addressed Jareds comments

megatron/inference/api.py

@@ -38,7 +38,7 @@ def generate_and_post_process(model,
                               temperature=1.0,
                               add_BOS=False,
                               use_eod_token_for_early_termination=True):
-    """Run inferecne and post-process outputs, i.e., detokenize,
+    """Run inference and post-process outputs, i.e., detokenize,
     move to cpu and convert to list."""
 
     # Main inference.

megatron/inference/forward_step.py

@@ -61,7 +61,8 @@ class ForwardStep:
                                                 max_sequence_len)
         # Pipelining arguments.
         args = get_args()
-        self.pipeline_size_larger_than_one = args.pipeline_model_parallel_size
+        self.pipeline_size_larger_than_one = (
+            args.pipeline_model_parallel_size > 1)
         # Threshold of pipelining.
         self.pipelining_batch_x_seqlen = \
             args.inference_batch_times_seqlen_threshold