Augmented Max input token

app.py

@@ -16,7 +16,7 @@ print(f"CUDA device: {torch.cuda.get_device_name(torch.cuda.current_device())}")
 MAX_MAX_NEW_TOKENS = 4096
 DEFAULT_MAX_NEW_TOKENS = 2048
 # MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
-MAX_INPUT_TOKEN_LENGTH = 20000
+MAX_INPUT_TOKEN_LENGTH = 100000
 
 DEFAULT_USER_QUERY = '''
 @prefix : <urn:webprotege:ontology:7272b2af-011f-4d40-8519-9fc3f830442e#> .
@@ -867,7 +867,7 @@ def generate(
 
 
         streamer = TextIteratorStreamer(tokenizer,
-                                        timeout=60.0,
+                                        timeout=2*60.0,
                                         skip_prompt=True,
                                         skip_special_tokens=True)