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

    Neuronpedia

    INDEX
    Explanations

    verb describing action followed by noun

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    elerinin
    0.83
     довольно
    0.82
     которые
    0.81
     경우에는
    0.80
     （
    0.80
    場合には
    0.78
     그러니까
    0.78
    比如说
    0.77
    的东西
    0.77
    這樣子
    0.76
    POSITIVE LOGITS
     same
    1.54
     own
    1.43
     largest
    1.38
     highest
    1.37
     latest
    1.34
     lowest
    1.28
     smallest
    1.23
     newest
    1.18
     biggest
    1.18
     multitude
    1.16
    Act Density 0.732%

    No Known Activations