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

    Neuronpedia

    INDEX
    Explanations

    describing formats

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ات
    1.70
    től
    1.61
    pping
    1.47
    σότε
    1.47
    వ
    1.47
    το
    1.45
     וח
    1.42
    trab
    1.41
    时候
    1.41
    tiempo
    1.41
    POSITIVE LOGITS
    いた
    1.92
    <0x80>
    1.89
    おります
    1.88
    v
    1.52
    ă
    1.51
     Bridges
    1.50
     Interfaces
    1.43
    ารย์
    1.41
    <0x81>
    1.41
    .
    1.40
    Act Density 0.019%

    No Known Activations