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

    Neuronpedia

    INDEX
    Explanations

    who followed by verb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     in
    0.67
    是
    0.66
    は
    0.64
     في
    0.61
     e
    0.58
     is
    0.57
    在
    0.57
     antara
    0.56
     و
    0.55
     عن
    0.55
    POSITIVE LOGITS
    y
    0.47
     Variation
    0.45
     That
    0.44
    t
    0.43
    '.
    0.42
     themselves
    0.39
     conspiring
    0.39
     Who
    0.39
    tım
    0.39
    यों
    0.38
    Act Density 0.266%

    No Known Activations