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

    Neuronpedia

    INDEX
    Explanations

    the model

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     menghindari
    0.67
     avoiding
    0.61
     trị
    0.61
     avoidance
    0.60
    ्तीय
    0.60
    ഡിയ
    0.60
     }=\
    0.59
     avoids
    0.59
     oh
    0.59
     reglamento
    0.59
    POSITIVE LOGITS
     Sabb
    0.70
     Independ
    0.64
    balances
    0.63
    یکیشن
    0.62
     Shou
    0.57
     Spacer
    0.56
    balanced
    0.56
    Satellite
    0.56
    gradation
    0.55
     қу
    0.55
    Act Density 0.218%

    No Known Activations