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

    Neuronpedia

    INDEX
    Explanations

    words associated with potential risks and hazards

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     unico
    -0.71
     Walpole
    -0.68
     Brücken
    -0.65
     apologize
    -0.65
     Giuli
    -0.64
    Bridges
    -0.64
    schirm
    -0.63
     Matsumoto
    -0.63
    ={`/
    -0.62
    endmodule
    -0.62
    POSITIVE LOGITS
     danger
    0.90
     Danger
    0.87
    danger
    0.86
    rungsseite
    0.86
     peligroso
    0.85
     dangereux
    0.84
     surla
    0.81
     perigo
    0.81
    ='".$
    0.81
     roger
    0.78
    Act Density 0.109%

    No Known Activations