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

    Neuronpedia

    INDEX
    Explanations

    code block delimiters and conjunctions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ্রান্ত
    0.39
    !).
    0.38
    רות
    0.38
     यदि
    0.38
    <unused730>
    0.38
     необходимых
    0.36
     умер
    0.36
    ridium
    0.36
    رير
    0.35
    전히
    0.35
    POSITIVE LOGITS
     ولكن
    0.45
     lakini
    0.44
    才可以
    0.43
    neze
    0.43
    可以
    0.42
    但不
    0.42
     ناج
    0.42
    才
    0.42
     but
    0.41
    ですが
    0.41
    Act Density 0.082%

    No Known Activations