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

    Neuronpedia

    INDEX
    Explanations

    gender and relationships

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Validator
    0.55
     टर्मिनल
    0.53
     deslig
    0.52
     अंग्रेज
    0.51
     Пер
    0.50
    Validator
    0.50
     strait
    0.50
     perone
    0.50
    是中国
    0.50
     Архивировано
    0.49
    POSITIVE LOGITS
     slope
    0.63
    slope
    0.60
    当該
    0.56
    తు
    0.56
    overwrite
    0.56
    Slope
    0.55
     Goto
    0.55
    Этот
    0.53
     refute
    0.53
     Aquarelle
    0.53
    Act Density 0.000%

    No Known Activations