Megatron-LM

Megatron-LM

优化GPU训练技术 加速大规模Transformer模型

Megatron-LM框架利用GPU优化技术实现Transformer模型的大规模训练。其Megatron-Core组件提供模块化API和系统优化,支持自定义模型训练。该项目可进行BERT、GPT、T5等模型预训练,支持数千GPU分布式训练百亿参数级模型,并提供数据预处理、模型评估和下游任务功能。

Megatron-LMMegatron-Core大语言模型GPU优化分布式训练Github开源项目
<div align="center">

Megatron-LM & Megatron-Core

<h4>GPU optimized techniques for training transformer models at-scale</h4>

Documentation version license

<div align="left">

Latest News

  • [2024/7] Megatron-Core v0.7 improves scalability and training resiliency and adds support for multimodal training (blog).
  • [2024/6] Megatron-Core added supports for Mamba-based models. Check out our paper An Empirical Study of Mamba-based Language Models and code example.
  • [2024/1 Announcement] NVIDIA has released the core capabilities in Megatron-LM into Megatron-Core in this repository. Megatron-Core expands upon Megatron-LM's GPU-optimized techniques with more cutting-edge innovations on system-level optimizations, featuring composable and modular APIs. Explore the Megatron-Core intro for more details.

Table of Contents

Megatron Overview

This repository comprises two essential components: Megatron-LM and Megatron-Core. Megatron-LM serves as a ressearch-oriented framework leveraging Megatron-Core for large language model (LLM) training. Megatron-Core, on the other hand, is a library of GPU optimized training techniques that comes with formal product support including versioned APIs and regular releases. You can use Megatron-Core alongside Megatron-LM or Nvidia NeMo Framework for an end-to-end and cloud-native solution. Alternatively, you can integrate Megatron-Core's building blocks into your preferred training framework.

Megatron-LM

First introduced in 2019, Megatron (1, 2, and 3) sparked a wave of innovation in the AI community, enabling researchers and developers to utilize the underpinnings of this library to further LLM advancements. Today, many of the most popular LLM developer frameworks have been inspired by and built directly leveraging the open-source Megatron-LM library, spurring a wave of foundation models and AI startups. Some of the most popular LLM frameworks built on top of Megatron-LM include Colossal-AI, HuggingFace Accelerate, and NVIDIA NeMo Framework. A list of projects that have directly used Megatron can be found here.

Megatron-Core

Megatron-Core is an open-source PyTorch-based library that contains GPU-optimized techniques and cutting-edge system-level optimizations. It abstracts them into composable and modular APIs, allowing full flexibility for developers and model researchers to train custom transformers at-scale on NVIDIA accelerated computing infrastructure. This library is compatible with all NVIDIA Tensor Core GPUs, including FP8 acceleration support for NVIDIA Hopper architectures.

Megatron-Core offers core building blocks such as attention mechanisms, transformer blocks and layers, normalization layers, and embedding techniques. Additional functionality like activation recomputation, distributed checkpointing is also natively built-in to the library. The building blocks and functionality are all GPU optimized, and can be built with advanced parallelization strategies for optimal training speed and stability on NVIDIA Accelerated Computing Infrastructure. Another key component of the Megatron-Core library includes advanced model parallelism techniques (tensor, sequence, pipeline, context, and MoE expert parallelism).

Megatron-Core can be used with NVIDIA NeMo, an enterprise-grade AI platform. Alternatively, you can explore Megatron-Core with the native PyTorch training loop here. Visit Megatron-Core documentation to learn more.

Training Speed and Scalability

Our codebase is capable of efficiently training large language models (i.e., models with hundreds of billions of parameters) with both model and data parallelism. To demonstrate how our software scales with multiple GPUs and model sizes, we consider GPT models ranging from 2 billion parameters to 462 billion parameters. All models use a vocabulary size of 131,072 and a sequence length of 4096. We vary hidden size, number of attention heads, and number of layers to arrive at a specific model size. As the model size increases, we also modestly increase batch size. Our experiments use up to 6144 H100 GPUs. We perform fine-grained overlapping of data-parallel (--overlap-grad-reduce --overlap-param-gather), tensor-parallel (--tp-comm-overlap) and pipeline-parallel communication (enabled by default) with computation to improve scalability. The reported throughputs are measured for end-to-end training and include all operations including data loading, optimizer steps, communication, and even logging. Note that we did not train these models to convergence.

