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

    Neuronpedia

    INDEX
    Explanations

    while introducing contrast

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Cuando
    0.68
     cuando
    0.63
    Когда
    0.61
     nếu
    0.61
     quando
    0.60
    When
    0.59
    cuando
    0.59
    Quando
    0.58
     when
    0.58
     когда
    0.58
    POSITIVE LOGITS
     there
    0.93
     they
    0.84
     we
    0.72
     it
    0.72
     this
    0.72
     यह
    0.66
     هناك
    0.66
    there
    0.66
     overshadowed
    0.63
     everyone
    0.61
    Act Density 0.293%

    No Known Activations