AllanYangZhou / midGPT
Distributed pretraining of large language models (LLMs) on cloud TPU slices, with Jax and Equinox.
☆24Updated 7 months ago
Alternatives and similar repositories for midGPT
Users that are interested in midGPT are comparing it to the libraries listed below
Sorting:
- Minimal but scalable implementation of large language models in JAX☆34Updated 6 months ago
- A simple library for scaling up JAX programs☆134Updated 6 months ago
- Einsum-like high-level array sharding API for JAX☆34Updated 10 months ago
- ☆109Updated this week
- JAX implementation of the Mistral 7b v0.2 model☆35Updated 10 months ago
- Experiment of using Tangent to autodiff triton☆78Updated last year
- ☆217Updated 10 months ago
- LoRA for arbitrary JAX models and functions☆136Updated last year
- Transformer with Mu-Parameterization, implemented in Jax/Flax. Supports FSDP on TPU pods.☆30Updated this week
- Two implementations of ZeRO-1 optimizer sharding in JAX☆14Updated last year
- A set of Python scripts that makes your experience on TPU better☆53Updated 10 months ago
- If it quacks like a tensor...☆58Updated 6 months ago
- Implementation of PSGD optimizer in JAX☆33Updated 4 months ago
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆123Updated last year
- Jax/Flax rewrite of Karpathy's nanoGPT☆57Updated 2 years ago
- ☆79Updated 10 months ago
- seqax = sequence modeling + JAX☆155Updated last month
- ☆14Updated 10 months ago
- Proof-of-concept of global switching between numpy/jax/pytorch in a library.☆18Updated 10 months ago
- A MAD laboratory to improve AI architecture designs 🧪☆115Updated 4 months ago
- 🧱 Modula software package☆189Updated last month
- An implementation of the Llama architecture, to instruct and delight☆21Updated 4 months ago
- ☆20Updated last year
- Named Tensors for Legible Deep Learning in JAX☆173Updated 2 weeks ago
- Train very large language models in Jax.☆204Updated last year
- This is a port of Mistral-7B model in JAX☆32Updated 10 months ago
- ☆16Updated 6 months ago
- The simplest, fastest repository for training/finetuning medium-sized GPTs.☆120Updated last week
- ☆37Updated last year
- supporting pytorch FSDP for optimizers☆80Updated 5 months ago