inference_example.py

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

# Load model and tokenizer
model = AutoModelForCausalLM.from_pretrained("PATH_TO_MODEL", device_map="auto", trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained("PATH_TO_MODEL", trust_remote_code=True)

# For PEFT models (if not merged)
# from peft import PeftModel
# model = PeftModel.from_pretrained(model, "PATH_TO_MODEL")

# Prepare input
system_prompt = "You are embodying the following persona: [PERSONA]"
user_prompt = "Survey Question: [QUESTION]\n\nPlease provide your honest and detailed response to this question."

messages = [
    {"role": "system", "content": system_prompt},
    {"role": "user", "content": user_prompt}
]

# Tokenize input
input_text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to(model.device)

# Generate response
with torch.no_grad():
    outputs = model.generate(
        input_ids=input_ids,
        max_new_tokens=256,
        temperature=0.7,
        top_p=0.9,
        do_sample=True
    )

# Decode response
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)