[seqfan] A079826 diagonal number concats

[Kevin Ryde]

Comparing some code and data I'm not sure if the sample values of
A079826 concat number diagonals are quite right.  Or if they're right
and might want more explanation of its triangle.

The samples show a(12) = 765753403227 but I would make that value
starting 78... instead of 76...  Is the triangle made by rows numbered
alternately rightward and leftward like the following?

    row  | triangle
     1   |   1
     2   |   3  2
     3   |   4  5  6
     4   |  10  9  8  7
     5   |  11 12 13 14 15
     6   |  21 20 19 18 17 16
     7   |  22 23 24 25 26 27 28
     8   |  36 35 34 33 32 31 30 29
     9   |  37 38 39 40 41 42 43 44 45
    10   |  55 54 53 52 51 50 49 48 47 46
    11   |  56 57 58 59 60 61 62 63 64 65 66
    12   |  78 77 76 75 74 73 72 71 70 69 68 67
   ...      ...

The a(12) upwards diagonal looks like 78,57,53,...etc.  Though I see
further parts of the sample values differ by 2 or 4 or 6, which makes me
wonder if the triangle is something more subtle I haven't understood.