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

    Neuronpedia

    INDEX
    Explanations

    trained on data

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     보내
    0.42
    我自己
    0.40
     خودم
    0.40
     சகோ
    0.38
     steadily
    0.38
     düşünüyorum
    0.37
     দিচ্ছেন
    0.37
     প্রয়োজন
    0.37
    exempt
    0.37
     ಬರುತ್ತ
    0.37
    POSITIVE LOGITS
     training
    1.37
    训练
    1.22
     trained
    1.20
     learnt
    1.19
    Training
    1.18
     learned
    1.17
     Training
    1.14
    training
    1.14
    訓練
    1.09
     TRAINING
    1.09
    Act Density 0.044%

    No Known Activations