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

    Neuronpedia

    INDEX
    Explanations

    information related to criminal activities and police reports

    oai_token-act-pair · gpt-3.5-turbo

    references to criminal activity and suspects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     UNESCO
    -0.70
    isSpecialOrderable
    -0.69
     correctness
    -0.68
    inction
    -0.66
    Score
    -0.66
     obscurity
    -0.65
     relegation
    -0.64
    Ħ¢
    -0.64
     winners
    -0.64
    alg
    -0.63
    POSITIVE LOGITS
     assaulted
    1.26
     stole
    1.20
     fled
    1.20
     drove
    1.17
     confessed
    1.16
     robbed
    1.11
     entered
    1.10
     walked
    1.10
     exited
    1.10
     allegedly
    1.10
    Act Density 0.234%

    No Known Activations