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

    Neuronpedia

    INDEX
    Explanations

    code, properties, position

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    _
    0.40
    \_
    0.37
    asures
    0.37
    This
    0.35
    减少
    0.35
    UTION
    0.34
    Index
    0.33
    англ
    0.33
    Sprites
    0.33
    𝐀
    0.33
    POSITIVE LOGITS
     اگر
    0.41
     ovako
    0.41
     misalnya
    0.41
     accanto
    0.38
     below
    0.38
     jeśli
    0.38
     beispielsweise
    0.38
     если
    0.36
     например
    0.36
     bijvoorbeeld
    0.36
    Act Density 0.000%

    No Known Activations