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

    Neuronpedia

    INDEX
    Explanations

    pragmatic, dogmatic, systematic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Armstrong
    0.42
    cycled
    0.38
    else
    0.38
    ossus
    0.37
    erode
    0.37
    Accordingly
    0.36
    iter
    0.35
    বৈ
    0.35
    urry
    0.35
    IRON
    0.35
    POSITIVE LOGITS
    matic
    1.44
    мати
    1.38
    MATIC
    1.25
    matism
    1.21
    matics
    1.17
    matically
    1.16
    MAT
    1.10
    mat
    1.09
    matik
    1.09
    matis
    1.06
    Act Density 0.012%

    No Known Activations