Pinned Repositories
360blockscope
Block scoping in python (ironically)
abnormal-floats
Code for the note "NF4 Isn't Information Theoretically Optimal (and that's Good)
asyncbots
A framework for simplifying writing RTM bots for Slack.
easy-lora-and-gptq
JAX notebook showing how to LoRA + GPTQ arbitrary models
haiku-mup
A port of muP to JAX/Haiku
jax-dropless-moe
WIP implementation of block-sparse dropless MoE in JAX
jax-gptq
JAX implementation of GPTQ quantization algorithm
lorax
LoRA for arbitrary JAX models and functions
qax
If it quacks like a tensor...
tf2-gradient-checkpointing
Simple gradient checkpointing for eager mode execution
davisyoshida's Repositories
davisyoshida/lorax
LoRA for arbitrary JAX models and functions
davisyoshida/qax
If it quacks like a tensor...
davisyoshida/tf2-gradient-checkpointing
Simple gradient checkpointing for eager mode execution
davisyoshida/haiku-mup
A port of muP to JAX/Haiku
davisyoshida/abnormal-floats
Code for the note "NF4 Isn't Information Theoretically Optimal (and that's Good)
davisyoshida/easy-lora-and-gptq
JAX notebook showing how to LoRA + GPTQ arbitrary models
davisyoshida/jax-gptq
JAX implementation of GPTQ quantization algorithm
davisyoshida/jax-dropless-moe
WIP implementation of block-sparse dropless MoE in JAX
davisyoshida/360blockscope
Block scoping in python (ironically)
davisyoshida/asyncbots
A framework for simplifying writing RTM bots for Slack.
davisyoshida/gpt-2-haiku
My port of GPT-2 to JAX/haiku. You probably want the HuggingFace FLAX one instead.
davisyoshida/bert
TensorFlow code and pre-trained models for BERT
davisyoshida/ctfp_work
davisyoshida/vgg16-haiku
VGG-16 in JAX and Haiku, ported from the torchvision
davisyoshida/vqgan-haiku
Port of VQGAN\Implemented in Haiku\Might still be some bugs
davisyoshida/1bwords_subset
davisyoshida/datasets
Tensorflow helpers for loading various datasets
davisyoshida/davisyoshida.github.io
davisyoshida/e2e-coref
End-to-end Neural Coreference Resolution
davisyoshida/finetune-transformer-lm
Code and model for the paper "Improving Language Understanding by Generative Pre-Training"
davisyoshida/flax
Flax is a neural network ecosystem for JAX that is designed for flexibility.
davisyoshida/gemma-reimpl
Implementation of Gemma in Jax/Flax
davisyoshida/jax
Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
davisyoshida/jax_single_use_rng
Simple wrapper to make RNG re-use bugs less likely
davisyoshida/llama-haiku
JAX/Haiku LLama implementatoin
davisyoshida/models
Models built with TensorFlow
davisyoshida/submitit
Python 3.6+ toolbox for submitting jobs to Slurm
davisyoshida/tensorflow
Computation using data flow graphs for scalable machine learning
davisyoshida/transformers
🤗 Transformers: State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch.