Added tokenizer with large sinhala corpus

config.json

@@ -1,9 +1,9 @@
 {
     "unknown_token": "<|unk|>",
     "pad_token": "<|pad|>",
-    "unknown_token_id": 1015,
-    "pad_token_id": 1016,
-    "max_length": 256,
-    "end_of_text_token": "<|endoftext|>",
-    "end_of_text_token_id": 1017
+    "unknown_token_id": 576,
+    "pad_token_id": 577,
+    "max_length": 10,
+    "end_of_text_token": "<|end_of_text|>",
+    "end_of_text_token_id": 578
 }