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

    Neuronpedia

    INDEX
    Explanations

    reveal leading to change

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    対策
    0.70
     Improvement
    0.65
    держи
    0.65
     Improving
    0.62
     stammt
    0.61
    奋斗
    0.60
     குறைவாக
    0.60
    Improving
    0.59
    improvement
    0.59
     पावसा
    0.58
    POSITIVE LOGITS
     catapult
    1.11
     unlock
    1.04
     reshape
    1.02
     shatter
    1.01
     unlocking
    1.00
     fundamentally
    0.99
     resh
    0.98
     awaken
    0.98
     unlocked
    0.97
     unlocks
    0.97
    Act Density 0.961%

    No Known Activations