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

    Neuronpedia

    INDEX
    Explanations

    global configuration

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ⴷ
    0.44
    ารถ
    0.44
    敵
    0.44
    він
    0.43
    ವರೆಗೆ
    0.43
    다라고
    0.42
    𝟐
    0.42
    ською
    0.41
    ృష్
    0.41
     RANGE
    0.41
    POSITIVE LOGITS
     If
    0.56
    ↵
    0.52
     if
    0.43
     El
    0.43
     When
    0.42
    c
    0.42
     eğer
    0.41
     It
    0.41
     tipped
    0.41
    ;
    0.41
    Act Density 0.001%

    No Known Activations