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

    Neuronpedia

    INDEX
    Explanations

    multi-lingual syllable beginnings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ijo
    0.44
    ത്വം
    0.44
    ום
    0.42
     Chapman
    0.41
     Dyke
    0.41
    Chap
    0.40
    loom
    0.39
    andosi
    0.39
    جل
    0.38
     Ultraviolet
    0.38
    POSITIVE LOGITS
    ᇂ
    0.48
    שא
    0.46
    低的
    0.41
    係る
    0.40
    ภัณฑ์
    0.40
    ેલી
    0.40
    𝗁
    0.39
    加えて
    0.38
     скло
    0.38
     practiced
    0.38
    Act Density 0.000%

    No Known Activations