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

    Neuronpedia

    INDEX
    Explanations

    in Development or level

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     years
    0.86
     beginning
    0.86
     further
    0.86
     races
    0.84
    گزاری
    0.84
     abandoning
    0.84
    一下子
    0.83
     try
    0.83
     worrying
    0.82
     never
    0.80
    POSITIVE LOGITS
    <unused2222>
    1.15
    <unused2140>
    1.14
    <unused1990>
    1.07
    <0xBC>
    1.07
    <unused2113>
    1.05
    <unused1991>
    1.03
    1.02
    난
    1.00
     INumber
    1.00
     বাহিরে
    1.00
    Act Density 0.490%

    No Known Activations