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

    Neuronpedia

    INDEX
    Explanations

    ranked lists described

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ir
    0.44
    porque
    0.42
    原因是
    0.42
    here
    0.41
    Anyway
    0.41
    wa
    0.40
    rt
    0.39
    Sand
    0.39
     زیرا
    0.39
    it
    0.38
    POSITIVE LOGITS
    явля
    0.42
    可谓
    0.42
     (\
    0.39
     consideradas
    0.39
     ALSO
    0.38
     também
    0.38
     nicknamed
    0.37
     imageUrl
    0.37
     lastly
    0.37
    $.\\
    0.36
    Act Density 0.004%

    No Known Activations