Resources

Useful links:

• FineWeb (pretraining dataset)
• Tiktokenizer
• Transformer Neural Net 3D visualizer
• llm.c Let's Reproduce GPT-2
• Llama 3 paper from Meta
• Hyperbolic, for inference of base model
• InstructGPT paper on SFT
• HuggingFace inference playground
• DeepSeek-R1 paper
• TogetherAI Playground for open model inference
• AlphaGo paper (PDF)
• AlphaGo Move 37 video
• LM Arena for model rankings
• AI News Newsletter
• LMStudio for local inference