Long reads on models, training & the math behind them.https://www.parth-shastri.com/https://www.parth-shastri.com/blog/open-training-recipes/https://www.parth-shastri.com/blog/open-training-recipes/The open strategies for building and refining capable LLMs, from OLMo's pre-training to the Tulu post-training recipe from AI2.Sun, 01 Jun 2025 00:00:00 GMTTraininghttps://www.parth-shastri.com/blog/phi2ning/https://www.parth-shastri.com/blog/phi2ning/A walkthrough of fine-tuning Microsoft's 2.7B parameter Phi-2 with supervised fine-tuning (SFT) and Direct Preference Optimization (DPO).Mon, 29 Jan 2024 00:00:00 GMTFine-tuninghttps://www.parth-shastri.com/blog/understanding-llms/https://www.parth-shastri.com/blog/understanding-llms/The Transformer architecture behind large language models, and the generative pre-training that brings them to life.Sun, 30 Jul 2023 00:00:00 GMTArchitecturehttps://www.parth-shastri.com/blog/demystifying-diffusion-models/https://www.parth-shastri.com/blog/demystifying-diffusion-models/How diffusion models like DALL-E and Midjourney work, from the forward noising process to the simple loss that trains them (DDPM).Fri, 14 Oct 2022 00:00:00 GMTGenerative