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

    Neuronpedia

    INDEX
    Explanations

    named entities taking action

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ہیں۔
    0.99
     हैं
    0.90
     Есть
    0.87
     אין
    0.87
    Есть
    0.84
     ہے۔
    0.83
     आहेत
    0.82
     podendo
    0.80
    есть
    0.79
    っています
    0.79
    POSITIVE LOGITS
     took
    2.24
     became
    2.14
     went
    2.08
     began
    2.02
     gave
    1.97
     was
    1.93
     came
    1.90
     drew
    1.81
     drove
    1.76
     flew
    1.72
    Act Density 0.082%

    No Known Activations