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

    Neuronpedia

    INDEX
    Explanations

    clone, corrupt, ensure, words, code, regardless of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     অনুরোধ
    0.52
    のように
    0.50
     линии
    0.49
    韋
    0.48
    ので
    0.48
    向量
    0.48
     свою
    0.47
     книгу
    0.46
    ержа
    0.45
    齜
    0.45
    POSITIVE LOGITS
    pti
    0.44
    pping
    0.44
     strs
    0.42
     factorial
    0.42
    pert
    0.42
    嗲
    0.41
     meats
    0.38
     factors
    0.38
     pert
    0.38
     kg
    0.38
    Act Density 0.000%

    No Known Activations