├── .github ├── ISSUE_TEMPLATE │ ├── bug.yml │ └── feature-request.yml ├── PULL_REQUEST_TEMPLATE.md └── workflows │ └── spellcheck.yml ├── .gitignore ├── CODE_OF_CONDUCT.md ├── CONTRIBUTING.md ├── LICENSE ├── README.md ├── UPDATES.md ├── USE_POLICY.md ├── WhatsApp_Finetune.ipynb ├── configs ├── __init__.py ├── datasets.py ├── fsdp.py ├── peft.py └── training.py ├── docs ├── Dataset.md ├── FAQ.md ├── LLM_finetuning.md ├── images │ ├── feature-based_FN.png │ ├── featurebased_FN_.png │ └── full-param-FN.png ├── inference.md ├── multi_gpu.md └── single_gpu.md ├── ft_datasets ├── __init__.py ├── alpaca_dataset.py ├── grammar_dataset │ ├── __init__.py │ ├── grammar_dataset.py │ └── grammar_dataset_process.ipynb ├── samsum_dataset.py ├── utils.py └── whatsapp_dataset.py ├── inference ├── README.md ├── chat_completion.py ├── chat_utils.py ├── chats.json ├── checkpoint_converter_fsdp_hf.py ├── code-llama │ ├── code_completion_example.py │ ├── code_completion_prompt.txt │ ├── code_infilling_example.py │ └── code_infilling_prompt.txt ├── hf-text-generation-inference │ ├── README.md │ └── merge_lora_weights.py ├── inference.py ├── model_utils.py ├── safety_utils.py ├── samsum_prompt.txt └── vLLM_inference.py ├── llama_finetuning.py ├── model_checkpointing ├── __init__.py └── checkpoint_handler.py ├── multi_node.slurm ├── policies ├── __init__.py ├── activation_checkpointing_functions.py ├── anyprecision_optimizer.py ├── mixed_precision.py └── wrapping.py ├── prepare_dataset.py ├── preprocessing.py ├── requirements.txt ├── scripts ├── markdown_link_check_config.json ├── spellcheck.sh └── spellcheck_conf │ ├── spellcheck.yaml │ └── wordlist.txt └── utils ├── __init__.py ├── config_utils.py ├── dataset_utils.py ├── fsdp_utils.py ├── memory_utils.py └── train_utils.py /.github/ISSUE_TEMPLATE/bug.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/.github/ISSUE_TEMPLATE/bug.yml -------------------------------------------------------------------------------- /.github/ISSUE_TEMPLATE/feature-request.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/.github/ISSUE_TEMPLATE/feature-request.yml -------------------------------------------------------------------------------- /.github/PULL_REQUEST_TEMPLATE.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/.github/PULL_REQUEST_TEMPLATE.md -------------------------------------------------------------------------------- /.github/workflows/spellcheck.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/.github/workflows/spellcheck.yml -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/.gitignore -------------------------------------------------------------------------------- /CODE_OF_CONDUCT.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/CODE_OF_CONDUCT.md -------------------------------------------------------------------------------- /CONTRIBUTING.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/CONTRIBUTING.md -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/README.md -------------------------------------------------------------------------------- /UPDATES.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/UPDATES.md -------------------------------------------------------------------------------- /USE_POLICY.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/USE_POLICY.md -------------------------------------------------------------------------------- /WhatsApp_Finetune.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/WhatsApp_Finetune.ipynb -------------------------------------------------------------------------------- /configs/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/configs/__init__.py -------------------------------------------------------------------------------- /configs/datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/configs/datasets.py -------------------------------------------------------------------------------- /configs/fsdp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/configs/fsdp.py -------------------------------------------------------------------------------- /configs/peft.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/configs/peft.py -------------------------------------------------------------------------------- /configs/training.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/configs/training.py -------------------------------------------------------------------------------- /docs/Dataset.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/docs/Dataset.md -------------------------------------------------------------------------------- /docs/FAQ.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/docs/FAQ.md -------------------------------------------------------------------------------- /docs/LLM_finetuning.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/docs/LLM_finetuning.md -------------------------------------------------------------------------------- /docs/images/feature-based_FN.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/docs/images/feature-based_FN.png -------------------------------------------------------------------------------- /docs/images/featurebased_FN_.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/docs/images/featurebased_FN_.png -------------------------------------------------------------------------------- /docs/images/full-param-FN.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/docs/images/full-param-FN.png -------------------------------------------------------------------------------- /docs/inference.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/docs/inference.md -------------------------------------------------------------------------------- /docs/multi_gpu.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/docs/multi_gpu.md -------------------------------------------------------------------------------- /docs/single_gpu.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/docs/single_gpu.md -------------------------------------------------------------------------------- /ft_datasets/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/ft_datasets/__init__.py -------------------------------------------------------------------------------- /ft_datasets/alpaca_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/ft_datasets/alpaca_dataset.py -------------------------------------------------------------------------------- /ft_datasets/grammar_dataset/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/ft_datasets/grammar_dataset/__init__.py -------------------------------------------------------------------------------- /ft_datasets/grammar_dataset/grammar_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/ft_datasets/grammar_dataset/grammar_dataset.py -------------------------------------------------------------------------------- /ft_datasets/grammar_dataset/grammar_dataset_process.