Ad

Privum

We build, secure and run your digital infrastructure.

Software, AI systems, penetration testing and secure cloud for companies.

Talk to Privum

Fine-tuning GPT-OSS-20B for multilingual reasoning on Amazon SageMaker HyperPod

23d

Age

Published
Collected

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.

Excerpt from the source

GPT-OSS-20B is an open-weight reasoning model that thinks in English. Ask it to reason in French, German, Spanish, or Italian and it acknowledges the instruction, then produces its chain-of-thought in English anyway. On a 50-example multilingual reasoning test, the base model reasoned in the correct target language only 22 percent of the time. That 22 percent was almost entirely the English examples (English 100 percent, French 0 percent, German 10 percent, Spanish 0 percent, Italian 0…

overfeed.news indexes and links. We publish a short excerpt — the full article stays at AWS Machine Learning Blog.

Log in to follow this source
Fine-tuning GPT-OSS-20B for multilingual reasoning on Amazon…