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

    Neuronpedia

    INDEX
    Explanations

    relative clauses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     konuda
    0.41
     veulent
    0.39
     రెడ్డి
    0.39
     kēia
    0.39
     Pacquiao
    0.37
     voorbeeld
    0.37
     theory
    0.36
     dieser
    0.36
     Bunu
    0.36
     żeby
    0.35
    POSITIVE LOGITS
    ซึ่ง
    1.04
     which
    1.02
    which
    0.94
     ซึ่ง
    0.80
     WHICH
    0.73
     whiche
    0.71
     אשר
    0.70
    Which
    0.66
     который
    0.66
     რომელიც
    0.64
    Act Density 0.087%

    No Known Activations