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

    Neuronpedia

    INDEX
    Explanations

    speaking and understanding requests

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    有意
    0.40
    喧
    0.39
     לו
    0.39
     希望
    0.39
     பரு
    0.38
    localize
    0.38
     সংয
    0.37
    Dynamic
    0.36
    }$-
    0.36
    ဂျ
    0.36
    POSITIVE LOGITS
     here
    0.44
     aquí
    0.42
    Pa
    0.40
    Gob
    0.39
     proving
    0.37
     دینا
    0.37
    here
    0.36
     snapping
    0.36
     remplir
    0.36
    pa
    0.35
    Act Density 0.001%

    No Known Activations