fix(llm): special tokens and leading space (#1831)

Author: pabloogc

private_gpt/components/llm/custom/sagemaker.py

@@ -243,12 +243,19 @@ def get_stream():
             event_stream = resp["Body"]
             start_json = b"{"
             stop_token = "<|endoftext|>"
+            first_token = True
 
             for line in LineIterator(event_stream):
                 if line != b"" and start_json in line:
                     data = json.loads(line[line.find(start_json) :].decode("utf-8"))
-                    if data["token"]["text"] != stop_token:
+                    special = data["token"]["special"]
+                    stop = data["token"]["text"] == stop_token
+                    if not special and not stop:
                         delta = data["token"]["text"]
+                        # trim the leading space for the first token if present
+                        if first_token:
+                            delta = delta.lstrip()
+                            first_token = False
                         text += delta
                         yield CompletionResponse(delta=delta, text=text, raw=data)