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

    Neuronpedia

    INDEX
    Explanations

    appreciate, highlight, notice things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     اوف
    0.82
    ")){
    0.77
    したがって
    0.77
     faid
    0.76
    Selon
    0.76
    Según
    0.74
    Relevant
    0.74
    ליו
    0.74
    িনীত
    0.73
     origina
    0.73
    POSITIVE LOGITS
     how
    1.86
    how
    1.37
     HOW
    1.24
     How
    1.17
     bagaimana
    1.14
     كيف
    1.11
     hvordan
    1.09
     कैसे
    1.08
     cómo
    1.07
    How
    1.05
    Act Density 0.274%

    No Known Activations