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

    Neuronpedia

    INDEX
    Explanations

    It's contractions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Ketika
    0.34
     Ketika
    0.31
    Jika
    0.31
    自分が
    0.31
    Summary
    0.30
    Other
    0.30
    Sementara
    0.30
    Parent
    0.29
     شوند
    0.29
     които
    0.29
    POSITIVE LOGITS
     hasn
    0.57
     wasn
    0.56
     could
    0.53
     isn
    0.51
     doesn
    0.50
     shouldn
    0.50
     has
    0.50
     thrives
    0.49
     exists
    0.48
     COULD
    0.48
    Act Density 0.410%

    No Known Activations