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

    Neuronpedia

    INDEX
    Explanations

    reports of violent incidents

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    itech
    -0.79
    apore
    -0.78
    enture
    -0.78
     ECO
    -0.76
     dividend
    -0.75
    ãĥij
    -0.74
    iggurat
    -0.74
     constraint
    -0.73
     tariff
    -0.73
     util
    -0.73
    POSITIVE LOGITS
     coroner
    1.50
     autopsy
    1.46
     detectives
    1.44
     eyewitness
    1.40
     homicide
    1.36
     gunman
    1.33
    Investigators
    1.31
     SWAT
    1.28
     killers
    1.27
     LAPD
    1.27
    Act Density 0.881%

    No Known Activations