feat: update model to support Chinese

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/mnt/huggingface/models/Qwen1.5-0.5B/",
+  "_name_or_path": "/mnt/data102_d2/huggingface/models/Qwen1.5-0.5B",
   "architectures": [
     "Qwen2ForCausalLM"
   ],
@@ -18,10 +18,10 @@
   "num_key_value_heads": 16,
   "rms_norm_eps": 1e-06,
   "rope_theta": 1000000.0,
-  "sliding_window": 32768,
+  "sliding_window": null,
   "tie_word_embeddings": true,
   "torch_dtype": "float16",
-  "transformers_version": "4.39.3",
+  "transformers_version": "4.43.3",
   "use_cache": true,
   "use_sliding_window": false,
   "vocab_size": 151936

generation_config.json

@@ -2,5 +2,5 @@
   "bos_token_id": 151643,
   "eos_token_id": 151643,
   "max_new_tokens": 2048,
-  "transformers_version": "4.39.3"
+  "transformers_version": "4.43.3"
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5ffbc926e1d343cec40a19e69df3bccd41ff39280bcfc4a05fb35c2c4879d18
-size 928069010
+oid sha256:0f83725c1a17de9c12ca2e102d5e29c0231200c3f660b58eb2db363680f5049d
+size 928073042

tokenizer.json

@@ -73,6 +73,7 @@
     "end_of_word_suffix": "",
     "fuse_unk": false,
     "byte_fallback": false,
+    "ignore_merges": false,
     "vocab": {
       "!": 0,
       "\"": 1,