LLM From Scratch (Educational Model)
I trained this model while following the Train LLM From Scratch guide. It is intended primarily for educational and demonstration purposes. While it is not suitable for production use, feel free to utilize it for your own presentations or to save yourself the time and resources of training a model from zero.
- Training Code: FareedKhan-dev/train-llm-from-scratch
- Training Code License: MIT License
- Objective: Educational demonstration of LLM development.
The model was trained on a subset (about 20GB) of the The Pile (Uncopyrighted).