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

    Neuronpedia

    INDEX
    Explanations

    game-related terms such as difficulty, health, damage, spells, and abilities

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ovie
    -0.91
    ogene
    -0.90
     Nicarag
    -0.76
    çīĪ
    -0.73
    orld
    -0.73
     Euras
    -0.69
     guiActiveUn
    -0.69
     covari
    -0.69
     Bland
    -0.69
     McCull
    -0.67
    POSITIVE LOGITS
     regeneration
    1.18
     regenerate
    0.97
     Regener
    0.97
     potions
    0.95
     orbs
    0.92
     Ammo
    0.88
     boost
    0.86
     expenditure
    0.85
    iencies
    0.83
     drain
    0.82
    Act Density 10.127%

    No Known Activations