name: Fine-Tuning Assistant
slug: fine-tuning-assistant
description: Guide model fine-tuning processes for customized AI performance
category: ai-ml
complexity: advanced
version: "1.0.0"
author: "ID8Labs"
triggers:
- "fine-tune model"
- "fine-tuning"
- "customize LLM"
- "train custom model"
- "adapt model"
tags:
- fine-tuning
- training
- customization
- LLM
- machine-learning

Fine-Tuning Assistant

The Fine-Tuning Assistant skill guides you through the process of adapting pre-trained models to your specific use case. Fine-tuning can dramatically improve model performance on specialized tasks, teach models your preferred style, and add capabilities that prompting alone cannot achieve.

This skill covers when to fine-tune versus prompt engineer, preparing training data, selecting base models, configuring training parameters, evaluating results, and deploying fine-tuned models. It applies modern techniques including LoRA, QLoRA, and instruction tuning to make fine-tuning practical and cost-effective.

Whether you are fine-tuning GPT models via API, running local training with open-source models, or using platforms like Hugging Face, this skill ensures you approach fine-tuning strategically and effectively.

Core Workflows

Workflow 1: Decide Whether to Fine-Tune

1. Assess the problem:
2. Can prompting achieve the goal?
3. Is the task format or style consistent?
4. Do you have quality training data?
5. Is this worth the investment?
6. Compare approaches:
   | Approach | When to Use | Investment |
   |----------|-------------|------------|
   | Better prompts | First attempt, variable tasks | Low |
   | Few-shot examples | Consistent format, limited data | Low |
   | RAG | Knowledge-intensive, dynamic data | Medium |
   | Fine-tuning | Consistent style, specialized task | High |
7. Evaluate requirements:
8. Minimum 100-1000 quality examples
9. Clear evaluation criteria
10. Budget for training and hosting
11. Decision: Fine-tune only if prompting/RAG insufficient

Workflow 2: Prepare Fine-Tuning Dataset

1. Collect training examples:
2. Representative of target use case
3. High quality (no errors in outputs)
4. Diverse coverage of task variations
5. Format for training:
   jsonl {"messages": [ {"role": "system", "content": "You are a helpful assistant..."}, {"role": "user", "content": "User input here"}, {"role": "assistant", "content": "Ideal response here"} ]}
6. Quality assurance:
7. Review sample of examples manually
8. Check for consistency in style/format
9. Remove duplicates and low-quality entries
10. Split train/validation/test sets
11. Validate dataset format

Workflow 3: Execute Fine-Tuning

1. Select base model:
2. Consider size vs capability tradeoff
3. Match model to task complexity
4. Check licensing for your use case
5. Configure training:
   ```python
   # OpenAI fine-tuning
   training_config = {
   "model": "gpt-4o-mini-2024-07-18",
   "training_file": "file-xxx",
   "hyperparameters": {
   "n_epochs": 3,
   "batch_size": "auto",
   "learning_rate_multiplier": "auto"
   }
   }

# LoRA fine-tuning (local)
lora_config = {
"r": 16, # Rank
"lora_alpha": 32,
"lora_dropout": 0.05,
"target_modules": ["q_proj", "v_proj"]
}
```
3. Monitor training:
- Watch loss curves
- Check for overfitting
- Validate on held-out set
4. Evaluate results:
- Compare to baseline model
- Test on diverse inputs
- Check for regressions

Quick Reference

Best Practices

• Start with Prompting: Fine-tuning is expensive; exhaust cheaper options first
• Can better prompts achieve 80% of the goal?
• Try few-shot examples in the prompt

• Consider RAG for knowledge tasks

• Quality Over Quantity: 100 excellent examples beat 10,000 mediocre ones

• Each example should be a gold standard
• Better to have humans verify examples

• Remove anything you wouldn't want the model to learn

• Match Format to Use Case: Training examples should mirror real usage

• Same prompt structure as production
• Realistic input variations

• Cover edge cases explicitly

• Don't Over-Train: More epochs isn't always better

• Watch validation loss for overfitting
• Start with 1-3 epochs

• Early stopping when validation plateaus

• Evaluate Properly: Training loss isn't the goal

• Use held-out test set
• Compare to baseline on same tests
• Check for capability regressions

• Test on edge cases explicitly

• Version Everything: Fine-tuning is iterative

• Version your training data
• Track experiment configurations
• Document what worked and what didn't

Advanced Techniques

LoRA (Low-Rank Adaptation)

Efficient fine-tuning for large models:

from peft import LoraConfig, get_peft_model

lora_config = LoraConfig(
    r=16,                           # Rank of update matrices
    lora_alpha=32,                  # Scaling factor
    target_modules=["q_proj", "v_proj", "k_proj", "o_proj"],
    lora_dropout=0.05,
    bias="none",
    task_type="CAUSAL_LM"
)

# Apply LoRA to base model
model = get_peft_model(base_model, lora_config)

# Only ~0.1% of parameters are trainable
trainable_params = sum(p.numel() for p in model.parameters() if p.requires_grad)

QLoRA (Quantized LoRA)

Fine-tune large models on consumer hardware:

from transformers import BitsAndBytesConfig

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 model in 4-bit
model = AutoModelForCausalLM.from_pretrained(
    "meta-llama/Llama-2-7b-hf",
    quantization_config=bnb_config
)

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

Instruction Tuning Dataset Creation

Convert raw data to instruction format:

def create_instruction_example(raw_data):
    return {
        "messages": [
            {
                "role": "system",
                "content": "You are a customer service agent for TechCorp..."
            },
            {
                "role": "user",
                "content": f"Customer inquiry: {raw_data['inquiry']}"
            },
            {
                "role": "assistant",
                "content": raw_data['ideal_response']
            }
        ]
    }

# Apply to dataset
instruction_dataset = [create_instruction_example(d) for d in raw_dataset]

Evaluation Framework

Comprehensive assessment of fine-tuned models:

def evaluate_fine_tuned_model(model, test_set, baseline_model=None):
    results = {
        "task_accuracy": [],
        "format_compliance": [],
        "style_match": [],
        "regression_check": []
    }

    for example in test_set:
        output = model.generate(example.input)

        # Task-specific accuracy
        results["task_accuracy"].append(
            check_correctness(output, example.expected)
        )

        # Format compliance
        results["format_compliance"].append(
            matches_expected_format(output)
        )

        # Style matching (for style transfer tasks)
        results["style_match"].append(
            style_similarity(output, example.expected)
        )

        # Regression on general capabilities
        if baseline_model:
            results["regression_check"].append(
                compare_general_capability(model, baseline_model, example)
            )

    return {k: np.mean(v) for k, v in results.items()}

Curriculum Learning

Order training data by difficulty:

def create_curriculum(dataset):
    # Score examples by complexity
    scored = [(score_complexity(ex), ex) for ex in dataset]
    scored.sort(key=lambda x: x[0])

    # Create epochs with increasing difficulty
    n = len(scored)
    curriculum = {
        "epoch_1": [ex for _, ex in scored[:n//3]],           # Easy
        "epoch_2": [ex for _, ex in scored[:2*n//3]],         # Easy + Medium
        "epoch_3": [ex for _, ex in scored],                   # All
    }
    return curriculum

Common Pitfalls to Avoid

• Fine-tuning when better prompting would suffice
• Using low-quality or inconsistent training examples
• Not holding out a proper test set
• Training for too many epochs (overfitting)
• Ignoring capability regressions from fine-tuning
• Not versioning training data and configurations
• Expecting fine-tuning to add factual knowledge (use RAG instead)
• Fine-tuning on data that doesn't match production use