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/ft_datasets/grammar_dataset/grammar_dataset_process.ipynb -------------------------------------------------------------------------------- /ft_datasets/samsum_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/ft_datasets/samsum_dataset.py -------------------------------------------------------------------------------- /ft_datasets/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/ft_datasets/utils.py -------------------------------------------------------------------------------- /ft_datasets/whatsapp_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/ft_datasets/whatsapp_dataset.py -------------------------------------------------------------------------------- /inference/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/README.md -------------------------------------------------------------------------------- /inference/chat_completion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/chat_completion.py -------------------------------------------------------------------------------- /inference/chat_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/chat_utils.py -------------------------------------------------------------------------------- /inference/chats.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/chats.json -------------------------------------------------------------------------------- /inference/checkpoint_converter_fsdp_hf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/checkpoint_converter_fsdp_hf.py -------------------------------------------------------------------------------- /inference/code-llama/code_completion_example.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/code-llama/code_completion_example.py -------------------------------------------------------------------------------- /inference/code-llama/code_completion_prompt.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/code-llama/code_completion_prompt.txt -------------------------------------------------------------------------------- /inference/code-llama/code_infilling_example.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/code-llama/code_infilling_example.py -------------------------------------------------------------------------------- /inference/code-llama/code_infilling_prompt.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/code-llama/code_infilling_prompt.txt -------------------------------------------------------------------------------- /inference/hf-text-generation-inference/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/hf-text-generation-inference/README.md -------------------------------------------------------------------------------- /inference/hf-text-generation-inference/merge_lora_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/hf-text-generation-inference/merge_lora_weights.py -------------------------------------------------------------------------------- /inference/inference.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/inference.py -------------------------------------------------------------------------------- /inference/model_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/model_utils.py -------------------------------------------------------------------------------- /inference/safety_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/safety_utils.py -------------------------------------------------------------------------------- /inference/samsum_prompt.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/samsum_prompt.txt -------------------------------------------------------------------------------- /inference/vLLM_inference.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/inference/vLLM_inference.py -------------------------------------------------------------------------------- /llama_finetuning.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/llama_finetuning.py -------------------------------------------------------------------------------- /model_checkpointing/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/model_checkpointing/__init__.py -------------------------------------------------------------------------------- /model_checkpointing/checkpoint_handler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/model_checkpointing/checkpoint_handler.py -------------------------------------------------------------------------------- /multi_node.slurm: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/multi_node.slurm -------------------------------------------------------------------------------- /policies/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/policies/__init__.py -------------------------------------------------------------------------------- /policies/activation_checkpointing_functions.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/policies/activation_checkpointing_functions.py -------------------------------------------------------------------------------- /policies/anyprecision_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/policies/anyprecision_optimizer.py -------------------------------------------------------------------------------- /policies/mixed_precision.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/policies/mixed_precision.py -------------------------------------------------------------------------------- /policies/wrapping.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/policies/wrapping.py -------------------------------------------------------------------------------- /prepare_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/prepare_dataset.py -------------------------------------------------------------------------------- /preprocessing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/preprocessing.py -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/requirements.txt -------------------------------------------------------------------------------- /scripts/markdown_link_check_config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/scripts/markdown_link_check_config.json -------------------------------------------------------------------------------- /scripts/spellcheck.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/scripts/spellcheck.sh -------------------------------------------------------------------------------- /scripts/spellcheck_conf/spellcheck.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/scripts/spellcheck_conf/spellcheck.yaml -------------------------------------------------------------------------------- /scripts/spellcheck_conf/wordlist.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/scripts/spellcheck_conf/wordlist.txt -------------------------------------------------------------------------------- /utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/utils/__init__.py -------------------------------------------------------------------------------- /utils/config_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/utils/config_utils.py -------------------------------------------------------------------------------- /utils/dataset_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/utils/dataset_utils.py -------------------------------------------------------------------------------- /utils/fsdp_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/utils/fsdp_utils.py -------------------------------------------------------------------------------- /utils/memory_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/utils/memory_utils.py -------------------------------------------------------------------------------- /utils/train_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ads97/WhatsApp-Llama/HEAD/utils/train_utils.py --------------------------------------------------------------------------------