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

    Neuronpedia

    INDEX
    Explanations

    results and confirmations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     لهذه
    0.49
     কৌশল
    0.41
     Requires
    0.39
    就开始
    0.39
     requires
    0.39
     lately
    0.39
     작업
    0.38
     যতক্ষণ
    0.38
     хотим
    0.38
     scoping
    0.38
    POSITIVE LOGITS
     regretted
    0.52
     satisfe
    0.50
     thanked
    0.49
    Ironically
    0.49
     irony
    0.47
     feliz
    0.46
    ಲಾಯಿತು
    0.46
    ardon
    0.46
    umumkan
    0.45
    thanks
    0.45
    Act Density 0.023%

    No Known Activations