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

    Neuronpedia

    INDEX
    Explanations

    code parsing for prediction

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    tti
    1.12
    ا
    1.11
    을
    1.07
    ть
    1.07
     대해서
    1.04
    წი
    1.02
    ా
    1.00
     préparation
    1.00
    ้ำ
    0.98
    اة
    0.97
    POSITIVE LOGITS
    ist
    0.98
     derogatory
    0.95
     leftist
    0.95
     hopelessness
    0.92
     deadliest
    0.90
    𝑖
    0.90
     Saudi
    0.89
     whereabouts
    0.89
     popular
    0.87
     discriminatory
    0.87
    Act Density 0.004%

    No Known Activations