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

    Neuronpedia

    INDEX
    Explanations

    allow write disruption find wildly

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ِ
    1.12
    ُ
    1.05
     يُ
    1.05
     اہلِ
    1.02
    નાં
    1.01
     تُ
    0.97
     —,
    0.97
     впоследствии
    0.93
    É
    0.93
    ஃப்
    0.91
    POSITIVE LOGITS
    但是
    1.04
     dva
    0.93
    非常的
    0.93
     grote
    0.85
    떄
    0.83
     goto
    0.83
     alps
    0.83
     kde
    0.82
    人和
    0.82
     इसको
    0.81
    Act Density 0.039%

    No Known Activations