Show HN: LLM post-training to speak like GenZ, costing less than a cup of coffee
A post-trained language model that responds in GenZ slang, built on top of Qwen2.5-0.5B-Instruct using SFT followed by GRPO reinforcement learning. Total training cost was under $2 and took 3-4 hours on Google Colab. Synthetic data was used, and while both SFT and SFT+RL models produce similar outputs, RL training may slightly increase keyword occurrence.
Show HN: LLM post-training to speak like GenZ, costing less than a cup of coffee
Report an error
The correction channel is not available yet. You can copy the article reference below for later.
Correction instructionsSummary only
Under the source content permission, this page provides an editorial summary, key points, and a link to the original article.