How to find duplicate bunches of rows for differrent group

SELECT DISTINCT
  SUBSTRING(Grp_kit_NO, 0, CHARINDEX('_', Grp_kit_NO)) GRP_KIT_NO,  0 rownum INTO #temp
FROM group_kit_detail
DECLARE @i int = 0
UPDATE #temp
SET @i = rownum = @i + 1

SELECT
  GRP_KIT_NO,
  CONVERT(varchar(max), '') comp,
  0 RW INTO #T1
FROM GROUP_KIT_BOM
WHERE 1 = 2
SELECT
  CONVERT(varchar(max), '') GRP_KIT_NO,
  CONVERT(varchar(max), '') Match_GRP_KIT_NO INTO #Final
FROM GROUP_KIT_BOM
WHERE 1 = 2


DECLARE @RowCounter int = 1
WHILE @RowCounter <= (SELECT
    COUNT(1)
  FROM #temp)
BEGIN
  DECLARE @MasterGroup varchar(50) = ''
  SELECT
    @MasterGroup = GRP_KIT_NO
  FROM #temp
  WHERE rownum = @RowCounter

  SELECT
    GRP_KIT_NO,
    COMPONENT + COMP_TYPE + QTY comp,
    0 Row1 INTO #TInput
  FROM GROUP_KIT_BOM
  WHERE SUBSTRING(Grp_kit_NO, 0, CHARINDEX('_', Grp_kit_NO)) = @MasterGroup
  ORDER BY GRP_KIT_NO, comp
  DECLARE @j int = 0
  UPDATE #TInput
  SET @j = Row1 = @j + 1

  INSERT INTO #T1
    SELECT
      GRP_KIT_NO,
      COMP = STUFF((SELECT
        ',' + COMP
      FROM #TInput A
      WHERE A.GRP_KIT_NO = B.GRP_KIT_NO     
      FOR xml PATH ('')), 1, 1, ''),
      0
    FROM #TInput B
    GROUP BY B.GRP_KIT_NO
    ORDER BY COMP

  DROP TABLE #TInput
  SET @RowCounter = @RowCounter + 1

END

DECLARE @k int = 0
UPDATE #T1
SET @k = RW = @k + 1


DECLARE @Counter int = 1-----------------matching comp

WHILE @Counter <= (SELECT
    COUNT(1)
  FROM #T1)
BEGIN

  DECLARE @Group varchar(50) = ''
  SELECT
    @Group = GRP_KIT_NO
  FROM #T1
  WHERE RW = @Counter

  INSERT INTO #Final
    SELECT DISTINCT
      A.GRP_KIT_NO,
      B.GRP_KIT_NO
    FROM #T1 A,
         #T1 B
    WHERE B.GRP_KIT_NO <> @Group
    AND A.comp = B.comp
    AND A.GRP_KIT_NO <> B.GRP_KIT_NO
    ORDER BY A.GRP_KIT_NO

  SET @Counter = @Counter + 1
END
 
SELECT
  *
FROM #Final
		
DROP TABLE #Final
DROP TABLE #T1
DROP TABLE #TEMP