Update README.md

README.md

@@ -133,7 +133,7 @@ attn_implementation = "eager"  # Or "flash_attention_2"
 model = SentenceTransformer(
     "nvidia/llama-embed-nemotron-8b",
     trust_remote_code=True,
-    model_kwargs={"attn_implementation": attn_implementation, "torch_dtype": "float32"},
+    model_kwargs={"attn_implementation": attn_implementation, "torch_dtype": "bfloat16"},
     tokenizer_kwargs={"padding_side": "left"},
 )
 
@@ -152,7 +152,7 @@ document_embeddings = model.encode_document(documents)
 scores = (query_embeddings @ document_embeddings.T)
 
 print(scores.tolist())
-# [[0.37646484375, 0.057891845703125]]
+# [[0.3770667314529419, 0.05808388814330101]]
 ```
 
 Or using Hugging Face Transformers like here: