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

    Neuronpedia

    INDEX
    Explanations

    /dev/null sequence

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     overtaken
    0.49
     непотпуним
    0.41
     unaltered
    0.39
     क्यू
    0.39
     unbroken
    0.38
     والدہ
    0.38
     shaken
    0.37
    প্প
    0.37
     incidental
    0.36
    鑾
    0.36
    POSITIVE LOGITS
     dev
    0.55
     abyss
    0.51
    dev
    0.51
    Dev
    0.49
    ブラック
    0.47
     discard
    0.47
     junk
    0.46
     чер
    0.46
     silence
    0.45
     Dev
    0.45
    Act Density 0.001%

    No Known Activations