unsloth VS transformers

Gemma 7b is 2.4x faster than HF + FA2.

Check out https://github.com/unslothai/unsloth for full benchmarks!

Here are the missing links:

* Gemma, a family of open models from Google: https://ai.google.dev/gemma

* Unsloth is a tool/method for training models faster (IIUC): https://github.com/unslothai/unsloth

Thanks! Again, partnerships over customers. If you're experienced and have the technical chops to make a MI300x sing, we want to work with you. Our model is that we are the capex/opex investor for businesses. As much as I love software, Hot Aisle is more of a hardware business. Running super high end large scale compute is an extreme challenge in itself. We are less interested in building the software side of things and want to foster those who can focus on that side.

https://github.com/unslothai/unsloth/issues/160

https://github.com/search?q=repo%3Apredibase%2Florax+rocm&ty...

https://github.com/sgl-project/sglang/issues/157

https://github.com/casper-hansen/AutoAWQ (supports rocm)

Unsloth’s colab notebooks for fine-tuning Mistral-7B are super easy to use and run fine in just about any colab instance:

https://github.com/unslothai/unsloth

It’s my default now for experimenting and basic training. If I want to get into the weeds with the training, I use axolotl, but 9/10, it’s not really necessary.

If anyone wants to finetune their own Mistral 7b model 2.2x faster and use 62% less memory - give our open source package Unsloth a try! https://github.com/unslothai/unsloth a try! :)

https://github.com/unslothai/unsloth seems good and more relevant to your aims perhaps but I haven't tried it.

The unsloth project offers some low-level optimizations for Llama et al, and as of today some prelim Mistral work (which I heard is the llama architecture?)

This seems to just be a link to the Unsloth Github repo[0], which in turn is the free version of Unsloth Pro/Max[1]. Maybe the link should be changed?

[0]: https://github.com/unslothai/unsloth

Fascinating work, very promising.

Can you summarise how the model in your paper differs from this one ?

https://github.com/huggingface/transformers/issues/27011

transformers is a package by Hugging Face, that helps you interact with models on HF Hub (GitHub)

Is t5x an encoder/decoder architecture?

Some more general options.

The Flax ecosystem

https://github.com/google/flax?tab=readme-ov-file

or dm-haiku

https://github.com/google-deepmind/dm-haiku

were some of the best developed communities in the Jax AI field

Perhaps the “trax” repo? https://github.com/google/trax

Some HF examples https://github.com/huggingface/transformers/tree/main/exampl...

Sadly it seems much of the work is proprietary these days, but one example could be Grok-1, if you customize the details. https://github.com/xai-org/grok-1/blob/main/run.py

The HuggingFace transformers library already has support for a similar method called prompt lookup decoding that uses the existing context to generate an ngram model: https://github.com/huggingface/transformers/issues/27722

I don't think it would be that hard to switch it out for a pretrained ngram model.

Most of this tutorial is based on Hugging Face course about Transformers and on Niels Rogge's Transformers tutorials: make sure to check their work and give them a star on GitHub, if you please ❤️

* Superconvergence + LR range finder + Fast AI's Ranger21 optimizer was the goto optimizer for CNNs, and worked fabulously well, but on transformers, the learning rate range finder sadi 1e-3 was the best, whilst 1e-5 was better. However, the 1 cycle learning rate stuck. https://github.com/huggingface/transformers/issues/16013

Thanks! :) I'm pushing them into transformers, pytorch-gemma and collabing with the Gemma team to resolve all the issues :)

The RoPE fix should already be in transformers 4.38.2: https://github.com/huggingface/transformers/pull/29285

My main PR for transformers which fixes most of the issues (some still left): https://github.com/huggingface/transformers/pull/29402