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

    Neuronpedia

    INDEX
    Explanations

    digits and code structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     шеран
    0.42
     rhenium
    0.40
    हन
    0.39
    тона
    0.39
     Quỳnh
    0.38
    ቱን
    0.38
    쇤
    0.38
     всему
    0.37
    ध्यात्म
    0.37
    ваясь
    0.37
    POSITIVE LOGITS
     ১
    0.65
     १
    0.60
    1
    0.54
    １
    0.54
     ۱
    0.50
    (-
    0.50
    ১
    0.49
     one
    0.46
    minus
    0.46
     minus
    0.45
    Act Density 0.031%

    No Known Activations