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

    Neuronpedia

    INDEX
    Explanations

    arithmetic operations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     alone
    0.79
     solo
    0.70
     recognition
    0.69
     around
    0.67
     बोलने
    0.63
     several
    0.63
     Android
    0.62
    alone
    0.62
     विचार
    0.60
     multiple
    0.60
    POSITIVE LOGITS
     Warming
    0.81
    тет
    0.74
    柒
    0.71
    ഌ
    0.71
     инг
    0.70
    тит
    0.70
     rohkem
    0.70
     divertida
    0.68
     enqueue
    0.68
    悆
    0.68
    Act Density 0.070%

    No Known Activations