EasyLM
trax
EasyLM | trax | |
---|---|---|
8 | 7 | |
2,247 | 7,964 | |
- | 0.4% | |
7.7 | 4.3 | |
5 months ago | 5 days ago | |
Python | Python | |
Apache License 2.0 | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
EasyLM
- Maxtext: A simple, performant and scalable Jax LLM
- How To Fine-Tune LLaMA, OpenLLaMA, And XGen, With JAX On A GPU Or A TPU
-
Open-sourced LLMs are adept at mimicking ChatGPT’s style but not its factuality. There exists a substantial capabilities gap, which requires better base LM.
Title: The False Promise of Imitating Proprietary LLLs Authors: Arnav Gudibande, Eric Wallace, Charlie Snell, Xinyang Geng, Hao Liu, Pieter Abbeel, Sergey Levine, Dawn Song Word Count: 3400 Average Reading Time: 18-20 minutes Source Code: https://github.com/young-geng/EasyLM Additional Links: https://huggingface.co/young-geng/koala-eval, https://huggingface.co/young-geng/koala
-
Paid dev gig: develop a basic LLM PEFT finetuning utility
Check out easyLM https://github.com/young-geng/EasyLM
-
OpenLLaMA Releases 7B/3B Checkpoints with 700B/600B Tokens
We release the weights in two formats: an EasyLM format to be use with our EasyLM framework, and a PyTorch format to be used with the Hugging Face transformers library.
-
OpenLLaMA: An Open Reproduction of LLaMA
I am quite new to this, I would like to get it running. Would the process roughly be:
1. Get a machine with decent GPU, probably rent cloud GPU.
2. On that machine download the weights/model/vocab files from https://huggingface.co/openlm-research/open_llama_7b_preview...
3. Install Anaconda. Clone https://github.com/young-geng/EasyLM/.
4. Install EasyLM:
conda env create -f scripts/gpu_environment.yml
- Koala: A Dialogue Model for Academic Research [Finetuned Llama-13B on a dataset generated by ChatGPT]
trax
-
Maxtext: A simple, performant and scalable Jax LLM
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
-
Replit's new Code LLM was trained in 1 week
and the implementation https://github.com/google/trax/blob/master/trax/models/resea... if you are interested.
Hope you get to look into this!
-
RedPajama: Reproduction of Llama with Friendly License
Thank you for developing the pipeline and amassing considerable compute for gathering and preprocessing this dataset!
I'm not sure if this is the right place to ask about this, but could you consider training an LLM using a more advanced, sparse transformer architecture (specifically, "Terraformer" from this paper https://arxiv.org/abs/2111.12763 and this codebase https://github.com/google/trax/blob/master/trax/models/resea... by Google Brain and OpenAI)? I understand the pressure to focus on training a straightforward LLaMA replication, but of course you see that it's a legacy dense architecture which limits its inference performance. This new architecture is not just an academic curiosity but is already validated at scale by Google, providing 10x+ inference performance boost on the same hardware.
Frankly, the community's compute budget - for training and for inference - isn't infinite, and neither is the public's interest in models that do not have advantage (at least in convenience) over closed-source ones; and so we should utilize both those resources as efficiently as possible. It could be a big step forward if you trained at least LLaMA-Terraformer-7B and 13B foundation models on the whole dataset.
-
The founder of Gmail claims that ChatGPT can “kill” Google in two years.
But a couple years later they came out with open source implementations yeah: https://github.com/google/trax/tree/master/trax/models/reformer
-
[D] Paper Explained - Sparse is Enough in Scaling Transformers (aka Terraformer) | Video Walkthrough
Code: https://github.com/google/trax/blob/master/trax/examples/Terraformer_from_scratch.ipynb
- Why would I want to develop yet another deep learning framework?
-
How to train large models on a normal laptop?
Training language models is expensive. Train the biggest model you can afford. I assume you've tried the colab from the reformer GitHub: https://github.com/google/trax/tree/master/trax/models/reformer
What are some alternatives?
mlc-llm - Enable everyone to develop, optimize and deploy AI models natively on everyone's devices.
flax - Flax is a neural network library for JAX that is designed for flexibility.
camel - 🐫 CAMEL: Communicative Agents for “Mind” Exploration of Large Language Model Society (NeruIPS'2023) https://www.camel-ai.org
dm-haiku - JAX-based neural network library
Open-Llama - The complete training code of the open-source high-performance Llama model, including the full process from pre-training to RLHF.
muzero-general - MuZero
brev-cli - Connect your laptop to cloud computers. Follow to stay updated about our product
ML-Optimizers-JAX - Toy implementations of some popular ML optimizers using Python/JAX
RWKV-LM - RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.
extending-jax - Extending JAX with custom C++ and CUDA code
modal-examples - Examples of programs built using Modal
objax