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

    Neuronpedia

    INDEX
    Explanations

    web portal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    齜
    -2.36
    >
    -2.34
     которое
    -2.22
     然
    -2.20
     vzduchu
    -1.97
    였
    -1.96
    大半
    -1.95
    ：「
    -1.95
    i
    -1.95
     большое
    -1.90
    POSITIVE LOGITS
    zembro
    2.34
    趼
    2.27
    过于
    2.22
    𖥧
    2.17
    $
    2.17
    ܜ
    2.14
     helicóptero
    2.06
     viewpoints
    2.05
     elástico
    2.02
    嘮
    2.02
    Act Density 0.002%

    No Known Activations