Here is the sequence logo predicted by the SAM server:

H = helix
E = sheet
C = loop (or coil)

The taller the letter at each position, the higher the probability of that specific secondary structure for that amino acid.

For example, the Valines at residues 38, 63, 111 & 116 are highly predicted to be part of a sheet. However, the Valines at residues 87 & 142 are not predicted to be anything!

The above predictions are already on the starting model (based on the horizontal lines in the above figure).