Edit model card

StoryLlama is an expert language model fine-tuned from the Llama-2-13B-chat model, focused on generating entertaining, imaginative children's picture book stories. By leveraging a proprietary high-quality picture book story dataset for Lora fine-tuning, StoryLlama retains the powerful language understanding and generation capabilities of the original model while specifically enhancing its ability to tell lively and interesting stories.

Currently, there are three versions of StoryLlama: v1, v2, v3, and v4. They have all been trained using the same picture book dataset, but with variations in the training parameters, which may result in subtle differences in performance when generating stories. Regardless of the version, StoryLlama can automatically create engaging and meaningful original picture book stories based on a sentence provided by the user.

The potential applications for StoryLlama are vast and can assist in the creation of children’s literature, the automatic generation of picture books, and support early childhood education, among others. Furthermore, with continued optimization and expansion of the training data, StoryLlama also has the potential to grow into an all-purpose story creation assistant, providing inspiration and material for various types of creative writing.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Examples
Unable to determine this model's library. Check the docs .

Dataset used to train Aiwensile2/StoryLlama