Model table

Our weak scaled results show superlinear scaling (MFU increases from 41% for the smallest model considered to 47-48% for the largest models); this is because larger GEMMs have higher arithmetic intensity and are consequently more efficient to execute.

Weak scaling

We also strong scaled the standard GPT-3 model (our version has slightly more than 175 billion parameters due to larger vocabulary size) from 96 H100 GPUs to 4608 GPUs, using the same batch size of 1152 sequences throughout. Communication becomes more exposed at larger scale, leading to a reduction in MFU from 47% to 42%.

Strong scaling

Setup

We strongly recommend using the latest release of NGC's PyTorch container with DGX nodes. If you can't use this for some reason, use the latest pytorch, cuda, nccl, and NVIDIA APEX releases. Data preprocessing requires NLTK, though this is not required for training, evaluation, or downstream tasks.

You can launch an instance of the PyTorch container and mount Megatron, your dataset, and checkpoints with the following Docker commands:

docker pull nvcr.io/nvidia/pytorch:xx.xx-py3
docker run --gpus all -it --rm -v /path/to/megatron:/workspace/megatron -v /path/to/dataset:/workspace/dataset -v /path/to/checkpoints:/workspace/checkpoints nvcr.io/nvidia/pytorch:xx.xx-py3

Downloading Checkpoints

We have provided pretrained BERT-345M and GPT-345M checkpoints to evaluate or for finetuning downstream tasks. To access these checkpoints, first sign up for and setup the NVIDIA GPU Cloud (NGC) Registry CLI. Further documentation for downloading models can be found in the NGC documentation.

Alternatively, you can directly download the checkpoints using:

<pre> BERT-345M-uncased: wget --content-disposition https://api.ngc.nvidia.com/v2/models/nvidia/megatron_bert_345m/versions/v0.1_uncased/zip -O megatron_bert_345m_v0.1_uncased.zip BERT-345M-cased: wget --content-disposition https://api.ngc.nvidia.com/v2/models/nvidia/megatron_bert_345m/versions/v0.1_cased/zip -O megatron_bert_345m_v0.1_cased.zip GPT-345M: wget --content-disposition https://api.ngc.nvidia.com/v2/models/nvidia/megatron_lm_345m/versions/v0.0/zip -O megatron_lm_345m_v0.0.zip </pre>

The models require vocabulary files to run. The BERT WordPiece vocab file can be extracted from Google's pretrained BERT models: uncased, cased. The GPT vocab file and merge table can be downloaded directly.

Usage

After installation, there are several possible workflows. The most comprehensive is:

  1. Data preprocessing
  2. Pretraining
  3. Finetuning (Optional for zero-shot tasks)
  4. Downstream task evaluation or text generation

However, steps 1 and 2 can be replaced by using one of the pretrained models mentioned above.

We've provided several scripts for pretraining both BERT and GPT in the examples directory, as well as scripts for both zero-shot and fine-tuned downstream tasks including MNLI, RACE, WikiText103, and LAMBADA evaluation. There is also a script for GPT interactive text generation.

Training

Data Preprocessing

The training data requires preprocessing. First, place your training data in a loose json format, with one json containing a text sample per line. For example:

<pre> {"src": "www.nvidia.com", "text": "The quick brown fox", "type": "Eng", "id": "0", "title": "First Part"} {"src": "The Internet", "text": "jumps over the lazy dog", "type": "Eng", "id": "42", "title": "Second Part"} </pre>

The name of the text field of the json can be changed by using the --json-key flag in preprocess_data.py The other metadata are optional and are not used in training.

The loose json is then processed into a binary format for training. To convert the json into mmap format use preprocess_data.py. An example script to prepare data for BERT training is:

<pre> python tools/preprocess_data.py \ --input my-corpus.json \ --output-prefix my-bert \ --vocab-file bert-vocab.txt \ --tokenizer-type BertWordPieceLowerCase \ --split-sentences </pre>

The output will be two files named, in this case, my-bert_text_sentence.bin and my-bert_text_sentence.idx. The --data-path specified in later BERT training is the full path and new filename, but without the file extension.

