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

    Neuronpedia

    INDEX
    Explanations

    stopping or inactive states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    akl
    0.40
     ভারসাম্য
    0.39
     shipbuilding
    0.38
     monadic
    0.38
    iglas
    0.37
     strengthening
    0.37
    Late
    0.37
    过滤器
    0.37
     constructing
    0.37
     Always
    0.37
    POSITIVE LOGITS
     restart
    0.51
     frozen
    0.47
     inactive
    0.46
    inactive
    0.46
     slumber
    0.46
     reactivation
    0.45
    Inactive
    0.45
     inert
    0.45
     Restart
    0.45
     idle
    0.44
    Act Density 0.066%

    No Known Activations