sergio sainz @sergiosp13

engineer, deep learning enthusiast , climber , and slow runner Joined September 2010

Tweets

249
Followers

38
Following

329
Likes

4K

Matthew Berman @MatthewBerman

8 months ago

1/ Google Research unveils new paper: "Titans: Learning to Memorize at Test Time" It introduces human-like memory structures to overcome the limits of Transformers, with one "SURPRISING" feature. Here's why this is huge for AI. 🧵👇

59 447 3K 431K 3K

Download Image

Eugene Vinitsky (@RLC) 🍒🦋 @EugeneVinitsky

a year ago

This page of common pytorch mistakes is pretty invaluable uvadlc-notebooks.readthedocs.io/en/latest/tuto…

4 106 826 80K 1K

Download Image

Richard Sutton @RichardSSutton

a year ago

Everything you know about the world is a belief about the statistics of your sensory input and how they depend on your output. There is nothing more to it, and understanding knowledge in this sense is one key to creating AI.

36 45 361 39K 121

Andrej Karpathy @karpathy

a year ago

🔥llm.c update: Our single file of 2,000 ~clean lines of C/CUDA code now trains GPT-2 (124M) on GPU at speeds ~matching PyTorch (fp32, no flash attention) github.com/karpathy/llm.c… On my A100 I'm seeing 78ms/iter for llm.c and 80ms/iter for PyTorch. Keeping in mind this is fp32,…

151 534 5K 638K 2K

Download Image

AI at Meta @AIatMeta

2 years ago

Today we’re releasing Code Llama 70B: a new, more performant version of our LLM for code generation — available under the same license as previous Code Llama models. Download the models ➡️ bit.ly/3Oil6bQ • CodeLlama-70B • CodeLlama-70B-Python • CodeLlama-70B-Instruct

164 1K 5K 891K 1K

Download Image

PyTorch @PyTorch

2 years ago

Do you want to adapt an LLM on your own data and domain? 🤔 Learn how to finetune a 7B parameter model on a typical consumer GPU (NVIDIA T4 16GB) with LoRA and tools from the PyTorch and Hugging Face ecosystem in our latest post. Details 👉 hubs.la/Q02g0x1N0

7 170 729 68K 551

Download Image

Sasha Rush @srush_nlp

2 years ago

Several people have told me over drinks that these puzzles are being used for ML tech interviews 🤣. github.com/srush/Tensor-P… github.com/srush/GPU-Puzz…

15 134 1K 211K 3K

Jim Fan @DrJimFan

2 years ago

One of the best tutorial-style repos since @karpathy's minGPT! GPT-Fast: a minimalistic, PyTorch-only decoding implementation loaded with best practices: int8/int4 quantization, speculative decoding, Tensor parallelism, etc. Boosts the "clock speed" of LLM OS by 10x with no model…

25 408 3K 407K 2K

Download Video

Yann LeCun @ylecun

2 years ago

The UAE Minister of AI @OmarSAlolama points to a historical precedent of premature technology regulation motivated by fear: the ban of the printing press in 1515 by Sultan Selim I led to the decline of the Ottoman Empire. “We overregulated a technology, which was the printing…

208 870 4K 749K 866

Bindu Reddy @bindureddy

2 years ago

OpenAI Developer Day Announcement and Implications for Open-Source AI development OpenAI's developer day was filled with a bunch of annoncements. Most notable amongst them was a 3x price cut on GPT-4-turbo, Custom GPTs and Assistants APIs Sadly and predictably, there was no…

91 169 1K 412K 524

Download Image

Matt Shumer @mattshumer_

2 years ago

Here is an incredible GPT-4 prompt for engineers. Use it to speed up any code by identifying inefficiencies and rectifying them: --- <prompt_explanation> You are a world expert in making code run faster. You use any resource you can to do so. Given some code, first, explain…

49 189 2K 388K 4K

hardmaru @hardmaru

2 years ago

Artificial lifeforms are super fascinating to watch. These self-organizing, self-replicating, “lifeforms” emerged from a continuous time cellular automata system called Flow-Lenia. Lenia is a family of CAs generalizing Conway’s Game of Life to continuous space, time and states.

33 513 3K 566K 740

Download Video

sergio sainz @sergiosp13

2 years ago

Beijing Fall 2023

1 0 6 37 0

Download Image

Beff – e/acc @BasedBeffJezos

2 years ago

@elonmusk Sama playbook: 1) scrape the open internet 2) train a large model to compress it 3) sell a slow drip back to users via API 4) make it illegal for other players to do the same The internet collectively getting mugged and being forced to buy back its assets from the bandits

16 42 525 1.1M 55

Sebastian Risi @risi1979

2 years ago

Introducing Neural Developmental Programs (NDPs)🧬🧠Instead of neural networks with fixed architectures, we allow neural networks to grow through a dynamic self-organizing process, inspired by how biological nervous systems develop👇 PDF: arxiv.org/abs/2307.08197

27 263 1K 239K 634

Download Video

Jerry Liu @jerryjliu0

2 years ago

A key way you can make your LLM/RAG chatbot more “advanced” over complex data sources is to add a router - dynamically decide which data to query / which parameters to use using LLMs. It’s a key step towards LLM workflow automation. It sounds simple, but there’s some trickiness:…

9 86 487 92K 594

Download Image

AI at Meta @AIatMeta

2 years ago

Today we’re releasing Code Llama, a large language model built on top of Llama 2, fine-tuned for coding & state-of-the-art for publicly available coding tools. Keeping with our open approach, Code Llama is publicly-available now for both research & commercial use. More ⬇️

170 976 3K 1.0M 776

Andrej Karpathy @karpathy

2 years ago

"How is LLaMa.cpp possible?" great post by @finbarrtimbers finbarr.ca/how-is-llama-c… llama.cpp surprised many people (myself included) with how quickly you can run large LLMs on small computers, e.g. 7B runs @ ~16 tok/s on a MacBook. Wait don't you need supercomputers to work…

81 730 5K 917K 3K

Download Image

Matt Shumer @mattshumer_

2 years ago

Built a notebook that makes it dumb simple to fine-tune LLaMA 2. Just load in a dataset, and run it! colab.research.google.com/drive/1Zmaceu6…

23 368 2K 189K 2K

Tivadar Danka @TivadarDanka

2 years ago

The Japanese multiplication method makes everybody feel "I wish they taught math like this in school." It's not just a cute visual tool: it illuminates how and why long multiplication works. Here is the full story.