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

    Neuronpedia

    INDEX
    Explanations

    some of the

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     перечис
    0.50
     these
    0.46
     этих
    0.46
     two
    0.45
     four
    0.45
     እነዚህ
    0.43
     three
    0.43
     них
    0.42
     esas
    0.41
     две
    0.41
    POSITIVE LOGITS
    Some
    0.43
    Different
    0.43
    不同的
    0.42
    Reduce
    0.40
    的不同
    0.39
    不同
    0.38
    Sil
    0.35
    h
    0.35
     forskellige
    0.35
    reduce
    0.35
    Act Density 0.001%

    No Known Activations