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

    Neuronpedia

    INDEX
    Explanations

    instances of repeated text occurring in several contexts, particularly those involving discussions of accountability in law enforcement

    oai_token-act-pair · gpt-3.5-turbo

    connecting conjunctions, particularly "and," within various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    marine
    -0.72
    inch
    -0.72
    anza
    -0.68
     Franch
    -0.66
    activity
    -0.66
    rir
    -0.65
    steen
    -0.63
     Doodle
    -0.62
    hov
    -0.62
    leon
    -0.62
    POSITIVE LOGITS
     reused
    1.07
     executed
    1.05
     photographed
    1.04
     interrogated
    1.04
     prosecuted
    1.00
     analyzed
    0.99
     analysed
    0.97
     withdrawn
    0.96
     transported
    0.96
     evaluated
    0.95
    Act Density 0.333%

    No Known Activations