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

    Neuronpedia

    INDEX
    Explanations

    if/then structures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     enablement
    0.52
    یند
    0.44
     ہیں۔
    0.43
    ്രീ
    0.42
     سائٹ
    0.40
    িলেন
    0.39
     imagery
    0.39
    都不
    0.39
     يحدث
    0.39
    ক্ষিতে
    0.39
    POSITIVE LOGITS
     takich
    0.45
     Shap
    0.45
     Johnny
    0.44
     P
    0.44
     Danny
    0.44
     Cyril
    0.43
     самим
    0.42
     Flash
    0.42
     Shar
    0.42
     Ti
    0.42
    Act Density 0.001%

    No Known Activations