Updates to documentation

[toddnief]

• Need to do full SFT on sub-types whenever you reinitialize

Do all the data files (train, eval and test) have to be in the same directory (data_dir) or should they be in different places?
I'm guessing I need a .env file with # includes SBATCH_MAIL, conda_yml, DSI_PARTITION and CGFP_DIR? What are each of those and how do I know what to put in each?
Where does ENV_NAME come from (it is in makefile)?
[lack of slurm knowledge] The directions state that I should "make a slurm script" and then run it using the make command. I'm guessing that I should take that code snippet and call it $(scripts_dir)train-cgfp.slurm