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

    Neuronpedia

    INDEX
    Explanations

    assistant followed by comma

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    posta
    0.39
    mono
    0.39
    mill
    0.39
     ingen
    0.39
    fold
    0.39
     pož
    0.38
     smak
    0.38
     gilt
    0.38
     mill
    0.38
     jem
    0.37
    POSITIVE LOGITS
     saj
    0.37
    Ker
    0.37
     ഏറെ
    0.37
    Enumerator
    0.36
     vendar
    0.35
     असल्याने
    0.35
    摡
    0.35
     Jerusalem
    0.34
     మార్
    0.34
    urgeon
    0.34
    Act Density 0.001%

    No Known Activations