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

    Neuronpedia

    INDEX
    Explanations

    type specification and conversion

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     commotion
    0.88
     newfound
    0.87
     impuls
    0.86
     acontecimientos
    0.86
     burgeoning
    0.85
     army
    0.85
     uphe
    0.85
     tumult
    0.84
     uprisings
    0.83
     besieged
    0.82
    POSITIVE LOGITS
    类型
    1.15
     especificar
    1.14
     specifies
    1.10
     specify
    1.07
     specifying
    1.01
    字段
    0.95
    精度
    0.95
     类型
    0.94
     specification
    0.94
    适配
    0.93
    Act Density 3.255%

    No Known Activations