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

    Neuronpedia

    INDEX
    Explanations

    tell application; limiting steps

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    鲷
    0.45
    isting
    0.42
    ၠ
    0.42
    నకు
    0.41
     өл
    0.41
    .".,
    0.41
     tocó
    0.40
     مشتمل
    0.40
    近代
    0.40
     Eren
    0.39
    POSITIVE LOGITS
     influence
    0.42
     present
    0.42
     slowest
    0.42
     up
    0.42
    tax
    0.42
     slow
    0.41
     attachment
    0.39
     crucial
    0.39
    off
    0.38
     exhaust
    0.38
    Act Density 0.000%

    No Known Activations