transformer

13 stories and discussions about transformer, aggregated from every source we track.

1.

A custom AI architecture being developed in rust . Contribute to Sparticle62ops/pssa development by creating an account on GitHub.

77 points•sparticle62•about 2 months ago•32 comments•
2.

Train a tiny GPT in under a minute (CUDA only). Contribute to lostmsu/TurboGPT development by creating an account on GitHub.

56 points•lostmsu•about 2 months ago•10 comments•
3.

Posted by Jakob Uszkoreit, Software Engineer, Natural Language Understanding Neural networks, in particular recurrent neural networks (RNNs), are n...

4 points•adsouza•about 9 years ago•0 comments
4.

Paper Tape is All You Need. Contribute to dbrll/ATTN-11 development by creating an account on GitHub.

3 points•raymii•7 months ago•0 comments
5.
2 points•qrify_app•about 2 months ago•0 comments•
6.

While Large Language Models (LLMs) rely on highly non-linear components, in this work we demonstrate that they exhibit fundamental linearity: when inputs from distinct text streams are linearly combined, the model…

2 points•sbulaev•about 2 months ago•0 comments•
7.

I've been learning about Vision Transformers, and it's been a ride. It's probably the densest...

2 points•marshateo•2 months ago•0 comments
8.

Performing inference on a Transformer can be very different from training. Partly this is because inference adds a new factor to consider: latency. In this section, we will go all the way from sampling a single new…

2 points•aray07•2 months ago•0 comments•
9.
1 points•ksec•about 2 months ago•0 comments•
10.
1 points•yarapavan•about 2 months ago•0 comments•
11.

Nemotron-H is a series of hybrid Mamba-Transformer models which offer either better or on-par accuracy and improved inference speed (up to 3x) compared to other similarly-sized state-of-the-art open-sourced pure…

1 points•Bluestein•2 months ago•0 comments•
12.

Paradigma releases Limite 1B - Violetto, a 1-billion parameter transformer for high-throughput mathematical reasoning, alongside its evaluations, a training-time value model, and a custom vLLM inference plugin.

1 points•E-Reverance•2 months ago•0 comments•
13.

A visual technical deep dive into Jev: transformer inference, typed probabilities, proper scoring rules, policy gradients, and what RLCD changes.

1 points•madmax108•2 months ago•0 comments•

Related topics