Code

Embedding#1209

Code

Open

rachitgarg91 wants to merge 51 commits intoNVIDIA:24.09-betaNVIDIA/Megatron-LM:24.09-betafrom rachitgarg91:embeddingrachitgarg91/Megatron-LM:embeddingCopy head branch name to clipboard

+8,884-1,647

Commits

Commits on Sep 19, 2024

• ADLR/megatron-lm!2033 - Online eval

  

  trintamaki

  authored and

  ericharper

  committed
  63be779

Commits on Oct 8, 2024

• ADLR/megatron-lm!1973 - MMMU multi-image support

  

  trintamaki

  authored and

  Rachit Garg

  committed
  69d4c44

• ADLR/megatron-lm!2113 - build: Use multi-stage for parallel builds

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  7754f56

• ADLR/megatron-lm!2126 - Only print warning when relevant

  

  deepakn94

  authored and

  Rachit Garg

  committed
  121d05e

• ADLR/megatron-lm!2124 - tests: Fix location of megatron

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  3a0ca4b

• ADLR/megatron-lm!2127 - ci: Bump sha

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  ca219ed

• ADLR/megatron-lm!2128 - ci: Improve cherry pick workflow

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  1899bb7

• ADLR/megatron-lm!2034 - ci: Introduce JET Python SDK

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  26f5c32

• ADLR/megatron-lm!2130 - ci: Improve cherry pick MR description

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  d626aeb

• ADLR/megatron-lm!2119 - Huvu/t5 te10 fix nemoci pr482

  Show description for ea83faa
  

  2 peopleauthored and

  Rachit Garg

  committed
  ea83faa

• ADLR/megatron-lm!2134 - ci: Set author and milestone for cherry-picks

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  ae1bffb

• ADLR/megatron-lm!2135 - ci: Send alerts on unit-tests-extended

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  baad0ad

• ADLR/megatron-lm!2133 - tests: Minor improvements to JET

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  460c6a9

• ADLR/megatron-lm!2136 - tests: Fix GPT test

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  a0799f4

• ADLR/megatron-lm!2139 - ci: Fix cherry-pick strings

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  4c3e06a

• ADLR/megatron-lm!2110 - Use torch dataloader in multimodal evaluation

  

  trintamaki

  authored and

  Rachit Garg

  committed
  71b2aa1

• ADLR/megatron-lm!2137 - ci: Enable dev container for new features

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  3ab6da9

• ADLR/megatron-lm!2005 - Fix performance regression brought by torch.bincount

  

  xxuwenc

  authored and

  Rachit Garg

  committed
  ff89e91

• ADLR/megatron-lm!2073 - Multimodal batched bug fix

  

  trintamaki

  authored and

  Rachit Garg

  committed
  6543004

• ADLR/megatron-lm!1581 - Add MLA support into MCore

  Show description for 29793cf
  

  2 peopleauthored and

  Rachit Garg

  committed
  29793cf

• ADLR/megatron-lm!1995 - Add freeze options to pretrain_vlm

  

  trintamaki

  authored and

  Rachit Garg

  committed
  754e0f0

• ADLR/megatron-lm!2145 - Improve logging when decreasing batch size

  

  deepakn94

  authored and

  Rachit Garg

  committed
  10350b6

• ADLR/megatron-lm!2148 - Add model.eval() to run_text_generation_server.py

  

  mathemakitten

  authored and

  Rachit Garg

  committed
  f54686a

• ADLR/megatron-lm!2111 - Mcore llama3.1 support

  Show description for b301e5f
  

  2 peopleauthored and

  Rachit Garg

  committed
  b301e5f

• ADLR/megatron-lm!2151 - ci: Run experimental UTs on dev image

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  dcdf804

• ADLR/megatron-lm!1953 - Mcore export to export models to TRTLLM (GPU and CPU version)

  Show description for 32b395e
  

  3 peopleauthored and

  Rachit Garg

  committed
  32b395e

• ADLR/megatron-lm!2154 - ci: Prune docker cache of mcore-docker-node-jet

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  8fbb30c

• ADLR/megatron-lm!2155 - Resolve release test failure caused by GroupedMLP distributed checkpointing

  

  xxuwenc

  authored and

  Rachit Garg

  committed
  9f06f06

• ADLR/megatron-lm!2156 - tests: Set better name for Wandb logging

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  015bffc

• ADLR/megatron-lm!1950 - Remove pkg_resources package

  Show description for 34f0f98
  

  3 peopleauthored and

  Rachit Garg

  committed
  34f0f98

• ADLR/megatron-lm!2142 - ci: Onboard CW

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  8103c4c

• ADLR/megatron-lm!2158 - Small changes to export

  Show description for 30aafee
  

  3 peopleauthored and

  Rachit Garg

  committed
  30aafee

• ADLR/megatron-lm!2152 - Fix rope backward compatibility

  

  BoxiangW

  authored and

  Rachit Garg

  committed
  30445f8

• ADLR/megatron-lm!2140 - [Bug fix] Don't trace graphs during inference

  

  jiemingz

  authored and

  Rachit Garg

  committed
  c6a0ec8

• ADLR/megatron-lm!2109 - Adding more MR tests for T5 (e.g., transformer_engine, distributed_checkpoint)

  Show description for d90956c
  

  2 peopleauthored and

  Rachit Garg

  committed
  d90956c

• ADLR/megatron-lm!2164 - ci: Download artifacts

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  77f62d8

• ADLR/megatron-lm!2165 - ci: Bump version

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  cf25e49

• ADLR/megatron-lm!2153 - Add the interface to set TP communication bootstrap backend

  Show description for a792575
  

  2 peopleauthored and

  Rachit Garg

  committed
  a792575

• ADLR/megatron-lm!2095 - Add support for SigLIP vision encoder to multimodal mcore

  

  Matthieu Le

  authored and

  Rachit Garg

  committed
  74d4b1a

• ADLR/megatron-lm!2175 - adding cu_seqlens_padded support in MCore

  Show description for d819b9c
  

  3 peopleauthored and

  Rachit Garg

  committed
  d819b9c

• ADLR/megatron-lm!2181 - Fixing attention mask dimenions to support TE versions > 1.9

  Show description for 4ded7ce
  

  2 peopleauthored and

  Rachit Garg

  committed
  4ded7ce

• ADLR/megatron-lm!2180 - rotary_scaling fix for llama3.1 and 3.2

  

  yueshen2016

  authored and

  Rachit Garg

  committed
  c302697

• ADLR/megatron-lm!2185 - chore: Improve generator for launch scripts

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  7619780

• ADLR/megatron-lm!2160 - Adding Inference pipeline for T5

  Show description for 52699ce
  

  3 peopleauthored and

  Rachit Garg

  committed
  52699ce

• ADLR/megatron-lm!2182 - ci: Group runs by model

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  691b323

• ADLR/megatron-lm!1862 - Cpu init te

  Show description for ce67659
  

  4 peopleauthored and

  Rachit Garg

  committed
  ce67659

• ADLR/megatron-lm!2186 - ci: Run script after export

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  73ef715

• ADLR/megatron-lm!2089 - Fix upcycling issues.

  

  RayWang96

  authored and

  Rachit Garg

  committed
  329c1d7

• ADLR/megatron-lm!2189 - tests: Fix ENV export

  

  ko3n1g

  authored and

  Rachit Garg

  committed
  99f63e8

• rebase to 24.09

  

  Rachit Garg

  committed
  fd20cda

• Merge branch 'embedding' of https://github.com/rachitgarg91/Megatron-LM into embedding

  

  Rachit Garg

  committed
  93dab11