Update README.md

README.md

@@ -28,7 +28,23 @@ We conducted continued pretraining using Japanese data on the English language m
 ## Usage
 
 ```python
-TODO
+from transformers import AutoModelForCausalLM, AutoTokenizer
+tokenizer = AutoTokenizer.from_pretrained("stabilityai/japanese-mistral-7b-v0.1-base")
+model = AutoModelForCausalLM.from_pretrained(
+  "stabilityai/japanese-mistral-7b-v0.1-base",
+  trust_remote_code=True,
+  torch_dtype="auto",
+)
+model.cuda()
+inputs = tokenizer("AI で科学研究を加速するには、", return_tensors="pt").to("cuda")
+tokens = model.generate(
+  **inputs,
+  max_new_tokens=64,
+  temperature=0.75,
+  top_p=0.95,
+  do_sample=True,
+)
+print(tokenizer.decode(tokens[0], skip_special_tokens=True))
 ```
 
 ## Model Details