LLM Tuning Patterns

Expert patterns for LoRA, QLoRA, instruction dataset preparation, DPO, and evaluation.

Pattern 1: QLoRA Fine-Tuning with SFTTrainer

Complete QLoRA setup for instruction fine-tuning a 7B model on 24 GB VRAM.

from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
from peft import LoraConfig, get_peft_model, prepare_model_for_kbit_training
from trl import SFTTrainer
from datasets import load_dataset
import torch