SimpleBooks: Long-term dependency book dataset with simplified English vocabulary for word-level language modeling
Paper
•
1911.12391
•
Published
Model Description
Small pretrained LLM on the SimplyBooks dataset.
- Developed by: [Ivan Torriani]
Model Sources [optional]
- Repository: [More Information Needed]
- Paper [SimplyBooks Dataset]: [Small dataset of children's stories made for shorter training. ]
Uses
Be used as an educational model accessible by almost all computers. Directly, it can be used for text generation. Downstream, it can be fine tuned for writing short stories.
Bias, Risks, and Limitations
Model evaluations demonstrated progressive split between training loss and validation loss.