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

    Neuronpedia

    INDEX
    Explanations

    quick wins by

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    р
    1.01
    на
    0.91
    т
    0.88
    릭터
    0.86
    ण्यात
    0.82
    wounded
    0.81
    Benzyl
    0.80
    তোমার
    0.80
    ять
    0.79
    yang
    0.77
    POSITIVE LOGITS
     gratification
    1.14
     paced
    1.06
    aneous
    1.02
     prototyping
    0.96
     speeds
    0.94
    aneously
    0.93
    बाजी
    0.89
    sand
    0.87
     turnaround
    0.86
     prêts
    0.86
    Act Density 0.500%

    No Known Activations