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

    Neuronpedia

    INDEX
    Explanations

    month names after of or in

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    лькі
    0.56
     гря
    0.55
    ‌ها
    0.54
    ્ર
    0.53
     zaak
    0.52
     stronę
    0.52
     যাওয়া
    0.52
     schönen
    0.50
    𝙚
    0.50
    𝗋
    0.50
    POSITIVE LOGITS
    alled
    0.67
    ो
    0.60
    ل
    0.58
    л
    0.57
    erver
    0.56
     TPU
    0.55
    ल
    0.54
    ↵
    0.52
     CEO
    0.52
    father
    0.52
    Act Density 0.004%

    No Known Activations