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

    Neuronpedia

    INDEX
    Explanations

    pronouns followed by verbs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     tentu
    0.43
     oczywiście
    0.40
     assumption
    0.39
     ovviamente
    0.38
    明らかに
    0.37
     assuming
    0.36
     presumption
    0.35
     certainement
    0.33
     commencent
    0.33
     একদম
    0.33
    POSITIVE LOGITS
     বুঝি
    0.49
     ਹੈ
    0.38
     है
    0.37
     possesses
    0.35
    lias
    0.33
    আছে
    0.33
     ہے
    0.33
     নাকি
    0.33
     COULD
    0.33
    indeed
    0.32
    Act Density 0.050%

    No Known Activations