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

    Neuronpedia

    INDEX
    Explanations

    reinforcement

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     affairs
    -0.09
     plonge
    -0.08
     জানা
    -0.08
    кан
    -0.08
     בנ
    -0.07
     nio
    -0.07
     جات
    -0.07
    -src
    -0.07
     berlin
    -0.07
    -defense
    -0.07
    POSITIVE LOGITS
     rewarded
    0.14
     reward
    0.14
    奖励
    0.13
    Reward
    0.13
    .reward
    0.12
     rewarding
    0.12
     incentiv
    0.12
     rewards
    0.12
    Rewards
    0.12
     Reward
    0.11
    Act Density 0.013%

    No Known Activations