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

    Neuronpedia

    INDEX
    Explanations

    specific formatting and keywords

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    芰
    0.41
     trav
    0.41
    荽
    0.41
    ர்ச்ச
    0.40
    Dia
    0.40
     സാധ്യത
    0.40
     kemungkinan
    0.40
    מו
    0.39
    Repair
    0.39
    בה
    0.39
    POSITIVE LOGITS
    ↵
    0.46
    alin
    0.46
    zing
    0.43
     всіх
    0.41
     нажмите
    0.40
     sucked
    0.39
     покри
    0.39
     всех
    0.39
     ይቀ
    0.38
     Jeter
    0.38
    Act Density 0.001%

    No Known Activations