Training a Small Model on Your Own Popular Tweets
Melvin Vivas · X post · 2026-09-14 · Open on X
Topics: Fine-tuning & Model Customization, Portfolio Projects · Level: intermediate
Summary
The creator is preparing data to train a small model on his most popular tweets, which suggests a personal style-tuning project. It is a short status update with no details on method.
Key points
- Your own high-performing content can be a fine-tuning dataset.
- Pick the popular tweets as the quality signal for training examples.
- Data prep comes first: collect, filter and format the examples into training pairs.
Try this
- Fine-tune a small model on your most popular social posts so it writes in your style.
More in Fine-tuning & Model Customization
- First LoRA Run on Qwen3.5-2B with Codex as Training Companion
- Using Codex to Generate Synthetic Training Pairs for Fine-tuning Qwen2.5
- Unsloth Studio: Fine-tuning Without Colab Notebooks
- Model Distillation Explained: A Teacher Model Trains a Smaller Student
- Idea: Fine-Tune a Small Model on Your Best Tweets
- Hugging Face Training Agents series: post-training coding agents