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

    Neuronpedia

    INDEX
    Explanations

    simulates, simulations, simulated

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ક
    3.13
     omissions
    2.96
     ilang
    2.93
    -\
    2.85
    𝑵
    2.82
    smallest
    2.75
     σχετικά
    2.68
     lau
    2.65
    Tidak
    2.63
     principais
    2.62
    POSITIVE LOGITS
    ت
    5.67
    т
    3.56
    р
    3.53
    تهم
    3.51
    ট
    3.44
    ingly
    3.32
    ر
    3.25
     annealing
    3.23
    د
    3.10
    ulating
    3.09
    Act Density 0.045%

    No Known Activations