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

    Neuronpedia

    INDEX
    Explanations

    patterns related to sports scores and statistics involving numbers with '-'

    oai_token-act-pair · gpt-3.5-turbo

    negative numerical values or statistics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ulhu
    -0.93
     photoc
    -0.81
    kson
    -0.79
     ®
    -0.77
    ibur
    -0.76
     moder
    -0.75
    atel
    -0.73
     Parenthood
    -0.71
    esson
    -0.67
     Paradox
    -0.66
    POSITIVE LOGITS
    season
    1.09
    game
    1.05
    team
    1.05
    player
    0.99
    scoring
    0.96
    winning
    0.95
    Miami
    0.93
    legged
    0.91
    sports
    0.91
    goal
    0.89
    Act Density 0.137%

    No Known Activations