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

    Neuronpedia

    INDEX
    Explanations

    Chinese, Spanish, and English prefixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    MUX
    0.38
    regarding
    0.37
     lâm
    0.37
     দমন
    0.36
    有關
    0.36
    ರಿಗೆ
    0.35
     concerning
    0.35
    ureshi
    0.34
    俘
    0.34
     Exempt
    0.34
    POSITIVE LOGITS
    一下
    0.50
    Vice
    0.44
    夗
    0.43
    वस्था
    0.39
     Stable
    0.38
    看一下
    0.38
    Under
    0.38
    vice
    0.38
    跑步
    0.38
    Chair
    0.37
    Act Density 0.004%

    No Known Activations