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

    Neuronpedia

    INDEX
    Explanations

    if ... then / greater than

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     inc
    0.41
    孔
    0.39
    frm
    0.37
     lám
    0.37
     illumination
    0.37
     món
    0.36
     Deaf
    0.36
     optimization
    0.35
     illumin
    0.35
    مثل
    0.35
    POSITIVE LOGITS
     surely
    0.52
    ということは
    0.48
    既然
    0.46
     Surely
    0.46
    那么
    0.45
    ன்றால்
    0.42
     Then
    0.42
    证明
    0.41
     allora
    0.41
     then
    0.41
    Act Density 0.052%

    No Known Activations