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

    Neuronpedia

    INDEX
    Explanations

    express and middleware

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     서버
    0.48
    サーバー
    0.46
     servidor
    0.45
     hostname
    0.45
    服务器
    0.43
     modelu
    0.41
    Hostname
    0.41
     сервера
    0.41
    服务端
    0.40
    砝
    0.39
    POSITIVE LOGITS
     bath
    0.42
     Darn
    0.39
     Chamb
    0.38
     foll
    0.38
     ihn
    0.38
     Sikh
    0.38
     remedial
    0.37
     Bath
    0.37
    äter
    0.37
     इंट्रो
    0.37
    Act Density 0.012%

    No Known Activations