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

    Neuronpedia

    INDEX
    Explanations

    has/have + past participle

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ולה
    0.55
    Z
    0.46
    if
    0.45
    I
    0.45
    which
    0.45
    Contains
    0.44
    и
    0.43
    Last
    0.43
    B
    0.43
    E
    0.43
    POSITIVE LOGITS
     been
    1.98
    been
    1.40
     BEEN
    1.39
     Been
    1.34
     gotten
    1.19
     været
    1.17
    Been
    1.16
     struggled
    1.13
     undergone
    1.12
     varit
    1.10
    Act Density 0.314%

    No Known Activations