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

    Neuronpedia

    INDEX
    Explanations

    good if or for

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    governmental
    0.40
    ("../../
    0.40
    unächst
    0.38
    originally
    0.38
     Adapun
    0.38
    wezig
    0.37
    гел
    0.37
    etés
    0.37
    हालय
    0.37
    patri
    0.37
    POSITIVE LOGITS
    适用于
    0.51
     spesso
    0.49
    适合
    0.48
     often
    0.45
     especially
    0.45
     if
    0.45
     ideal
    0.44
     useful
    0.44
     часто
    0.43
    เหมาะ
    0.43
    Act Density 0.198%

    No Known Activations