SQL - return top 10 rows by category

Hello experts,

I have a script here, that seems to be working and returning me good data, but I only want the top 10 rows by pm.description column.  When I add the rownumber column, I can't seem to use an effective where <= 10 on it.:


DECLARE @MONTH1 VARCHAR(20),
            @MONTH2 VARCHAR(20),
            @MONTH3 VARCHAR(20),
            @MONTH4 VARCHAR(20),
            @MONTH5 VARCHAR(20),
            @MONTH6 VARCHAR(20),
            @MONTH7 VARCHAR(20),
            @MONTH8 VARCHAR(20),
            @MONTH9 VARCHAR(20),
            @MONTH10 VARCHAR(20),
            @MONTH11 VARCHAR(20),
            @MONTH12 VARCHAR(20)
            

set @MONTH1 =  DATENAME(month, getdate())
set @MONTH2 =  DATENAME(month, getdate()-30)
set @MONTH3 =  DATENAME(month, getdate()-60)
set @MONTH4 =  DATENAME(month, getdate()-90)
set @MONTH5 =  DATENAME(month, getdate()-120)
set @MONTH6 =  DATENAME(month, getdate()-150)
set @MONTH7 =  DATENAME(month, getdate()-180)
set @MONTH8 =  DATENAME(month, getdate()-210)
set @MONTH9 =  DATENAME(month, getdate()-240)
set @MONTH10 =  DATENAME(month, getdate()-270)
set @MONTH11 =  DATENAME(month, getdate()-300)
set @MONTH12 =  DATENAME(month, getdate()-330)


exec('select
         ROW_NUMBER() OVER ( PARTITION BY pm.description ORDER BY count( pd.diagnosis_code_id) DESC ) AS RowNumber,
       pm.description
       , pd.diagnosis_code_id
       , count( pe.enc_timestamp ) as count_orders
       , sum(case when pd.create_timestamp >= dateadd(day,-30,cast(getdate() as datetime))
                  then 1 else 0 end)
         as [' + @MONTH1 + ']
       , sum(case when pd.create_timestamp >= dateadd(day,-60,cast(getdate() as datetime))
                   and pd.create_timestamp < dateadd(day,-30,cast(getdate() as datetime))
                  then 1 else 0 end)
         as [' + @MONTH2 + ']
       , sum(case when pd.create_timestamp >= dateadd(day,-90,cast(getdate() as datetime))
                   and pd.create_timestamp < dateadd(day,-60,cast(getdate() as datetime))
                  then 1 else 0 end)
         as [' + @MONTH3 + ']
                , sum(case when pd.create_timestamp >= dateadd(day,-120,cast(getdate() as datetime))
                   and pd.create_timestamp < dateadd(day,-90,cast(getdate() as datetime))
                  then 1 else 0 end)
         as [' + @MONTH4 + ']
                , sum(case when pd.create_timestamp >= dateadd(day,-150,cast(getdate() as datetime))
                   and pd.create_timestamp < dateadd(day,-120,cast(getdate() as datetime))
                  then 1 else 0 end)
         as [' + @MONTH5 + ']
                , sum(case when pd.create_timestamp >= dateadd(day,-180,cast(getdate() as datetime))
                   and pd.create_timestamp < dateadd(day,-150,cast(getdate() as datetime))
                  then 1 else 0 end)
         as [' + @MONTH6 + ']
                , sum(case when pd.create_timestamp >= dateadd(day,-210,cast(getdate() as datetime))
                   and pd.create_timestamp < dateadd(day,-180,cast(getdate() as datetime))
                  then 1 else 0 end)
         as [' + @MONTH7 + ']
                , sum(case when pd.create_timestamp >= dateadd(day,-240,cast(getdate() as datetime))
                   and pd.create_timestamp < dateadd(day,-210,cast(getdate() as datetime))
                  then 1 else 0 end)
         as [' + @MONTH8 + ']
                , sum(case when pd.create_timestamp >= dateadd(day,-270,cast(getdate() as datetime))
                   and pd.create_timestamp < dateadd(day,-240,cast(getdate() as datetime))
                  then 1 else 0 end)
         as [' + @MONTH9 + ']
                , sum(case when pd.create_timestamp >= dateadd(day,-300,cast(getdate() as datetime))
                   and pd.create_timestamp < dateadd(day,-270,cast(getdate() as datetime))
                  then 1 else 0 end)
         as [' + @MONTH10 + ']
                , sum(case when pd.create_timestamp >= dateadd(day,-330,cast(getdate() as datetime))
                   and pd.create_timestamp < dateadd(day,-300,cast(getdate() as datetime))
                  then 1 else 0 end)
         as [' + @MONTH11 + ']
                , sum(case when pd.create_timestamp >= dateadd(day,-360,cast(getdate() as datetime))
                   and pd.create_timestamp < dateadd(day,-330,cast(getdate() as datetime))
                  then 1 else 0 end)
         as [' + @MONTH12 + ']
from patient_diagnosis pd
join patient_encounter pe
on pe.enc_id = pd.enc_id and pe.person_id = pd.person_id
join provider_mstr pm
on pd.provider_id = pm.provider_id
where
      pd.diagnosis_code_id is not null
      and RowNumber <= 10
group by
      pm.description
      , pd.diagnosis_code_id
      order by pm.description,  count( pd.diagnosis_code_id) desc, pd.diagnosis_code_id')

I get an invalid column name Row Number on the "and RowNumber <= 10" at the bottom.

Thoughts?

Thanks!