meant to set always_save_checkpoint to False instead, so we only writ…

…e when val improves

config/train_shakespeare_char.py

@@ -7,7 +7,7 @@
 log_interval = 10 # don't print too too often
 
 # we expect to overfit on this small dataset, so only save when val improves
-always_save_checkpoint = True
+always_save_checkpoint = False
 
 wandb_log = False # override via command line if you like
 wandb_project = 'shakespeare-char'