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

    Neuronpedia

    INDEX
    Explanations

    contributing to future opportunities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     usually
    0.96
     typically
    0.88
     meestal
    0.85
     Usually
    0.83
     عادة
    0.82
     solito
    0.81
     complained
    0.81
     patchy
    0.77
    usually
    0.77
     varies
    0.74
    POSITIVE LOGITS
     contribuir
    1.08
     возможности
    1.05
     contribuer
    1.04
    能够
    1.04
    能夠
    1.02
     столь
    1.00
     Contributing
    0.97
    加入
    0.95
    能够在
    0.93
     jestem
    0.93
    Act Density 0.216%

    No Known Activations