0.86
    proof
    0.85
     disposal
    0.85
    astically
    0.85
     explodes
    0.85
    Act Density 0.018%

    No Known Activations