Kite
🎉 You are looking at Kite 7, which has better data, tokenization, and a longer context window!
Kite is a small, trained, 15 million parameter language model.
Training
It was trained on the first shard of semran1/cosmopedia-v2-subset, using 1 epoch, 8 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
- -