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

    Neuronpedia

    INDEX
    Explanations

    Input: in instructions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     изготовления
    1.09
    Ар
    0.94
    wolves
    0.93
    Тре
    0.93
    цу
    0.93
    们
    0.93
     ireo
    0.90
     шар
    0.90
     sniff
    0.90
     underlining
    0.90
    POSITIVE LOGITS
    ting
    1.57
    ح
    1.24
    ಾನ್
    1.23
    ures
    1.21
    ted
    1.21
    sko
    1.18
     Producing
    1.16
    1.14
    isierte
    1.12
    rechte
    1.11
    Act Density 0.116%

    No Known Activations