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

    Neuronpedia

    INDEX
    Explanations

    housing, support, cables

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     порядка
    0.96
     торже
    0.96
     Ин
    0.94
     такую
    0.93
     должна
    0.93
    пы
    0.91
     الوضع
    0.91
     эту
    0.91
     была
    0.88
     слишком
    0.88
    POSITIVE LOGITS
    人员
    0.97
    ken
    0.95
     dietro
    0.89
    言語
    0.87
     frutto
    0.87
    難
    0.85
     otten
    0.84
    甶
    0.83
    ت
    0.82
     nữa
    0.82
    Act Density 0.001%

    No Known Activations