llama-recipes
llama-recipes copied to clipboard
Scripts for fine-tuning Meta Llama with composable FSDP & PEFT methods to cover single/multi-node GPUs. Supports default & custom datasets for applications such as summarization and Q&A. Supporting a...
Marine
### 🚀 The feature, motivation and pitch marineman.eu ### Alternatives Navy Seal ### Additional context Military Program
# What does this PR do? 1. fixed "INPUT_FILE" path 2. fix create_word_bounded_chunks was called before read in the "text" variables 3. processed_text not defined ## Before submitting - [yes]...
# What does this PR do? - fix "INPUT_FILE" path (the initial output of extracted_text.txt and the input path doesn't match, therefore not able to read in here) - missing...
# What does this PR do? This pull request (PR) adds a new notebook tutorial that showcases the latest capabilities of LLaMA 3.2. ## Before submitting - [ ] This...
### System Info - ### Information - [ ] The official example scripts - [ ] My own modified scripts ### 🐛 Describe the bug Would be great to get...
# What does this PR do? Fixes # (issue) ## Feature/Issue validation/testing Please describe the tests that you ran to verify your changes and relevant result summary. Provide instructions so...
Added an open source implementation of NotebookLM that uses Together AI.
# What does this PR do? This PR adds an AI data analyst that uses Llama APIs or Ollama to analyze data and generate charts. It uses Llama to generate...
### 🚀 The feature, motivation and pitch Notice that in the original paper "[The Llama 3 Herd of Models](https://scontent.ftpe8-1.fna.fbcdn.net/v/t39.2365-6/463020162_522238820565582_8192401983671993921_n.pdf?_nc_cat=108&ccb=1-7&_nc_sid=3c67a6&_nc_ohc=7glwXFfGh-gQ7kNvgFOooyU&_nc_zt=14&_nc_ht=scontent.ftpe8-1.fna&_nc_gid=AMGP-tQbgYRcaDMcQVNGQgO&oh=00_AYBi-c4K56zBN43Y1yHg2uGDhwh3vUqUBJpYb_9cidTrbA&oe=672CF519)", section 7.5.2 on vision model SFT states that only the vision...
### System Info 8 gpu on A100 ### Information - [X] The official example scripts - [ ] My own modified scripts ### 🐛 Describe the bug on the finetuning.py...