JAX, every mention
12 scenes · ← back to JAX
tap a year for its mentions
every year anyone Kevin Wang 4Dylan Patel 4Shawn Wang 3Mark Huang 3Jeremy Howard 3Yi Tay 2Nathan Lambert 2
Verbatim, from the transcripts: the passages where JAX comes up
Claude Code for Finance + The Global Memory Shortage: Doug O'Laughlin, SemiAnalysis
- ▶ 1:36:59 Shawn Wang I, I don't know if this is something that affects your analysis at all, because I don't have any appreciation for the, or the, the sizes that we're talking about here, that JAX is helping TPUs win, or JAX is winning relatively to PyTorch,… 2 times in the scene
[NeurIPS Best Paper] 1000 Layer Networks for Self-Supervised RL — Kevin Wang et al, Princeton
- ▶ 15:55 Kevin Wang And so, in our environment, we, in our research, we specifically used the JAX-GCRL environment, which is a JAX-based GPU-accelerate environment, so we can collect, like, thousands of, like, environment trajectories, like, in parallel at… 4 times in the scene
Information Theory for Language Models: Jack Morris
- ▶ 13:50 Shawn Wang A lot of people switch from like JAX to CUDA, but like the thing is just like being able to experiment very quickly on a limited budget.
[LLM Paper Club] Llama 3.1 Paper: The Llama Family of Models
- ▶ 18:29 unnamed speaker I thought Google has spiralism in their, uh, Jax, the distributed training repositories.
The 10,000x Yolo Researcher Metagame — with Yi Tay of Reka
How to train a Million Context LLM — with Mark Huang of Gradient.ai
- ▶ 25:49 Mark Huang Like, it was, uh, um, uh, I would say the original authors, you know, Matai and all the folks at, uh, uh, Berkeley, they created the JAX implementation for it. 3 times in the scene
A Comprehensive Overview of Large Language Models - Latent Space Paper Club
- ▶ 24:12 unnamed speaker So, things like, um, the libraries that we're using, um, JAX, PyTorch, TensorFlow, amongst others, um, there's this idea of distributed training, which means that, um, can we use multiple GPUs to train our models so that we are able to…
The Origin and Future of RLHF: the secret ingredient for ChatGPT - with Nathan Lambert
- ▶ 1:19:17 Nathan Lambert Like, I showed up in the grad student, Hamish Iveson, and I need to learn how to pronounce last names better, but he had some Jack's DPO code built on this EZLM framework, and we have, like, have deep TPUs that we could access for research… 2 times in the scene
The State of Silicon and the GPU Poors - with Dylan Patel of SemiAnalysis
- ▶ 11:33 Dylan Patel I mean, Google internally, and I think, you know, is, is obviously on JAX and XLA and all that kind of stuff, right? 4 times in the scene
The End of Finetuning — with Jeremy Howard of Fast.ai
- ▶ 1:02:03 Jeremy Howard You know, they have JAX, which was never a strategy. 3 times in the scene
RWKV: Reinventing RNNs for the Transformer Era
- ▶ 1:06:24 unnamed speaker Is, uh, is JAX interesting to people? 2 times in the scene