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

    Neuronpedia

    INDEX
    Explanations

    thread_rng, OsRng, Read, Seed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Produces
    0.42
    creates
    0.40
     तळ
    0.39
    UWORD
    0.38
     воздуха
    0.38
    ture
    0.38
     dard
    0.38
    तल
    0.37
    piro
    0.37
    Pyrid
    0.37
    POSITIVE LOGITS
     രണ്ടു
    0.38
     Propagation
    0.37
     modesty
    0.36
    ੍ਰ
    0.36
     mengisi
    0.36
     ನ್ಯ
    0.36
     deeln
    0.36
     강
    0.35
    гістра
    0.35
    bg
    0.35
    Act Density 0.000%

    No Known Activations