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

    Neuronpedia

    INDEX
    Explanations

    batch

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Königreich
    0.66
    uo
    0.66
     législ
    0.63
    henes
    0.63
    烃
    0.62
     மய
    0.62
    beschreibung
    0.62
     обязан
    0.61
     कुठे
    0.61
    0.61
    POSITIVE LOGITS
     batch
    4.27
    batch
    3.94
     Batch
    3.87
    Batch
    3.86
     batches
    3.75
    BATCH
    3.11
    batches
    3.00
     बैच
    2.84
    批量
    2.75
    批
    2.74
    Act Density 0.231%

    No Known Activations