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

    Neuronpedia

    INDEX
    Explanations

    terms being called

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    世界的
    0.45
    以外の
    0.39
    मेटिक
    0.37
    居然
    0.37
    সমস্ত
    0.37
    ამის
    0.37
    выя
    0.36
     দেশটি
    0.36
     அறிய
    0.36
    领域的
    0.36
    POSITIVE LOGITS
     appunto
    0.76
    ',
    0.63
    と呼ばれる
    0.62
     genoemd
    0.61
     genannt
    0.59
     라고
    0.59
     colloqu
    0.57
     bezeichnet
    0.56
     interchangeably
    0.55
     이라고
    0.54
    Act Density 0.052%

    No Known Activations