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

    Neuronpedia

    INDEX
    Explanations

    dialogue start marker

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Because
    0.88
     Then
    0.85
     During
    0.85
     When
    0.84
     Before
    0.81
     But
    0.79
     Without
    0.79
     To
    0.79
     Which
    0.78
     Thats
    0.76
    POSITIVE LOGITS
     పరీక్ష
    0.80
    ロット
    0.78
    сюда
    0.77
    更大
    0.76
     Fundación
    0.75
     Allí
    0.74
    ቛ
    0.74
    ীরে
    0.73
     núcleo
    0.73
     краса
    0.73
    Act Density 0.101%

    No Known Activations