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

    Neuronpedia

    INDEX
    Explanations

    how it works

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     chỉ
    0.37
     only
    0.37
     zeer
    0.35
    ہ
    0.34
    ﻪ
    0.33
    有很多
    0.33
     nothing
    0.32
     sehr
    0.32
     parecía
    0.31
     teníamos
    0.31
    POSITIVE LOGITS
     behaves
    0.48
     behave
    0.47
     differ
    0.44
     behaved
    0.44
    看待
    0.43
     differs
    0.43
     fonctionne
    0.41
     interacts
    0.40
     relates
    0.39
     функциони
    0.39
    Act Density 0.452%

    No Known Activations