Updated
Updated · O'Reilly Media · Sep 21
OpenAI Details ChatGPT’s 3-Stage Post-Training Pipeline With 13,000 Prompts and RLHF
Updated
Updated · O'Reilly Media · Sep 21

OpenAI Details ChatGPT’s 3-Stage Post-Training Pipeline With 13,000 Prompts and RLHF

3 articles · Updated · O'Reilly Media · Sep 21

Summary

  • OpenAI said ChatGPT’s original post-training stack had three layers: supervised fine-tuning, a reward model, and RLHF—turning base GPT-3 from erratic text generation into a conversational assistant.
  • About 40 labelers wrote ideal answers for roughly 13,000 prompts in the SFT stage, while OpenAI mixed in 10% pretraining data and later used 31,000 user prompts to drive RL rollouts.
  • A 6B-parameter reward model learned from rankings of 4–9 responses per prompt, yielding about 200,000 to 1.2 million pairwise comparisons to score outputs humans preferred.
  • PPO-based RLHF then optimized the policy against that reward model while adding a KL-divergence penalty to keep it close to the SFT model and reduce reward hacking.
  • OpenAI said the approach improved helpfulness, safety, tool use and reasoning; in cited tests, even a 1.3B RLHF model often beat a 175B SFT model, though gains were smaller with held-out evaluators.

Insights

Could the secret to next-generation AI lie not in massive data centers, but in the hidden biases of just 40 human labelers?
If AI models learn to please humans rather than seek truth, are we merely engineering the ultimate digital sycophants?
As AI reasoning demands explosive computing power, will the staggering cost of machine thought outpace the actual value of its answers?