For T5 use the same preprocessing as BERT, perhaps renaming it to:

<pre> --output-prefix my-t5 \ </pre>

Some minor modifications are required for GPT data preprocessing, namely, the addition of a merge table, an end-of-document token, removal of sentence splitting, and a change to the tokenizer type:

<pre> python tools/preprocess_data.py \ --input my-corpus.json \ --output-prefix my-gpt2 \ --vocab-file gpt2-vocab.json \ --tokenizer-type GPT2BPETokenizer \ --merge-file gpt2-merges.txt \ --append-eod </pre>

Here the output files are named my-gpt2_text_document.bin and my-gpt2_text_document.idx. As before, in GPT training, use the longer name without the extension as --data-path.

Further command line arguments are described in the source file preprocess_data.py.

BERT Pretraining

The examples/pretrain_bert.sh script runs single GPU 345M parameter BERT pretraining. Debugging is the primary use for single GPU training, as the code base and command line arguments are optimized for highly distributed training. Most of the arguments are fairly self-explanatory. By default, the learning rate decays linearly over the training iterations starting at --lr to a minimum set by --min-lr over --lr-decay-iters iterations. The fraction of training iterations used for warmup is set by --lr-warmup-fraction. While this is single GPU training, the batch size specified by --micro-batch-size is a single forward-backward path batch-size and the code will perform gradient accumulation steps until it reaches global-batch-size which is the batch size per iteration. The data is partitioned into a 949:50:1 ratio for training/validation/test sets (default is 969:30:1). This partitioning happens on the fly, but is consistent across runs with the same random seed (1234 by default, or specified manually with --seed). We use train-iters as the training iterations requested. Alternatively, one can provide --train-samples which is total number of samples to train on. If this option is present, then instead of providing --lr-decay-iters, one will need to provide --lr-decay-samples.

The logging, checkpoint-saving, and evaluation interval options are specified. Note that the --data-path now includes the additional _text_sentence suffix added in preprocessing, but does not include the file extensions.

Further command line arguments are described in the source file arguments.py.

To run examples/pretrain_bert.sh, make any desired modifications including setting the environment variables for CHECKPOINT_PATH, VOCAB_FILE, and DATA_PATH. Make sure to set these variables to their paths in the container. Then launch the container with Megatron and necessary paths mounted (as explained in Setup) and run the example script.

GPT Pretraining

The examples/pretrain_gpt.sh script runs single GPU 345M parameter GPT pretraining. As mentioned above, single GPU training is primarily intended for debugging purposes, as the code is optimized for distributed training.

It follows largely the same format as the previous BERT script with a few notable differences: the tokenization scheme used is BPE (which requires a merge table and a json vocabulary file) instead of WordPiece, the model architecture allows for longer sequences (note that the max position embedding must be greater than or equal to the maximum sequence length), and the --lr-decay-style has been set to cosine decay. Note that the --data-path now includes the additional _text_document suffix added in preprocessing, but does not include the file extensions.

Further command line arguments are described in the source file arguments.py.

examples/pretrain_gpt.sh can be launched the same way as described for BERT. Set the env vars and make any other modifications, launch the container with appropriate mounts, and run the script.

T5 Pretraining

Very similar to

编辑推荐精选

AEE

AEE

AI Excel全自动制表工具

AEE 在线 AI 全自动 Excel 编辑器,提供智能录入、自动公式、数据整理、图表生成等功能,高效处理 Excel 任务,提升办公效率。支持自动高亮数据、批量计算、不规则数据录入,适用于企业、教育、金融等多场景。

UI-TARS-desktop

UI-TARS-desktop

基于 UI-TARS 视觉语言模型的桌面应用,可通过自然语言控制计算机进行多模态操作。

UI-TARS-desktop 是一款功能强大的桌面应用,基于 UI-TARS(视觉语言模型)构建。它具备自然语言控制、截图与视觉识别、精确的鼠标键盘控制等功能,支持跨平台使用(Windows/MacOS),能提供实时反馈和状态显示,且数据完全本地处理,保障隐私安全。该应用集成了多种大语言模型和搜索方式,还可进行文件系统操作。适用于需要智能交互和自动化任务的场景,如信息检索、文件管理等。其提供了详细的文档,包括快速启动、部署、贡献指南和 SDK 使用说明等,方便开发者使用和扩展。

