mshumer/gpt-llm-trainer
An automated pipeline that fine-tunes LLaMA 2 and GPT-3.5 models from task descriptions by generating datasets and handling the full training workflow.
Not currently ranked — collecting fresh signals.
star history
The repository provides a notebook-based system that abstracts away the complexity of training task-specific models. Users input a task description, and the system uses Claude 3 or GPT-4 to generate training prompts and responses, creates appropriate system messages, splits the data into training and validation sets, and fine-tunes either LLaMA 2 7B or GPT-3.5 models. The pipeline is designed to streamline the process of going from an idea to a deployed fine-tuned model.
Frequently asked
- What is mshumer/gpt-llm-trainer?
- An automated pipeline that fine-tunes LLaMA 2 and GPT-3.5 models from task descriptions by generating datasets and handling the full training workflow.
- Is gpt-llm-trainer open source?
- Yes — mshumer/gpt-llm-trainer is open source, released under the MIT license.
- What language is gpt-llm-trainer written in?
- mshumer/gpt-llm-trainer is primarily written in Jupyter Notebook.
- How popular is gpt-llm-trainer?
- mshumer/gpt-llm-trainer has 4.2k stars on GitHub.
- Where can I find gpt-llm-trainer?
- mshumer/gpt-llm-trainer is on GitHub at https://github.com/mshumer/gpt-llm-trainer.