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

    Neuronpedia

    INDEX
    Explanations

    made of or from

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0.91
    ﻮ
    0.84
    ﻤ
    0.74
    ﻪ
    0.71
    ון
    0.70
    ﻜ
    0.70
     alveolar
    0.68
     данные
    0.65
     zatem
    0.64
     suro
    0.63
    POSITIVE LOGITS
    ت
    1.15
    ed
    1.02
    س
    0.95
    ts
    0.84
    सँग
    0.82
    تهم
    0.82
    k
    0.82
    с
    0.79
    et
    0.78
    da
    0.77
    Act Density 0.003%

    No Known Activations