If the text was in Unicode format, then cutting it in a "random" place, you spoiled it. The fact is that Unicode characters have a length in bytes greater than 1 (in some cases it is always 2, in some a variable number), and a BOM is possible at the beginning of the file (as far as I know, it usually has a length of 3 - odd). Therefore, it is likely that your cut came in the middle of the symbol. Faced with this, the editor will not perceive the file as Unicode, but will try to interpret it as text.
A simple solution is this: open the file as a text file, read the first x characters, and write it down, saving the format to another file. Should work.