Update README.md

README.md

@@ -77,11 +77,15 @@ European Championship => competitions
 
 If you want to use flash attention or increase sequence length, please, check the following code:
 ```python
+from gliner import GLiNER
+import torch
+
 model = GLiNER.from_pretrained("knowledgator/gliner-llama-1B-v1.0",
                                 _attn_implementation = 'flash_attention_2',
-                                                max_len = 2048).to('cuda:0')
+                                                max_len = 2048).to('cuda:0', dtype=torch.float16)
 ```
 
+
 ### Benchmarks
 Below you can see the table with benchmarking results on various named entity recognition datasets: