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

    Neuronpedia

    INDEX
    Explanations

    improvements and

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     COMPOUNDS
    0.39
    Depending
    0.39
     যদি
    0.38
     atrophy
    0.38
    normally
    0.38
    terbury
    0.37
     underdog
    0.37
     wither
    0.37
     depending
    0.36
     यदि
    0.36
    POSITIVE LOGITS
     improved
    0.70
    improved
    0.66
     verbess
    0.65
     amélior
    0.64
     Verbesser
    0.63
     added
    0.62
     добавлен
    0.61
     улуч
    0.60
    改进
    0.59
     updated
    0.59
    Act Density 0.126%

    No Known Activations