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

    Neuronpedia

    INDEX
    Explanations

    tell me

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     данного
    1.76
     अर्ज
    1.73
     ð
    1.67
     该
    1.65
    ilege
    1.54
    RCP
    1.52
    ции
    1.50
    oea
    1.49
    ்ச
    1.49
    /***
    1.47
    POSITIVE LOGITS
     congregate
    2.22
     goodbye
    2.02
    ி
    1.95
    实话
    1.92
     বাহুল্য
    1.91
     topLeft
    1.88
     tale
    1.72
    tale
    1.69
     rife
    1.69
     särsk
    1.66
    Act Density 0.115%

    No Known Activations