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

    Neuronpedia

    INDEX
    Explanations

    AI, RNN, Transformers, information, timestamp

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     always
    0.52
     S
    0.51
     sempre
    0.50
    ždy
    0.49
     ATP
    0.46
     Inland
    0.46
     također
    0.46
     आजकल
    0.46
     $\
    0.46
     titular
    0.46
    POSITIVE LOGITS
    లి
    0.50
    ленных
    0.48
     ਵੱ
    0.48
    ᅵ
    0.48
    找到了
    0.46
    出
    0.46
    著
    0.42
    鸴
    0.42
    rived
    0.42
    镞
    0.42
    Act Density 0.001%

    No Known Activations