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

    Neuronpedia

    INDEX
    Explanations

    most interesting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     spezielle
    0.40
     Corpus
    0.39
     Же
    0.39
     Nunca
    0.39
     डिजी
    0.38
     for
    0.38
     Jr
    0.38
    stored
    0.38
     Nig
    0.37
     uber
    0.37
    POSITIVE LOGITS
    3
    0.49
     değişt
    0.46
    4
    0.43
     పెరుగు
    0.41
     аргу
    0.40
    6
    0.40
     പ്രതീ
    0.40
    峮
    0.39
    ང
    0.39
     కా
    0.37
    Act Density 0.004%

    No Known Activations