LLM Thesis Toolkit 🧰

🧠 Fundamentals of ML 3Blue1Brown: Neural Networks series 📖 Fundamentals of LLMs Intro to LLMs Language Modeling Tokenization BPE Tokenization Vectorizing Words (Word Embeddings) Intro to Large Language Models Developing an LLM: Building, Training, Finetuning Visualizing transformers and attention, by Grant Sanderson How large language models work, a visual intro to transformers Text Generation and Sampling Strategies 👁‍🗨 Visualizations Transformer Visualization Another Transformer Visualization 🐱 Hands-on Tutorials Huggingface Sampling methods and text generation 🚀 Advanced Topics 🐱 More advanced topics: LoRA, Quantization, etc. Various skills resources 📚 Recommended Courses Advanced NLP CMU LLM Understanding NN Zero to Hero by Karpathy 🍃 Writing Planning and Gantt chart Task management 🚦 Principles of Research reproducibility: TODO minimum experiment: TODO benchmarks and baselines evaluations analyzing

🔐 Privacy & Security of LLMs 📑 Key Papers Undesirable Memorization in LLMs: A Survey Privacy Issues in LLMs Original data extraction paper 🎥 Talks & Tutorials Video by carlini More detailed version of the same video Eric Wallace on Memorization 🧩 Snippets Calculating perplexity Finetuning Large Models (more than 1 GPU) ⚙️ Implementation 🦅 Hábrók mini-wiki 🖥 Programming conventions 📟 Bash Little Cheatsheet 🗽 Models Pythia and GPT-Neo Hubble Suite Opt family EL-MIA Qwen finetuned on Epstien files 📚 Datasets EL-MIA: Entity-level memorization Ai4Privacy Hubble Enron Corp. Emails ECtHR: European Court of Human Rights YAGO Biographies