from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

repo = "ozashu/mistral-3b-sft-prompt-hack-merged"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(repo, torch_dtype=torch.bfloat16, device_map="auto")

prompt = "こんにちは。自己紹介してください。"
inputs = tok(prompt, return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=128)
print(tok.decode(out[0], skip_special_tokens=True))