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

    Neuronpedia

    INDEX
    Explanations

    right and left direction

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ļ
    0.79
     Risiko
    0.79
     يا
    0.75
    ια
    0.74
    佣
    0.74
    0.73
    اس
    0.72
     Global
    0.71
    ่ย
    0.71
     б
    0.70
    POSITIVE LOGITS
    eous
    1.87
    leftharpoons
    1.50
     sided
    1.42
    most
    1.36
    衽
    1.35
     ascension
    1.32
     handed
    1.30
    wing
    1.30
     wing
    1.30
     winger
    1.28
    Act Density 0.094%

    No Known Activations