初始化项目，由ModelHub XC社区提供模型

Model: ray0rf1re/Nano-Nano_v5.1
Source: Original Platform

tokenizer/README.md

@@ -0,0 +1,5 @@
+# nano-X1 Tokenizer
+
+ByteLevel BPE tokenizer based on EleutherAI/gpt-neox-20b with ChatML + `<think>`/`</think>` special tokens added.
+
+Load: `AutoTokenizer.from_pretrained('./nano_v5_tok_nanox1')`

tokenizer/chat_template.jinja

@@ -0,0 +1,6 @@
+{%- for message in messages %}<|im_start|>{{ message['role'] }}
+{%- if message['thinking'] is defined and message['thinking'] %}<think>{{ message['thinking'] }}</think>
+{%- endif %}{{ message['content'] }}<|im_end|>
+{%- endfor %}{%- if add_generation_prompt %}<|im_start|>assistant
+<think>
+{%- endif %}

tokenizer/tokenizer_config.json

@@ -0,0 +1,26 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>",
+    "<pad>",
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|system|>",
+    "<|user|>",
+    "<|assistant|>",
+    "<think>",
+    "</think>"
+  ],
+  "is_local": false,
+  "local_files_only": false,
+  "model_max_length": 2048,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}