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

    Neuronpedia

    INDEX
    Explanations

    scientific observation and empirical evidence

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     komfort
    0.47
     savait
    0.45
     bezel
    0.44
    逛
    0.44
     menyimpan
    0.43
    任務
    0.43
    voicing
    0.42
    枢
    0.42
     timezone
    0.42
     وګټ
    0.42
    POSITIVE LOGITS
     empirical
    1.23
     empir
    1.06
     observation
    1.05
     Empirical
    1.04
    empirical
    1.03
     empirically
    0.94
     experiment
    0.93
    observation
    0.88
     scientific
    0.86
     observación
    0.86
    Act Density 0.028%

    No Known Activations