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

    Neuronpedia

    INDEX
    Explanations

    number ranges expressed in the format "start - end"

    oai_token-act-pair · gpt-3.5-turbo

    patterns related to year-long durations or comparisons

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    anium
    -0.71
    esson
    -0.71
     Bullets
    -0.70
    SourceFile
    -0.70
     Shooter
    -0.67
    eele
    -0.65
    keley
    -0.65
     Mechdragon
    -0.65
    ãĥĦ
    -0.64
    idas
    -0.64
    POSITIVE LOGITS
    based
    1.10
    sized
    1.03
    time
    0.97
    years
    0.97
    level
    0.95
    long
    0.90
    appointed
    0.89
    day
    0.88
    to
    0.88
    release
    0.88
    Act Density 0.099%

    No Known Activations