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

    Neuronpedia

    INDEX
    Explanations

    very strict

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     asynchronously
    0.38
    చన
    0.37
    瓤
    0.34
     exponentes
    0.33
     Larger
    0.33
     handy
    0.32
    கை
    0.32
     удобно
    0.32
    বাদ
    0.32
     دب
    0.32
    POSITIVE LOGITS
     strict
    3.75
    strict
    3.31
     Strict
    3.25
    严格
    3.06
    Strict
    3.06
     stringent
    3.03
     strictest
    2.95
     stricter
    2.94
    厳
    2.89
    严
    2.86
    Act Density 0.126%

    No Known Activations