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

    Neuronpedia

    INDEX
    Explanations

    phrases related to criminal activities and investigations

    oai_token-act-pair · gpt-3.5-turbo

    references to planned or intentional criminal activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Tokens
    -0.88
    notation
    -0.73
     Cosponsors
    -0.68
    helps
    -0.67
    urn
    -0.67
    imei
    -0.67
    then
    -0.66
    oise
    -0.64
    ometers
    -0.63
    anguages
    -0.63
    POSITIVE LOGITS
     spontaneous
    1.18
     accidental
    1.11
     retaliation
    1.08
     perpetrated
    1.07
     arson
    1.07
     revenge
    1.03
     suicides
    1.00
     robbery
    0.99
     suicide
    0.98
     deliberate
    0.96
    Act Density 0.270%

    No Known Activations