#计算机科学#DeepSpeed Chat: 一键式RLHF训练，让你的类ChatGPT千亿大模型提速省钱15倍
An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries
翻译 - 基于DeepSpeed库的GPU上类似于GPT-3的模型并行模型的实现。设计为能够训练成千上亿个参数或更大参数的模型。
Princeton NLP's pre-training library based on fairseq with DeepSpeed kernel integration 🚃
Simple and efficient RevNet-Library for PyTorch with XLA and DeepSpeed support and parameter offload
Use the two different methods (deepspeed and SageMaker model parallelism library) to fine tune llama model on Sagemaker. Then deploy the fine tuned llama on Sagemaker with server side batch.
Tiny-DeepSpeed, a minimalistic re-implementation of the DeepSpeed library
Experiments with DeepSpeed MII library
A library that experiments with different DeepSpeed options for Inference
Using the deepspeed library to prepare a new fine-tuner script
Example models using DeepSpeed
chatglm多gpu用deepspeed和
A plug-in of Microsoft DeepSpeed to fix the bug of DeepSpeed pipeline
llama2 finetuning with deepspeed and lora
Teacher - student distillation using DeepSpeed
Intel® Extension for DeepSpeed* is an extension to DeepSpeed that brings feature support with SYCL kernels on Intel GPU(XPU) device. Note XPU is already supported by stock DeepSpeed.
#计算机科学#MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.
Testing DeepSpeed integration in 🤗 Accelerate
LLaMa Tuning with Stanford Alpaca Dataset using Deepspeed and Transformers
Alpaca-lora for huggingface implementation using Deepspeed and FullyShardedDataParallel
Train 🤗transformers with DeepSpeed: ZeRO-2, ZeRO-3
Deepspeed、LLM、Medical_Dialogue、医疗大模型、预训练、微调
Ongoing research training transformer language models at scale, including: BERT & GPT-2
Finetuning LLaMA with RLHF (Reinforcement Learning with Human Feedback) based on DeepSpeed Chat
Implementation of autoregressive language model using improved Transformer and DeepSpeed pipeline parallelism.