src/test.py

# coding: utf8
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_path = "vinai/PhoGPT-4B"  

device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
print(device)

model = AutoModelForCausalLM.from_pretrained(model_path)
tokenizer = AutoTokenizer.from_pretrained(model_path)

inputs = tokenizer('### Câu hỏi: Viết bài văn nghị luận xã hội về an toàn giao thông \n### Trả lời:', return_tensors='pt').to(device)
print(inputs)

outputs = model.generate(  
    inputs=inputs["input_ids"].to(device),  
    attention_mask=inputs["attention_mask"].to(device),  
    do_sample=True,  
    temperature=1.0,  
    top_k=50,  
    top_p=0.9,  
    max_new_tokens=1024,  
    eos_token_id=tokenizer.eos_token_id,  
    pad_token_id=tokenizer.pad_token_id  
)  

response = tokenizer.decode(outputs[0])
print(response)