READMD.md modify

README.md

@@ -1,3 +1,62 @@
 ---
 license: llama3
 ---
+
+- Foundation Model [Bllossom 8B](https://huggingface.co/MLP-KTLim/llama-3-Korean-Bllossom-8B)
+- datasets
+  -  [Koalpaca v1.1a](https://huggingface.co/datasets/beomi/KoAlpaca-v1.1a)
+  -  [jojo0217/korean_safe_conversation](https://huggingface.co/datasets/jojo0217/korean_safe_conversation)
+
+# Query
+```python
+
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
+
+BASE_MODEL = "sh2orc/llama-3-korean-8b"
+
+model = AutoModelForCausalLM.from_pretrained(BASE_MODEL,device_map="auto")
+
+tokenizer = AutoTokenizer.from_pretrained(BASE_MODEL)
+tokenizer.pad_token = tokenizer.eos_token
+tokenizer.padding_side = 'right'
+
+instruction = "한강에는 대교가 몇 개 있어?"
+
+pipe = pipeline("text-generation", 
+                model=model, 
+                tokenizer=tokenizer, 
+                max_new_tokens=1024)
+
+messages = [
+    {"role": "user", "content": instruction},
+]
+
+prompt = pipe.tokenizer.apply_chat_template(
+        messages, 
+        tokenize=False, 
+        add_generation_prompt=True
+)
+
+outputs = pipe(
+    prompt,
+    do_sample=True,
+    temperature=0.8,
+    top_k=10,
+    top_p=0.9,
+    add_special_tokens=True,
+    eos_token_id = [ 
+        pipe.tokenizer.eos_token_id,
+        pipe.tokenizer.convert_tokens_to_ids("<|eot_id|>")
+    ]
+)
+
+print(outputs[0]['generated_text'][len(prompt):])
+```
+
+# Result
+<pre>
+
+한강에는 총 8개의 다리(교)가 있습니다. 그 중 3개는 북쪽으로 향해 있고, 나머지 5개는 남쪽으로 향해 있습니다.
+
+</pre>