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

    Neuronpedia

    INDEX
    Explanations

    informaion about airplane crashes

    oai_token-act-pair · gpt-3.5-turbo

    mentions of plane crashes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    icrobial
    -0.72
    Intern
    -0.71
     reconc
    -0.67
     Cosmos
    -0.67
    ilyn
    -0.66
    Pr
    -0.64
    Balt
    -0.63
     hats
    -0.63
     ILCS
    -0.63
    atem
    -0.60
    POSITIVE LOGITS
    worthiness
    1.06
     involving
    0.81
     hazard
    0.75
     Victims
    0.73
     victims
    0.71
    prone
    0.70
     inflicted
    0.69
     perpetrated
    0.69
     crashes
    0.68
     caused
    0.68
    Act Density 0.055%

    No Known Activations