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

    Neuronpedia

    INDEX
    Explanations

    new client console alert

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𝙙
    0.84
    𝐥
    0.80
    ف
    0.79
    ปี
    0.78
    at
    0.77
    ا
    0.77
    𝐝
    0.75
     ->
    0.74
    𝙩
    0.74
    4
    0.73
    POSITIVE LOGITS
    ol
    0.77
    മാര്‍
    0.74
    '
    0.71
    ing
    0.71
     ০৯
    0.70
    কর্ম
    0.69
    ET
    0.68
     Atención
    0.67
    മാർ
    0.66
    cion
    0.66
    Act Density 0.007%

    No Known Activations