Vatsa PandeyEntering the LLM world with RNNs: CharRNNRecurrent neural networks (RNNs) are a type of artificial intelligence system that is modeled loosely after the neurons in the human brain…Oct 13, 2023Oct 13, 2023
Vatsa PandeyinTowards Generative AITiny Llama, A 1.1B model trained on three trillion tokensToday, I just saw a really interesting project, TinyLlama. The project is based on Llama-2 Architecture, and it aims to “pretrain a 1.1B…Sep 12, 2023Sep 12, 2023
Vatsa PandeyinTowards Generative AIUnagami, A Mainstream-like LLM, in 350 million parametersSep 7, 2023Sep 7, 2023
Vatsa PandeyA full tutorial on turning GPT-2 into a Chatty AIA couple days ago, I built NanoChatGPT, a model fine tuned on GPT-2-medium. When most people see GPT-2 they think of an autocomplete…Aug 31, 2023Aug 31, 2023