Commits

Breadcrumbs

History for

onmain

User selector

All users

Datepicker

All time

Commit History

Commits on May 2, 2024

• [MLPerf] GPT dataset features: drop last partial validation sequence, drop extra token, return sample with 1s loss mask, mock dataset testing

  

  BoxiangW

  authored and

  ericharper

  committed
  c90aa16

  Copy full SHA for c90aa16

Commits on Apr 17, 2024

• Support missing size and missing weights in dataset creation and blending

  

  John Kamalu

  authored and

  ericharper

  committed
  dc52e84

  Copy full SHA for dc52e84

Commits on Mar 29, 2024

• Bug fix (Parallel output should be set to false, so that we gather the output after the last stage ).

  shanmugamr1992

  committed
  5e22048

  Copy full SHA for 5e22048

Commits on Mar 27, 2024

• Retro example script update.

  

  lmcafee-nvidia

  authored and

  jaredcasper

  committed
  baf9e53

  Copy full SHA for baf9e53

Commits on Mar 26, 2024

• Simple bug fix

  

  Shanmugam Ramasamy

  committed
  45fcea7

  Copy full SHA for 45fcea7

• Merge branch 'jcasper/refactor' into 'main'

  Show description for da0afe0
  jaredcasper

  committed
  da0afe0

  Copy full SHA for da0afe0

• Refactor everything outside of core to be out of the main megatron. namespace.

  jaredcasper

  committed
  38644dd

  Copy full SHA for 38644dd

• move is_built_on_rank from config to builder

  

  John Kamalu

  authored and

  ericharper

  committed
  9475bab

  Copy full SHA for 9475bab

Commits on Mar 22, 2024

• Retro M-core Data

  

  lmcafee-nvidia

  authored and

  shanmugamr1992

  committed
  e89cce4

  Copy full SHA for e89cce4

Commits on Mar 13, 2024

• Addresssing Jared and Erics comments

  

  Shanmugam Ramasamy

  committed
  bdf1b5e

  Copy full SHA for bdf1b5e

Commits on Mar 11, 2024

• Mcore pretrain vision language model

  

  trintamaki

  authored and

  ericharper

  committed
  d144db4

  Copy full SHA for d144db4

Commits on Mar 5, 2024

• Merge branch 'yaml' into 'main'

  Show description for 8957468
  jaredcasper

  committed
  8957468

  Copy full SHA for 8957468

• Experimental Yaml configs

  

  wdykas

  authored and

  jaredcasper

  committed
  47cb630

  Copy full SHA for 47cb630

Commits on Mar 1, 2024

• [OMNIML-614] AMMO ptq + TensorRT-LLM export examples for megatron-lm

  

  ChenhanYu

  authored and

  jaredcasper

  committed
  80e180d

  Copy full SHA for 80e180d

Commits on Feb 15, 2024

• Fixing examples

  

  Shanmugam Ramasamy

  committed
  1b6ae27

  Copy full SHA for 1b6ae27

Commits on Feb 1, 2024

• Fixing bugs in inference and adding mcore support

  shanmugamr1992

  committed
  d91c5a6

  Copy full SHA for d91c5a6

• Fixing bugs in inference and adding mcore support

  shanmugamr1992

  committed
  376337d

  Copy full SHA for 376337d

• Fixing bugs in inference and adding mcore support

  shanmugamr1992

  committed
  7329f73

  Copy full SHA for 7329f73

• Fixing bugs in inference and adding mcore support

  shanmugamr1992

  committed
  50f8384

  Copy full SHA for 50f8384

Commits on Nov 17, 2023

• change image readme size

  huvunvidia

  committed
  5677267

  Copy full SHA for 5677267

• addressing Eric's comments

  huvunvidia

  committed
  dabfe1f

  Copy full SHA for dabfe1f

Commits on Nov 16, 2023

• Merge remote-tracking branch 'origin/lmcafee/retro-mcore' into huvu/t5

  huvunvidia

  committed
  974d0fa

  Copy full SHA for 974d0fa

Commits on Nov 4, 2023

• address Lawrence's comments

  huvunvidia

  committed
  26edc85

  Copy full SHA for 26edc85

Commits on Nov 2, 2023

• Merge branch 'main' into bert_refactor

  

  Shanmugam Ramasamy

  committed
  3794ed2

  Copy full SHA for 3794ed2

Commits on Nov 1, 2023

• fix examples

  wdykas

  committed
  244c8b4

  Copy full SHA for 244c8b4

Commits on Oct 30, 2023

• Addressed jared's comments

  

  Shanmugam Ramasamy

  committed
  fa56f2b

  Copy full SHA for fa56f2b

Commits on Oct 29, 2023

• update T5 to use methods from common

  huvunvidia

  committed
  c426f19

  Copy full SHA for c426f19

Commits on Oct 26, 2023

• Refactor dataset code and move to core

  

  John Kamalu

  authored and

  jaredcasper

  committed
  a0e190c

  Copy full SHA for a0e190c

Commits on Oct 22, 2023

• commit before push to huy_megatron:huvu/t5 oct21

  huvunvidia

  committed
  f010997

  Copy full SHA for f010997

Commits on Oct 21, 2023

• reverse chmod 100755 to 100644

  huvunvidia

  committed
  4076ab3

  Copy full SHA for 4076ab3

Commits on Oct 20, 2023

• unit tests and functional tests

  huvunvidia

  committed
  d2c5e8b

  Copy full SHA for d2c5e8b

Commits on Oct 19, 2023

• Refactoring bert

  shanmugamr1992

  committed
  2f3afc7

  Copy full SHA for 2f3afc7

• Addressing Jared's comments

  shanmugamr1992

  committed
  68ee266

  Copy full SHA for 68ee266

Commits on Oct 18, 2023

• Update train_gpt3_175b_distributed.sh

  shanmugamr1992

  committed
  705ba1f

  Copy full SHA for 705ba1f

• Refactoring gpt3 examples

  shanmugamr1992

  committed
  be32366

  Copy full SHA for be32366

Pagination

Previous

Next