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

    Neuronpedia

    INDEX
    Explanations

    language fluency

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     които
    0.42
     Writes
    0.41
     пре
    0.39
     си
    0.39
    Из
    0.39
     създа
    0.38
    ্যালি
    0.37
    дона
    0.37
    ко
    0.36
     въз
    0.36
    POSITIVE LOGITS
     isso
    0.58
     meios
    0.57
     அது
    0.54
     criteri
    0.54
     hasilnya
    0.54
    積極的に
    0.54
     velocidad
    0.52
     ഇത്
    0.52
     ainda
    0.52
     aún
    0.52
    Act Density 0.001%

    No Known Activations