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

    Neuronpedia

    INDEX
    Explanations

    introducing explanations or lists

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     فى
    0.92
     katika
    0.90
    Trong
    0.87
     në
    0.86
    utilise
    0.85
    യ്ക്ക്
    0.84
    utilisation
    0.84
     dalam
    0.83
     використовувати
    0.82
     kwenye
    0.82
    POSITIVE LOGITS
     because
    1.73
     indicating
    1.56
     stating
    1.48
     reminding
    1.46
     explaining
    1.45
     emphasizing
    1.45
     informing
    1.43
     inquiring
    1.41
     mentioning
    1.40
     insisting
    1.40
    Act Density 0.000%

    No Known Activations