Create tokenizer.json

tokenizer.json

@@ -0,0 +1,12 @@
+{
+  "model": {
+    "type": "BPE",
+    "vocab_size": 32000
+  },
+  "normalizer": {
+    "type": "Lowercase"
+  },
+  "pre_tokenizer": {
+    "type": "Whitespace"
+  }
+}