Add huggingface login func. in example code

Author: opus24

examples/lpu_inference.py

@@ -1,4 +1,7 @@
 from vllm import LLM, SamplingParams
+from huggingface_hub._login import login
+
+login(token="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", add_to_git_credential=True)
 
 # Sample prompts.
 prompts = [
@@ -8,10 +11,8 @@
 sampling_params = SamplingParams(temperature=0.8, top_p=0.8, top_k=1, repetition_penalty=1.2, max_tokens=60)
 
 # Create an LLM.
-#llm = LLM(model="facebook/opt-1.3b", device="fpga", pipeline_parallel_size=2)
-#llm = LLM(model="meta-llama/Meta-Llama-3-8B", device="fpga", tensor_parallel_size=1)
-#llm = LLM(model="TinyLlama/TinyLlama-1.1B-Chat-v1.0", device="fpga", tensor_parallel_size=1)
-llm = LLM(model="TinyLlama/TinyLlama-1.1B-Chat-v1.0", device="fpga", num_lpu_devices=2, num_gpu_devices=1)
+#llm = LLM(model="TinyLlama/TinyLlama-1.1B-Chat-v1.0", device="fpga", num_lpu_devices=2, num_gpu_devices=1)
+llm = LLM(model="meta-llama/Meta-Llama-3-8B", device="fpga", num_lpu_devices=2, num_gpu_devices=0)
 
 # Generate texts from the prompts. The output is a list of RequestOutput objects
 # that contain the prompt, generated text, and other information.