-
-
jax-dropless-moe Public
WIP implementation of block-sparse dropless MoE in JAX
-
-
lorax Public
LoRA for arbitrary JAX models and functions
-
jax Public
Forked from jax-ml/jaxComposable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
Python Apache License 2.0 UpdatedJan 13, 2024 -
-
-
easy-lora-and-gptq Public
JAX notebook showing how to LoRA + GPTQ arbitrary models
-
jax-gptq Public
JAX implementation of GPTQ quantization algorithm
-
-
abnormal-floats Public
Code for the note "NF4 Isn't Information Theoretically Optimal (and that's Good)
-
haiku-mup Public
A port of muP to JAX/Haiku
-
gpt-2-haiku Public
My port of GPT-2 to JAX/haiku. You probably want the HuggingFace FLAX one instead.
-
submitit Public
Forked from facebookincubator/submititPython 3.6+ toolbox for submitting jobs to Slurm
Python MIT License UpdatedApr 20, 2022 -
jax_single_use_rng Public
Simple wrapper to make RNG re-use bugs less likely
Python MIT License UpdatedApr 12, 2022 -
vqgan-haiku Public
Port of VQGAN\Implemented in Haiku\Might still be some bugs
Python MIT License UpdatedJul 15, 2021 -
vgg16-haiku Public
VGG-16 in JAX and Haiku, ported from the torchvision
Python BSD 3-Clause "New" or "Revised" License UpdatedJul 6, 2021 -
-
flax Public
Forked from google/flaxFlax is a neural network ecosystem for JAX that is designed for flexibility.
Python Apache License 2.0 UpdatedApr 8, 2021 -
tf2-gradient-checkpointing Public
Simple gradient checkpointing for eager mode execution
-
transformers Public
Forked from huggingface/transformers🤗 Transformers: State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch.
Python Apache License 2.0 UpdatedDec 12, 2020 -
bert Public
Forked from google-research/bertTensorFlow code and pre-trained models for BERT
Python Apache License 2.0 UpdatedApr 24, 2019 -
-
finetune-transformer-lm Public
Forked from openai/finetune-transformer-lmCode and model for the paper "Improving Language Understanding by Generative Pre-Training"
Python MIT License UpdatedDec 10, 2018 -
asyncbots Public
A framework for simplifying writing RTM bots for Slack.
-
e2e-coref Public
Forked from kentonl/e2e-corefEnd-to-end Neural Coreference Resolution
Python Apache License 2.0 UpdatedJan 16, 2018 -
-
models Public
Forked from tensorflow/modelsModels built with TensorFlow
Python Apache License 2.0 UpdatedJan 29, 2017 -
tensorflow Public
Forked from tensorflow/tensorflowComputation using data flow graphs for scalable machine learning
C++ Apache License 2.0 UpdatedAug 29, 2016