LLM Alchemy Chamber 🧙‍♂️✨

Welcome to a friendly neighborhood repository featuring diverse experiments and adventures in the world of LLMs. This collection is no ordinary repository; it's an alchemical blend of scripts, notebooks, and experiments dedicated to the mystical realm of Language Models (LLMs).

Alchemical Scripts

Projects	GitHub Link	Colab Link	Blog Link	Description
Youtube Cloner	Folder	Fireship GPT	Blog coming soon	An Attempt at cloning youtubers using LLMs by Finetuning

Finetuning	GitHub Link	Colab Link	Blog Link	Description
Gemma Finetuning	GitHub	Colab	A Beginner’s Guide to Fine-Tuning Gemma	Notebook to Finetune Gemma Models
Mistral-7b Finetuning	GitHub	Colab	A Beginner’s Guide to Fine-Tuning Mistral 7B Instruct Model	Notebook to Finetune Mistral-7b Model
Mixtral Finetuning	GitHub	Colab	A Beginner’s Guide to Fine-Tuning Mixtral Instruct Model	Notebook to Finetune Mixtral-7b Models
LLama2 Finetuning	GitHub	Colab		Notebook to Finetune Llama2-7b Model

Quantization	GitHub Link	Colab Link	Blog Link	Description
AWQ Quantization	GitHub	Colab	Squeeze Every Drop of Performance from Your LLM with AWQ	quantise LLM using AWQ.
GGUF Quantization	GitHub	Colab	Run any Huggingface model locally	quantise LLM to GGUF formate.

Data Prep	GitHub Link	Colab Link	Description
Documents -> Dataset	GitHub	Colab	Given Documents generate Instruction/QA dataset for finetuning LLMs
Topic -> Dataset	GitHub	Colab	Given a Topic generate a dataset to finetune LLMs
Alpaca Dataset Generation	GitHub	Colab	The original implementation of generating instruction dataset followed in the alpaca paper

Repo Structure

├── DataPrep (Notebook to generate synthetic data)
│   ├── dataset_prep.ipynb
│   └── ...
├── Deployment (TGI/VLLM scripts for testing)
│   └── ...
├── Finetuning (Finalized Finetuning Scripts)
│   ├── Gemma_finetuning_notebook.ipynb
│   ├── Llama2_finetuning_notebook.ipynb
│   ├── Mistral_finetuning_notebook.ipynb
│   ├── Mixtral_finetuning_notebook.ipynb
│   └── ...
├── LLMS (LLM experiments)
│   ├── ambari
│   │   └── ...
│   ├── CodeLLama
│   │   └── ...
│   ├── Gemma
│   │   ├── finetune-gemma.ipynb
│   │   └── gemma-sft.py
│   ├── Llama2
│   │   └── ...
│   ├── Mistral-7b
│   │   └── ...
│   └── Mixtral
│       └── ...
├── Projects (Upcoming ideas to explore)
│   └── YT_Clones
│       ├── Fireship_clone.ipynb
│       ├── youtube_channel_scraper.py
│       └── ...
├── Quantization
│   └── ...
├── utils
│   └── streaming_inference_hf.ipynb
└── RAG (Retrieval Augmented Generation)
    ├── 1_Naive_RAG.ipynb
    ├── 2_Semantic_Chunking_RAG.ipynb
    ├── 3_Sentence_Window_Retrieval_RAG.ipynb
    ├── 4_Auto_Merging_Retrieval_RAG.ipynb
    ├── 5_Agentic_RAG.ipynb
    └── 6_Visual_RAG.ipynb