├── .gitignore ├── README.md ├── images └── oreilly.png ├── notebooks ├── .DS_Store ├── prompt_injection.ipynb ├── rl_flan_t5_summaries.ipynb ├── rlaif.ipynb ├── sawyer_1_instruction_ft.ipynb ├── sawyer_2_train_reward_model.ipynb ├── sawyer_3_rl.ipynb └── sawyer_4_use_sawyer.ipynb ├── requirements.txt └── streamlit └── chat ├── README.md ├── app.py └── requirements.txt /.gitignore: -------------------------------------------------------------------------------- 1 | .ipynb_checkpoints/ 2 | .DS_Store 3 | results/ 4 | wandb/ 5 | *.streamlit/ -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/sinanuozdemir/oreilly-llm-alignment/HEAD/README.md -------------------------------------------------------------------------------- /images/oreilly.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/sinanuozdemir/oreilly-llm-alignment/HEAD/images/oreilly.png -------------------------------------------------------------------------------- /notebooks/.DS_Store: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/sinanuozdemir/oreilly-llm-alignment/HEAD/notebooks/.DS_Store -------------------------------------------------------------------------------- /notebooks/prompt_injection.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/sinanuozdemir/oreilly-llm-alignment/HEAD/notebooks/prompt_injection.ipynb -------------------------------------------------------------------------------- /notebooks/rl_flan_t5_summaries.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/sinanuozdemir/oreilly-llm-alignment/HEAD/notebooks/rl_flan_t5_summaries.ipynb -------------------------------------------------------------------------------- /notebooks/rlaif.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/sinanuozdemir/oreilly-llm-alignment/HEAD/notebooks/rlaif.ipynb -------------------------------------------------------------------------------- /notebooks/sawyer_1_instruction_ft.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/sinanuozdemir/oreilly-llm-alignment/HEAD/notebooks/sawyer_1_instruction_ft.ipynb -------------------------------------------------------------------------------- /notebooks/sawyer_2_train_reward_model.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/sinanuozdemir/oreilly-llm-alignment/HEAD/notebooks/sawyer_2_train_reward_model.ipynb -------------------------------------------------------------------------------- /notebooks/sawyer_3_rl.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/sinanuozdemir/oreilly-llm-alignment/HEAD/notebooks/sawyer_3_rl.ipynb -------------------------------------------------------------------------------- /notebooks/sawyer_4_use_sawyer.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/sinanuozdemir/oreilly-llm-alignment/HEAD/notebooks/sawyer_4_use_sawyer.ipynb -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- 1 | transformers 2 | datasets 3 | streamlit -------------------------------------------------------------------------------- /streamlit/chat/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/sinanuozdemir/oreilly-llm-alignment/HEAD/streamlit/chat/README.md -------------------------------------------------------------------------------- /streamlit/chat/app.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/sinanuozdemir/oreilly-llm-alignment/HEAD/streamlit/chat/app.py -------------------------------------------------------------------------------- /streamlit/chat/requirements.txt: -------------------------------------------------------------------------------- 1 | bitsandbytes 2 | accelerate 3 | peft --------------------------------------------------------------------------------