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

    Neuronpedia

    INDEX
    Explanations

    expressing i or we statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    하였다
    0.61
    其
    0.56
     하였다
    0.54
    하였습니다
    0.52
    原来
    0.52
     एवं
    0.52
    ത്വം
    0.52
    使其
    0.51
    하였
    0.50
    原來
    0.48
    POSITIVE LOGITS
     definitely
    0.79
     aren
    0.78
     KNOW
    0.73
     know
    0.73
     certainly
    0.73
    确实
    0.72
     knows
    0.71
     isn
    0.67
    know
    0.64
     recognize
    0.62
    Act Density 0.114%

    No Known Activations