AWS Machine Learning Blog· Mijanur Palash·· 26 d ago
Fine-tuning GPT-OSS-20B for multilingual reasoning on Amazon SageMaker HyperPod
Fine-tuning GPT-OSS-20B for multilingual reasoning on Amazon SageMaker HyperPod
AI summary
Learn how we fine-tuned GPT-OSS-20B for multilingual reasoning on Amazon SageMaker HyperPod, and how to decide between supervised fine-tuning (SFT) and reinforcement learning with GRPO. SFT took reasoning-language accuracy from 22% to 100%, and GRPO added measurable gains where headroom existed.
Selection record
Threshold 60Official, first-handFirst 34Second 38
Not admittedSum of both 72 < twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed：GPT-OSS-20B微调与GRPO训练技术详解
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: AWS Machine Learning Blog · aws.amazon.com