fine-tuning

# Fine-Tuning

Adapt LLMs to specific tasks and domains efficiently.

## Quick Start

### LoRA Fine-Tuning with PEFT
```python
from transformers import AutoModelForCausalLM, AutoTokenizer, TrainingArguments
from peft import LoraConfig, get_peft_model, TaskType
from datasets import load_dataset
from trl import SFTTrainer

# Load base model
model_name = "meta-llama/Llama-2-7b-hf"
model = AutoModelForCausalLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
tokenizer.pad_token = tokenizer.eos_token

# Configure LoRA
lora_config = LoraConfig(
    r=16,                          # Rank
    lora_alpha=32,                 # Alpha scaling
    target_modules=["q_proj", "v_proj", "k_proj", "o_proj"],
    lora_dropout=0.05,
    bias="none",
    task_type=TaskType.CAUSAL_LM
)

# Apply LoRA
model = get_peft_model(model, lora_config)
model.print_trainable_parameters()
# trainable params: 4,194,304 || all params: 6,742,609,920 || trainable%: 0.06%

# Training arguments
training_args = TrainingArguments(
    output_dir="./output",
    num_train_epochs=3,
    per_device_train_batch_size=4,
    gradient_accumulation_steps=4,
    learning_rate=2e-4,
    warmup_ratio=0.03,
    logging_steps=10,
    save_strategy="epoch"
)

# Train
trainer = SFTTrainer(
    model=model,
    args=training_args,
    train_dataset=dataset,
    tokenizer=tokenizer,
    max_seq_length=512
)
trainer.train()
```

### QLoRA (4-bit Quantized LoRA)
```python
from transformers import BitsAndBytesConfig
import torch

# Quantization config
bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.bfloat16,
    bnb_4bit_use_double_quant=True
)

# Load quantized model
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    quantization_config=bnb_config,
    device_map="auto"
)

# Apply LoRA on top of quantized model
model = get_peft_model(model, lora_config)
```

## Dataset Preparation

### Instruction Dataset Form