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

    Neuronpedia

    INDEX
    Explanations

    leaving worlds behind

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    力を
    0.44
    ्हान
    0.44
     درست
    0.41
     nhiên
    0.39
    されていない
    0.38
    screwdriver
    0.37
    ⎠
    0.37
    لمين
    0.36
    escence
    0.36
    功率
    0.36
    POSITIVE LOGITS
     From
    0.41
     Four
    0.41
     ISBN
    0.40
     Cl
    0.39
     Wine
    0.39
     Maybe
    0.38
    astic
    0.38
     Leaving
    0.37
    all
    0.37
     Vielleicht
    0.37
    Act Density 0.007%

    No Known Activations