zphang

minimal-llama

minimal-gpt-neox-20b

bert_on_stilts

Fork of huggingface/pytorch-pretrained-BERT for BERT on STILTs

minimal-opt

transformers

Code and models for BERT on STILTs

llm_feedback

lrqa

minimal-t5

hyperllama

my_pefty_llama

Minimal implementation of multiple PEFT methods for LLaMA fine-tuning

sndict

Structured Nested Dictionaries

hpt

llama_peft

zphang.github.io

Github Page

architecture-objective

DeepSpeed

DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.

doc-chat-ui

FLAN

gpt-neox

An implementation of model parallel autoregressive transformers on GPUs, based on the DeepSpeed library.

GPTQ-for-LLaMa

4 bits quantization of LLaMa using GPTQ

hf_benchmark_sample

jiant

The jiant toolkit for general-purpose text understanding models

lm_evaluation_harness

Megatron-DeepSpeed

Ongoing research training transformer language models at scale, including: BERT & GPT-2

mpi4py

Python bindings for MPI

peft

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

pegasus

summarization_experiments

t5x

zutils