Base model

meta-llama/Llama-3.2-3B-Instruct

Description

QLoRA adapter — Llama-3.2-3B-Instruct fine-tuned on NLNG LNG domain Q&A

Training data covers NLNG's internal domain: gas supply chain, cargo scheduling, shipping logistics, LNG production, HSE compliance, thermodynamics, and commercial operations.

Usage

python
Copy code
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
import torch

bnb_cfg   = BitsAndBytesConfig(load_in_4bit=True, bnb_4bit_quant_type="nf4",
                                bnb_4bit_compute_dtype=torch.bfloat16)
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-3.2-3B-Instruct")
base      = AutoModelForCausalLM.from_pretrained(
                "meta-llama/Llama-3.2-3B-Instruct", quantization_config=bnb_cfg, device_map="auto")
model     = PeftModel.from_pretrained(base, "Bluechip-AI/nlng-slm-v3-llama")

prompt = "What are the key safety considerations for LNG cargo loading operations?"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
output = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(output[0], skip_special_tokens=True))

Training details