Wan2.1

Wan2.1

开源且先进的大规模视频生成模型项目

Wan2.1 是一个开源且先进的大规模视频生成模型项目,支持文本到图像、文本到视频、图像到视频等多种生成任务。它具备丰富的配置选项,可调整分辨率、扩散步数等参数,还能对提示词进行增强。使用了多种先进技术和工具,在视频和图像生成领域具有广泛应用前景,适合研究人员和开发者使用。

爱图表

爱图表

全流程 AI 驱动的数据可视化工具,助力用户轻松创作高颜值图表

爱图表(aitubiao.com)就是AI图表,是由镝数科技推出的一款创新型智能数据可视化平台,专注于为用户提供便捷的图表生成、数据分析和报告撰写服务。爱图表是中国首个在图表场景接入DeepSeek的产品。通过接入前沿的DeepSeek系列AI模型,爱图表结合强大的数据处理能力与智能化功能,致力于帮助职场人士高效处理和表达数据,提升工作效率和报告质量。

Qwen2.5-VL

Qwen2.5-VL

一款强大的视觉语言模型,支持图像和视频输入

Qwen2.5-VL 是一款强大的视觉语言模型,支持图像和视频输入,可用于多种场景,如商品特点总结、图像文字识别等。项目提供了 OpenAI API 服务、Web UI 示例等部署方式,还包含了视觉处理工具,有助于开发者快速集成和使用,提升工作效率。

HunyuanVideo

HunyuanVideo

HunyuanVideo 是一个可基于文本生成高质量图像和视频的项目。

HunyuanVideo 是一个专注于文本到图像及视频生成的项目。它具备强大的视频生成能力,支持多种分辨率和视频长度选择,能根据用户输入的文本生成逼真的图像和视频。使用先进的技术架构和算法,可灵活调整生成参数,满足不同场景的需求,是文本生成图像视频领域的优质工具。

WebUI for Browser Use

WebUI for Browser Use

一个基于 Gradio 构建的 WebUI,支持与浏览器智能体进行便捷交互。

WebUI for Browser Use 是一个强大的项目,它集成了多种大型语言模型,支持自定义浏览器使用,具备持久化浏览器会话等功能。用户可以通过简洁友好的界面轻松控制浏览器智能体完成各类任务,无论是数据提取、网页导航还是表单填写等操作都能高效实现,有利于提高工作效率和获取信息的便捷性。该项目适合开发者、研究人员以及需要自动化浏览器操作的人群使用,在 SEO 优化方面,其关键词涵盖浏览器使用、WebUI、大型语言模型集成等,有助于提高网页在搜索引擎中的曝光度。

xiaozhi-esp32

xiaozhi-esp32

基于 ESP32 的小智 AI 开发项目,支持多种网络连接与协议,实现语音交互等功能。

xiaozhi-esp32 是一个极具创新性的基于 ESP32 的开发项目,专注于人工智能语音交互领域。项目涵盖了丰富的功能,如网络连接、OTA 升级、设备激活等,同时支持多种语言。无论是开发爱好者还是专业开发者,都能借助该项目快速搭建起高效的 AI 语音交互系统,为智能设备开发提供强大助力。

olmocr

olmocr

一个用于 OCR 的项目,支持多种模型和服务器进行 PDF 到 Markdown 的转换,并提供测试和报告功能。

olmocr 是一个专注于光学字符识别(OCR)的 Python 项目,由 Allen Institute for Artificial Intelligence 开发。它支持多种模型和服务器,如 vllm、sglang、OpenAI 等,可将 PDF 文件的页面转换为 Markdown 格式。项目还提供了测试框架和 HTML 报告生成功能,方便用户对 OCR 结果进行评估和分析。适用于科研、文档处理等领域,有助于提高工作效率和准确性。

飞书多维表格

飞书多维表格

飞书多维表格 ×DeepSeek R1 满血版

飞书多维表格联合 DeepSeek R1 模型,提供 AI 自动化解决方案,支持批量写作、数据分析、跨模态处理等功能,适用于电商、短视频、影视创作等场景,提升企业生产力与创作效率。关键词:飞书多维表格、DeepSeek R1、AI 自动化、批量处理、企业协同工具。

下拉加载更多