Update README.md

README.md

@@ -1,29 +1,60 @@
-### Run Huggingface RWKV World Model
+### Run Huggingface RWKV5 World Model
 
-> This model is developed and converted through https://github.com/BBuf/RWKV-World-HF-Tokenizer. If you have any issues, you can raise them in this project. You're also welcome to star it to follow the subsequent development progress.
 
 #### CPU
 
 ```python
+import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-model = AutoModelForCausalLM.from_pretrained("RWKV/rwkv-4-world-7b")
+def generate_prompt(instruction, input=""):
+    instruction = instruction.strip().replace('\r\n','\n').replace('\n\n','\n')
+    input = input.strip().replace('\r\n','\n').replace('\n\n','\n')
+    if input:
+        return f"""Instruction: {instruction}
+
+Input: {input}
+
+Response:"""
+    else:
+        return f"""User: hi
+
+Assistant: Hi. I am your assistant and I will provide expert full response in full details. Please feel free to ask any question and I will always answer it.
+
+User: {instruction}
+
+Assistant:"""
+
+
+model = AutoModelForCausalLM.from_pretrained("RWKV/rwkv-4-world-7b", trust_remote_code=True).to(torch.float32)
 tokenizer = AutoTokenizer.from_pretrained("RWKV/rwkv-4-world-7b", trust_remote_code=True)
 
-text = "\nIn a shocking finding, scientist discovered a herd of dragons living in a remote, previously unexplored valley, in Tibet. Even more surprising to the researchers was the fact that the dragons spoke perfect Chinese."
-prompt = f'Question: {text.strip()}\n\nAnswer:'
+text = "请介绍北京的旅游景点"
+prompt = generate_prompt(text)
 
 inputs = tokenizer(prompt, return_tensors="pt")
-output = model.generate(inputs["input_ids"], max_new_tokens=256)
+output = model.generate(inputs["input_ids"], max_new_tokens=333, do_sample=True, temperature=1.0, top_p=0.3, top_k=0, )
 print(tokenizer.decode(output[0].tolist(), skip_special_tokens=True))
 ```
 
 output:
 
 ```shell
-Question: In a shocking finding, scientist discovered a herd of dragons living in a remote, previously unexplored valley, in Tibet. Even more surprising to the researchers was the fact that the dragons spoke perfect Chinese.
+User: hi
+
+Assistant: Hi. I am your assistant and I will provide expert full response in full details. Please feel free to ask any question and I will always answer it.
+
+User: 请介绍北京的旅游景点
 
-Answer: 科学家在一个未曾探索过的山谷中发现了一群能说流利中文的龙。科学家惊讶地发现，这些龙是在一个完全未被探索的地区生活的。
+Assistant: 北京是中国的首都，拥有众多的旅游景点，以下是其中一些著名的景点：
+1. 故宫：位于北京市中心，是明清两代的皇宫，内有大量的文物和艺术品。
+2. 天安门广场：是中国最著名的广场之一，是中国人民政治协商会议的旧址，也是中国人民政治协商会议的中心。
+3. 颐和园：是中国古代皇家园林之一，有着悠久的历史和丰富的文化内涵。
+4. 长城：是中国古代的一道长城，全长约万里，是中国最著名的旅游景点之一。
+5. 北京大学：是中国著名的高等教育机构之一，有着悠久的历史和丰富的文化内涵。
+6. 北京动物园：是中国最大的动物园之一，有着丰富的动物资源和丰富的文化内涵。
+7. 故宫博物院：是中国最著名的博物馆之一，收藏了大量的文物和艺术品，是中国最重要的文化遗产之一。
+8. 天坛：是中国古代皇家
 ```
 
 #### GPU
@@ -32,22 +63,45 @@ Answer: 科学家在一个未曾探索过的山谷中发现了一群能说流利
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-model = AutoModelForCausalLM.from_pretrained("RWKV/rwkv-4-world-7b", torch_dtype=torch.float16).to(0)
+def generate_prompt(instruction, input=""):
+    instruction = instruction.strip().replace('\r\n','\n').replace('\n\n','\n')
+    input = input.strip().replace('\r\n','\n').replace('\n\n','\n')
+    if input:
+        return f"""Instruction: {instruction}
+
+Input: {input}
+
+Response:"""
+    else:
+        return f"""User: hi
+
+Assistant: Hi. I am your assistant and I will provide expert full response in full details. Please feel free to ask any question and I will always answer it.
+
+User: {instruction}
+
+Assistant:"""
+
+
+model = AutoModelForCausalLM.from_pretrained("RWKV/rwkv-4-world-7b", trust_remote_code=True, torch_dtype=torch.float16).to(0)
 tokenizer = AutoTokenizer.from_pretrained("RWKV/rwkv-4-world-7b", trust_remote_code=True)
 
-text = "你叫什么名字？"
-prompt = f'Question: {text.strip()}\n\nAnswer:'
+text = "乌兰察布"
+prompt = generate_prompt(text)
 
 inputs = tokenizer(prompt, return_tensors="pt").to(0)
-output = model.generate(inputs["input_ids"], max_new_tokens=40)
+output = model.generate(inputs["input_ids"], max_new_tokens=128, do_sample=True, temperature=1.0, top_p=0.3, top_k=0, )
 print(tokenizer.decode(output[0].tolist(), skip_special_tokens=True))
 ```
 
 output:
 
 ```shell
-Question: 你叫什么名字？
+User: hi
+
+Assistant: Hi. I am your assistant and I will provide expert full response in full details. Please feel free to ask any question and I will always answer it.
+
+User: 乌兰察布
 
-Answer: 我是一个人工智能语言模型，没有具体的名字。
+Assistant: 乌兰察布市是中国新疆维吾尔自治区的一个地级市，位于新疆维吾尔自治区西南部，毗邻青海省。乌兰察布市是新疆维吾尔自治区的重要城市之一，也是新疆维吾尔自治区的第二大城市。乌兰察布市是新疆的重要经济中心之一，拥有丰富的自然资源和人口密度，是新疆的重要交通枢纽和商
 ```