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

    Neuronpedia

    INDEX
    Explanations

    words related to negative or tragic events

    oai_token-act-pair · gpt-3.5-turbo

    descriptive words indicating extreme negativity or suffering

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    pai
    -0.88
    ratulations
    -0.73
     Leilan
    -0.73
    Lean
    -0.72
    amaru
    -0.68
    sector
    -0.68
    yi
    -0.67
    yip
    -0.65
    ersen
    -0.65
    UP
    -0.64
    POSITIVE LOGITS
     ordeal
    1.04
    omic
    1.01
     atrocities
    0.97
     miscarriage
    0.96
     injust
    0.94
     injustice
    0.94
     nightmares
    0.91
     consequences
    0.90
    ally
    0.90
     tragedy
    0.89
    Act Density 0.096%

    No Known Activations