Menyiapkan himpunan data preferensi

Dalam latihan ini, Anda akan bekerja dengan himpunan data yang berisi umpan balik manusia dalam bentuk keluaran "chosen" dan "rejected". Tugas Anda adalah mengekstrak prompt dari kolom "chosen" dan menyiapkan data untuk melatih model penghargaan.

Fungsi load_dataset dari datasets telah diimpor sebelumnya

Latihan ini adalah bagian dari kursus

Reinforcement Learning from Human Feedback (RLHF)

Petunjuk latihan

Muat himpunan data trl-internal-testing/hh-rlhf-helpful-base-trl-style dari Hugging Face.
Tulis fungsi yang mengekstrak prompt dari field 'content', dengan asumsi prompt berada pada indeks 0 dari masukan ke fungsi.
Terapkan fungsi yang mengekstrak prompt pada subset himpunan data 'chosen'.

Latihan interaktif praktis

Cobalah latihan ini dengan menyelesaikan kode contoh berikut.

# Load the dataset
preference_data = ____

# Define a function to extract the prompt
def extract_prompt(text):
    ____
    return prompt

# Apply the function to the dataset 
preference_data_with_prompt = ____(
    lambda sample: {**sample, 'prompt': ____(sample['chosen'])}
)

sample = preference_data_with_prompt.select(range(1))
print(sample['prompt'])

Edit dan Jalankan Kode

Latihan ini adalah bagian dari kursus

Reinforcement Learning from Human Feedback (RLHF)

SkillTag.level.advancedSkillTag.label

4.8+

Mulai Kursus Gratis

This chapter introduces the basics of Reinforcement Learning with Human Feedback (RLHF), a technique that uses human input to help AI models learn more effectively. Get started with RLHF by understanding how it differs from traditional reinforcement learning and why human feedback can enhance AI performance in various domains.

Exercise 1: Pengantar RLHF Exercise 2: Generasi teks dengan RLHF Exercise 3: Mengklasifikasikan teks yang dihasilkan untuk RLHF Exercise 4: RL vs. RLHF Exercise 5: Menjelajahi LLM pralatih Exercise 6: Tokenisasi himpunan data teks Exercise 7: Fine-tuning untuk klasifikasi ulasan Exercise 8: Menyiapkan data untuk RLHF Exercise 9: Menyiapkan himpunan data preferensi

Latihan Saat Ini

Exercise 10: Mengekstrak prompt

Discover how to set up systems for gathering human feedback in this Chapter. Learn best practices for collecting high-quality data, from pairwise comparisons to uncertainty sampling, and explore strategies for enhancing your data collection.

Exercise 1: Methods for high-quality feedback gathering Exercise 2: Understanding comparison and rating in RLHF Exercise 3: Comparing slogans for a gym campaign Exercise 4: Measuring feedback quality and relevance Exercise 5: Low confidence Exercise 6: K-means for feedback clustering Exercise 7: Active learning Exercise 8: Implementing an active learning pipeline Exercise 9: Active learning loop

In this Chapter, you'll get into the core of Reinforcement Learning from Human Feedback training. This includes exploring fine-tuning with PPO, techniques to train efficiently, and handling potential divergences from your metrics' objectives.

Exercise 1: Reward models explored Exercise 2: Initializing the reward Exercise 3: Setting up the reward trainer Exercise 4: Training with PPO Exercise 5: Initialize the PPO trainer Exercise 6: PPO fine-tuning Exercise 7: Efficient fine-tuning in RLHF Exercise 8: Prepare for 8-bit Training Exercise 9: Train with LoRA

Explore key techniques for assessing and improving model performance in this last Chapter of Reinforcement Learning from Human Feedback (RLHF): from fine-tuning metrics to incorporating diverse feedback sources, you'll be provided with a comprehensive toolkit to refine your models effectively.

Exercise 1: Model metrics and adjustments Exercise 2: Mitigating negative KL divergence Exercise 3: Checking the reward model Exercise 4: Incorporating diverse feedback sources Exercise 5: Majority voting on multiple data sources Exercise 6: Unreliable data source identification Exercise 7: Evaluating RLHF models Exercise 8: Interpreting curves Exercise 9: Evaluating RLHF with metrics Exercise 10: Wrapping up your RLHF journey