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

    Neuronpedia

    INDEX
    Explanations

    examples and conditional clauses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     necessariamente
    0.98
    基本的に
    0.97
     principally
    0.96
     especially
    0.89
     ataupun
    0.86
     particularly
    0.85
     primarily
    0.82
     maupun
    0.81
     necessity
    0.80
     intricacies
    0.79
    POSITIVE LOGITS
     Suppose
    1.28
    Suppose
    1.12
     Když
    1.02
     เงี้ย
    0.95
    say
    0.94
     Duitse
    0.93
     Seorang
    0.93
     Using
    0.91
     Lotion
    0.89
     Let
    0.88
    Act Density 0.348%

    No Known Activations