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

    Neuronpedia

    INDEX
    Explanations

    vibrates and instability

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     funcionalidad
    0.94
     пробе
    0.94
     ausschließlich
    0.93
     fonctionnalité
    0.92
     निर्धारक
    0.92
    OpportunitiesBy
    0.92
    loginfo
    0.91
     utilidad
    0.90
     meghatá
    0.89
     Datenschutz
    0.88
    POSITIVE LOGITS
     unstable
    2.06
     instability
    2.04
     instabilities
    1.81
     stability
    1.71
     shaky
    1.59
     Stability
    1.55
    stability
    1.50
    Stability
    1.49
    不稳定
    1.48
     fragile
    1.45
    Act Density 0.174%

    No Known Activations