Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Jan 23, 2025

• Merge branch 'inference_api_simplifications' into 'main'

  Show description for 2167226
  jaredcasper

  committed
  2167226

  Copy full SHA for 2167226

• ADLR/megatron-lm!2567 - Add prep_batch_for_inference_input function and make inference_input a dict

  

  santhnm2

  authored and

  jaredcasper

  committed
  a407351

  Copy full SHA for a407351

• Merge branch 'ko3n1g/ci/add-max-memory-to-checks' into 'main'

  Show description for 9e8690f
  ko3n1g

  committed
  9e8690f

  Copy full SHA for 9e8690f

• ADLR/megatron-lm!2577 - ci: Enable checks for mem-max-allocated-bytes

  ko3n1g

  committed
  46fdcd5

  Copy full SHA for 46fdcd5

Commits on Jan 22, 2025

• Merge branch 'fix_1d_flatten_tensor' into 'main'

  Show description for 0d59157
  ko3n1g

  committed
  0d59157

  Copy full SHA for 0d59157

• ADLR/megatron-lm!2468 - Handle 1d flatten shard-tensor edge-case

  Show description for 244ec97
  

  shjwudp

  authored and

  ko3n1g

  committed
  244ec97

  Copy full SHA for 244ec97

• Merge branch 'ko3n1g/fix/setup' into 'main'

  Show description for 5c12382
  ko3n1g

  committed
  5c12382

  Copy full SHA for 5c12382

• ADLR/megatron-lm!2579 - fix: Setup requirements

  ko3n1g

  committed
  4df1108

  Copy full SHA for 4df1108

• Merge branch 'feature/wandb_artifact_log_checkpoint' into 'main'

  Show description for 27756f4
  ko3n1g

  committed
  27756f4

  Copy full SHA for 27756f4

• ADLR/megatron-lm!2575 - Feature/wandb artifact log checkpoint

  Show description for e2a0e9e
  

  Talor Abramovich

  authored and

  ko3n1g

  committed
  e2a0e9e

  Copy full SHA for e2a0e9e

• Merge branch 'ko3n1g/chore/bump-pyt-24.10' into 'main'

  Show description for 66e5306
  ko3n1g

  committed
  66e5306

  Copy full SHA for 66e5306

• ADLR/megatron-lm!2299 - chore: Bump PyT to 24.10

  ko3n1g

  committed
  ffdb6dc

  Copy full SHA for ffdb6dc

• Merge branch 'add_llama_support' into 'main'

  Show description for ae1c43d
  jon-barker

  committed
  ae1c43d

  Copy full SHA for ae1c43d

• ADLR/megatron-lm!2550 - Add llama 3.1 support for mmodal example

  Show description for 2b61030
  

  2 peopleauthored and

  jon-barker

  committed
  2b61030

  Copy full SHA for 2b61030

• Merge branch 'mike/pp_fix_1' into 'main'

  Show description for 564dbd7
  ko3n1g

  committed
  564dbd7

  Copy full SHA for 564dbd7

• ADLR/megatron-lm!2569 - Bug fix in get_pipeline_model_parallel_last_rank

  Show description for cf1b0d4
  

  2 peopleauthored and

  ko3n1g

  committed
  cf1b0d4

  Copy full SHA for cf1b0d4

• Merge branch 'ko3n1g/ci/catch-log-error' into 'main'

  Show description for 33de8a5
  ko3n1g

  committed
  33de8a5

  Copy full SHA for 33de8a5

• ADLR/megatron-lm!2578 - ci: Catch UnicodeDecodeError

  ko3n1g

  committed
  dc251d7

  Copy full SHA for dc251d7

• Merge branch 'matthieul/fail_on_missing_token' into 'main'

  Show description for 4064396
  jon-barker

  committed
  4064396

  Copy full SHA for 4064396

• ADLR/megatron-lm!2568 - Assert image token exists in multimodal example

  

  Matthieu Le

  authored and

  jon-barker

  committed
  f8e4d27

  Copy full SHA for f8e4d27

• Merge branch 'fix_moe_drop_and_pad' into 'main'

  Show description for f73f20c
  jaredcasper

  committed
  f73f20c

  Copy full SHA for f73f20c

• ADLR/megatron-lm!2532 - Fix bug in !2426

  

  guyueh1

  authored and

  jaredcasper

  committed
  9bbb40d

  Copy full SHA for 9bbb40d

Commits on Jan 21, 2025

• Merge branch 'ko3n1g/ci/write-max-allocated-mem-to-tensorboard' into 'main'

  Show description for f8887ce
  ko3n1g

  committed
  f8887ce

  Copy full SHA for f8887ce

• ADLR/megatron-lm!2576 - feat: Log max-allocated-mem to TB

  ko3n1g

  committed
  f950178

  Copy full SHA for f950178

• Merge branch 'add_pg_desc' into 'main'

  Show description for df70c00
  ko3n1g

  committed
  df70c00

  Copy full SHA for df70c00

• ADLR/megatron-lm!2513 - add group_desc when invoking new_group()

  

  sanshang-nv

  authored and

  ko3n1g

  committed
  e8336b1

  Copy full SHA for e8336b1

Commits on Jan 20, 2025

• Merge branch 'dnarayanan/speed_and_functional_tests' into 'main'

  Show description for 7dd2658
  ko3n1g

  committed
  7dd2658

  Copy full SHA for 7dd2658

• ADLR/megatron-lm!2572 - New GPT memory and speed tests

  Show description for f29bf42
  

  deepakn94

  and

  ko3n1g

  committed
  f29bf42

  Copy full SHA for f29bf42

• Merge branch 'ko3n1g/chore/bump-versions' into 'main'

  Show description for 4fb4c3d
  ko3n1g

  committed
  4fb4c3d

  Copy full SHA for 4fb4c3d

• ADLR/megatron-lm!2510 - chore: Bump versions

  ko3n1g

  committed
  9c11ab4

  Copy full SHA for 9c11ab4

Commits on Jan 18, 2025

• Merge branch 'ko3n1g/ci/ci-output' into 'main'

  Show description for 7ba0d6d
  ko3n1g

  committed
  7ba0d6d

  Copy full SHA for 7ba0d6d

• ADLR/megatron-lm!2571 - ci: Better output

  ko3n1g

  committed
  57c392b

  Copy full SHA for 57c392b

Commits on Jan 17, 2025

• Merge branch 'dpykhtar/fix_load_ckpt' into 'main'

  Show description for c7bf403
  jaredcasper

  committed
  c7bf403

  Copy full SHA for c7bf403

• ADLR/megatron-lm!2555 - set weight_only to False

  

  dimapihtar

  authored and

  jaredcasper

  committed
  37a900f

  Copy full SHA for 37a900f

• Merge branch 'ko3n1g/refactor/make-get_mlp_module_spec-public' into 'main'

  Show description for fa35226
  ko3n1g

  committed
  fa35226

  Copy full SHA for fa35226

Pagination

Previous

Next