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

    Neuronpedia

    INDEX
    Explanations

    notes and closing statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     net
    0.75
     her
    0.70
    Object
    0.67
     newItem
    0.65
     happiness
    0.64
     Ret
    0.64
     Def
    0.61
     Exception
    0.60
     Mohawk
    0.60
     Patrimonio
    0.60
    POSITIVE LOGITS
    anded
    1.04
    ونَ
    0.97
     úgy
    0.96
    安排
    0.91
    ென்று
    0.90
    ຕິດຕໍ່
    0.89
     llamó
    0.88
    ፡
    0.88
    «,
    0.88
    riamo
    0.88
    Act Density 0.035%

    No Known Activations