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

    Neuronpedia

    INDEX
    Explanations

    a cmd command

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
      
    0.94
     Conversely
    0.64
     Importantly
    0.59
     -,
    0.59
     lingers
    0.55
    Comb
    0.55
     Notably
    0.55
    ல்ப
    0.54
    Ultimately
    0.53
    ്ലാ
    0.51
    POSITIVE LOGITS
     poems
    1.36
     函数
    1.25
    函数
    1.23
     pokemon
    1.23
     Poems
    1.23
     python
    1.23
     poem
    1.21
     Python
    1.19
     فونبٹ
    1.19
     SQL
    1.18
    Act Density 2.018%

    No Known Activations