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

    Neuronpedia

    INDEX
    Explanations

    code comments and closing brackets

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     និង
    0.51
     आणि
    0.46
    정과
    0.45
     ಮತ್ತು
    0.44
    yta
    0.44
    indest
    0.43
    ᱭ
    0.43
    ികളും
    0.42
     helst
    0.41
     и
    0.41
    POSITIVE LOGITS
    ↵↵↵↵↵
    0.55
    ↵↵↵
    0.55
    ↵↵
    0.53
    ↵↵↵↵
    0.52
     Advantages
    0.46
    ↵↵↵↵↵↵
    0.44
    ↵↵↵↵↵↵↵↵↵
    0.43
    ↵↵↵↵↵↵↵↵↵↵↵
    0.42
    ↵↵↵↵↵↵↵↵
    0.41
     vagy
    0.41
    Act Density 0.021%

    No Known Activations