Merge pull request EleutherAI#993 from mayankjobanputra/patch-1

default tokenizer_type should be string
ReloJeffrey · Jul 8, 2023 · 303d7be · 303d7be
2 parents 700219b + 90df174
commit 303d7be
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/megatron/tokenizer/train_tokenizer.py b/megatron/tokenizer/train_tokenizer.py
@@ -102,7 +102,7 @@ def parse_args():
         type=str,
         help="type of tokenizer to train, currently only BPE is supported",
         choices=["BPE"],
-        default=["BPE"],
+        default="BPE",
     )
     parser.add_argument(
         "-v",