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

    Neuronpedia

    INDEX
    Explanations

    naked eye

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    r
    0.67
    па
    0.66
    ,
    0.66
    ン
    0.64
    ли
    0.63
    ।
    0.62
    st
    0.58
    ти
    0.58
    パリ
    0.57
    וס
    0.57
    POSITIVE LOGITS
    at
    0.62
     billion
    0.57
     by
    0.56
     l
    0.55
     detection
    0.54
    3
    0.54
     of
    0.53
     особое
    0.52
     do
    0.51
     commander
    0.49
    Act Density 0.000%

    No Known Activations