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

    Neuronpedia

    INDEX
    Explanations

    allocate a little

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Plugins
    0.73
    s
    0.68
     eve
    0.66
     địch
    0.66
     faire
    0.66
     utilisateur
    0.66
     દિવસે
    0.64
     Doremi
    0.64
     be
    0.63
     chi
    0.62
    POSITIVE LOGITS
    ﻘ
    0.95
    ﻟ
    0.93
    ﻠ
    0.91
    ﻌ
    0.90
    ﺤ
    0.90
    ﻰ
    0.89
    ﺭ
    0.88
     inglesa
    0.86
    ﺔ
    0.84
    краё
    0.83
    Act Density 0.001%

    No Known Activations