Update README.md

README.md

@@ -28,6 +28,23 @@ Mistral-7B-v0.1 is a transformer model, with the following architecture choices:
 - Sliding-Window Attention
 - Byte-fallback BPE tokenizer
 
+
+#### Running the model on a GPU
+
+
+```python
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+tokenizer = AutoTokenizer.from_pretrained("Hemanth-thunder/Tamil-Mistral-7B-v0.1")
+model = AutoModelForCausalLM.from_pretrained("Hemanth-thunder/Tamil-Mistral-7B-v0.1")
+
+input_text = "ஒரு கிராமத்தில் பண்ணையார் ஒருவர் வாழ்ந்து வந்தார்."
+input_ids = tokenizer(input_text, return_tensors="pt")
+
+outputs = model.generate(**input_ids)
+print(tokenizer.decode(outputs[0]))
+```
+
 ## Troubleshooting
 
 - If you see the following error: