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

    Neuronpedia

    INDEX
    Explanations

    eat, eating, ate, eaten

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    보
    0.64
    ৫
    0.64
    V
    0.61
    ু
    0.61
     لاک
    0.60
    م
    0.60
    ک
    0.60
    Ч
    0.59
    Nicol
    0.57
    З
    0.56
    POSITIVE LOGITS
     eat
    0.88
     eating
    0.88
     먹
    0.81
     Eat
    0.80
     Eating
    0.79
     eats
    0.78
     makan
    0.75
     Eats
    0.74
    吃了
    0.73
     eaten
    0.72
    Act Density 0.046%

    No Known Activations