Recent Posts

• Implementing DeepSeek R1's GRPO algorithm from scratch
  by xcodevnon 4/13/2025, 6:33:05 PM with 1 comment
• The LLM pre-training data wall
  by xcodevnon 11/23/2024, 5:51:32 PM with 0 comments
• PodcastLM: An open-source AI podcast creator
  by xcodevnon 11/2/2024, 10:44:28 AM with 0 comments
• Scaling up self-attention inference
  by xcodevnon 9/29/2024, 5:49:50 AM with 0 comments
• Scaling up self-attention inference
  by xcodevnon 9/28/2024, 9:23:39 AM with 0 comments
• Letter from Professors Bengio, Hinton, Lessig, & Russell
  by xcodevnon 8/8/2024, 5:21:29 PM with 0 comments
• Logit Prisms: Decomposing Transformer Outputs for Mechanistic Interpretability
  by xcodevnon 6/17/2024, 11:57:50 AM with 4 comments
• Exploring MLP neurons inside Llama3 model
  by xcodevnon 6/8/2024, 7:32:55 PM with 1 comment
• Self-reference is so cool
  by xcodevnon 3/31/2022, 12:22:04 PM with 0 comments
• Pax: a PyTorch-like pytree module on top of Jax
  by xcodevnon 9/4/2021, 3:42:06 PM with 0 comments
• Google Colab Pro+ ($50/month)
  by xcodevnon 8/13/2021, 2:50:48 AM with 0 comments
• Prompting GTP-3 is like saying in your head
  by xcodevnon 4/21/2021, 4:38:07 PM with 0 comments
• GPT 9000 for Speech Signal
  by xcodevnon 3/11/2021, 4:17:58 PM with 1 comment
• Generative MNIST-RNN
  by xcodevnon 9/21/2020, 2:30:33 AM with 0 comments
• Parallel Worlds Probably Exist. Here’s Why [video]
  by xcodevnon 3/6/2020, 7:37:47 PM with 0 comments
• GPT-2: 6-Month Follow-Up
  by xcodevnon 8/20/2019, 4:14:17 PM with 12 comments
• Finding the Brain’s Compass
  by xcodevnon 8/19/2019, 2:38:10 PM with 1 comment
• Reinforcement learning: Fast and slow – Matthew Botvinick [video]
  by xcodevnon 11/18/2018, 5:13:19 AM with 0 comments