Kite
🎉 You are looking at Kite 6.3, which uses FineWeb-Edu, Qwen2, and an overhauled config!
Kite is a small, trained, 15 million parameter language model.
Training
It was trained on a 50K row subset of HuggingFaceFW/fineweb-edu, using 2 epochs, 32 batch size, 1e-3 learning rate, and the pika 5 tokenizer.
Limitations
Due to its size, the model is not suitable for production workloads.
- Downloads last month
- 379