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

    Neuronpedia

    INDEX
    Explanations

    resolving or removing errors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     blessed
    0.41
    KeyPress
    0.39
    कर
    0.38
     Chart
    0.36
     પર
    0.35
     clocked
    0.34
    بھ
    0.34
     CHART
    0.34
     HAZ
    0.33
    สอบ
    0.32
    POSITIVE LOGITS
     pyrite
    0.44
     ভারী
    0.42
    starch
    0.41
    erweise
    0.41
     造
    0.40
     մի
    0.40
    ՜
    0.39
     :).
    0.39
    ⁡
    0.39
    濘
    0.39
    Act Density 0.001%

    No Known Activations