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

    Neuronpedia

    INDEX
    Explanations

    how to improve things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ed
    2.21
    ة
    1.92
    ый
    1.67
    ت
    1.66
    न
    1.56
    ۰۰
    1.51
    変更
    1.50
    evich
    1.47
    берите
    1.46
     equated
    1.44
    POSITIVE LOGITS
    re
    1.53
    𝙖
    1.51
     upon
    1.48
     vastly
    1.46
     radically
    1.46
    𝑎
    1.41
    ਕ
    1.40
    ں
    1.38
    ra
    1.37
     Upon
    1.37
    Act Density 0.041%

    No Known Activations