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

    Neuronpedia

    INDEX
    Explanations

    how it works

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    どのような
    0.83
    oribus
    0.82
    omiast
    0.78
    เพียง
    0.78
     véritable
    0.78
    らかじめ
    0.77
    unable
    0.75
    没有任何
    0.74
    வதும்
    0.74
     inability
    0.74
    POSITIVE LOGITS
     works
    1.96
     work
    1.68
     Works
    1.67
     worked
    1.45
    Works
    1.43
    works
    1.41
     Work
    1.36
     Worked
    1.31
     WORKS
    1.29
     WORK
    1.27
    Act Density 0.355%

    No Known Activations