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

    Neuronpedia

    INDEX
    Explanations

    short-term

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    byter
    0.83
    Padding
    0.77
     এছ
    0.75
    Beyond
    0.74
     ಮೊದಲ
    0.72
    Grpc
    0.72
    ParallelGroup
    0.71
     egalitarian
    0.71
     allere
    0.71
    ższe
    0.70
    POSITIVE LOGITS
    هلا
    0.58
    any
    0.56
    om
    0.56
     információ
    0.54
     unable
    0.53
     thôi
    0.53
     flanked
    0.52
    حيان
    0.52
     trabal
    0.52
     insufficient
    0.51
    Act Density 0.001%

    No Known Activations