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

    Neuronpedia

    INDEX
    Explanations

    changing mimic data tuning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    remove
    0.41
    chlor
    0.40
     た
    0.39
    一
    0.38
    purch
    0.38
    ittaa
    0.37
    pages
    0.37
    圖片
    0.37
    một
    0.37
    istaa
    0.37
    POSITIVE LOGITS
    Output
    0.43
    Weight
    0.40
     bioquímica
    0.40
     erziel
    0.38
     tonal
    0.38
     franche
    0.37
    敚
    0.37
     केली
    0.37
     flancs
    0.36
     coerced
    0.36
    Act Density 0.058%

    No Known Activations