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

    Neuronpedia

    INDEX
    Explanations

    reviewed and approved

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     dạng
    0.75
    aan
    0.72
    ר
    0.67
     reh
    0.64
     wyją
    0.61
    सि
    0.61
     перво
    0.60
    ar
    0.59
    ゥ
    0.59
     invers
    0.58
    POSITIVE LOGITS
     Tính
    0.96
    ڍ
    0.95
    горь
    0.93
     gesam
    0.91
     وتم
    0.91
     फ्लावर
    0.91
    และ
    0.90
     También
    0.89
     Gentile
    0.88
    xavier
    0.87
    Act Density 0.001%

    No Known Activations