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

    Neuronpedia

    INDEX
    Explanations

    bullet points followed by colon

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Adapun
    1.15
     （
    1.05
     ۔۔۔
    1.02
    𝗶
    1.02
     关于
    1.00
     まず
    1.00
     從
    0.99
     proposé
    0.99
     Lastly
    0.99
     whopping
    0.98
    POSITIVE LOGITS
    _
    0.82
    :
    0.80
    ::
    0.68
    -
    0.66
    └──
    0.65
    >):
    0.60
    <eos>
    0.59
    ):
    0.58
    :$
    0.58
    ,
    0.57
    Act Density 0.240%

    No Known Activations