Initial commit

.gitignore

+.idea
+Megatron-LM-*
+LM-Evaluation-Harness*
+Bigcode-Evaluation-Harness*
+**/__pycache__
+.vscode

.gitmodules

+[submodule "LM-Evaluation-Harness-240310"]
+	path = LM-Evaluation-Harness-240310
+	url = https://github.com/jerryli1981/lm-evaluation-harness.git
+[submodule "Bigcode-Evaluation-Harness-240327"]
+	path = Bigcode-Evaluation-Harness-240327
+	url = https://github.com/jerryli1981/bigcode-evaluation-harness
+[submodule "Megatron-LM-240405"]
+	path = Megatron-LM-240405
+	url = https://github.com/NVIDIA/Megatron-LM.git
+[submodule "Megatron-LM-231007"]
+	path = Megatron-LM-231007
+	url = https://github.com/NVIDIA/Megatron-LM.git
+[submodule "Megatron-LM-240126"]
+	path = Megatron-LM-240126
+	url = https://github.com/NVIDIA/Megatron-LM.git
+[submodule "Megatron-LM-MegaBlocks"]
+	path = Megatron-LM-MegaBlocks
+	url = https://github.com/jerryli1981/Megatron-LM-MegaBlocks
+[submodule "Megatron-LM-240612"]
+	path = Megatron-LM-240612
+	url = https://github.com/NVIDIA/Megatron-LM.git
+[submodule "PAI-Megatron-LM-240718"]
+	path = PAI-Megatron-LM-240718
+	url = https://github.com/jerryli1981/PAI-Megatron-LM-240718
+[submodule "Megatron-LM-241113"]
+	path = Megatron-LM-241113
+	url = https://github.com/NVIDIA/Megatron-LM.git

LICENSE

+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

NOTICE

+=============================================================
+pai-megatron-patch is an open source tool developed by Alibaba PAI Team
+Copyright (c) 2022-2023, Alibaba Group Holding Limited.
+Licensed under the Apache License, Version 2.0
+
+=============================================================
+This toolkit implements some modules referring to some repositories under
+the same/different open source licenses
+
+-----------------------------
+model, data, tokenization
+Apache License, Version 2.0
+Copyright (c)  The HuggingFace Inc. team
+
+------------------
+model, toolkits, generation, tokenization
+Apache License, Version 2.0
+The Megatron-LM Team Authors
+The Transformer Engine Team Authors

PAI-Megatron-LM-240718/.coveragerc

+[html]
+directory = coverage
+
+[run]
+data_file = .coverage_$LOCAL_RANK

PAI-Megatron-LM-240718/.github/ISSUE_TEMPLATE/bug.md

+---
+name: BUG
+about: Report a bug that needs attention
+title: "[BUG]"
+labels: ''
+assignees: ''
+
+---
+
+**Describe the bug**
+A clear and concise description of what the bug is.
+
+**To Reproduce**
+Steps to reproduce the behavior. The easier it is to reproduce the faster it will get maintainer attention.
+
+**Expected behavior**
+A clear and concise description of what you expected to happen.
+
+**Stack trace/logs**
+If applicable, add the stack trace or logs from the time of the error.
+
+**Environment (please complete the following information):**
+ - Megatron-LM commit ID
+ - PyTorch version
+ - CUDA version
+ - NCCL version
+
+**Proposed fix**
+If you have a proposal for how to fix the issue state it here or link to a PR.
+
+**Additional context**
+Add any other context about the problem here.

PAI-Megatron-LM-240718/.github/ISSUE_TEMPLATE/enhancement.md

+---
+name: ENHANCEMENT
+about: Suggest an idea to improve this project
+title: "[ENHANCEMENT]"
+labels: ''
+assignees: ''
+
+---
+
+**Is your feature request related to a problem? Please describe.**
+A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
+
+**Describe the solution you'd like**
+A clear and concise description of what you want to happen.
+
+**Describe alternatives you've considered**
+A clear and concise description of any alternative solutions or features you've considered.
+
+**Proposed implementation**
+If you have a proposed implementation for the feature state it here or link to a PR.
+
+**Additional context**
+Add any other context or screenshots about the feature request here.

PAI-Megatron-LM-240718/.github/ISSUE_TEMPLATE/question.md

+---
+name: QUESTION
+about: Ask a question about Megatron-LM that is not a bug, regression or enhancement
+  request
+title: "[QUESTION]"
+labels: ''
+assignees: ''
+
+---
+
+**Your question**
+Ask a clear and concise question about Megatron-LM.

PAI-Megatron-LM-240718/.github/ISSUE_TEMPLATE/regression.md

+---
+name: REGRESSION
+about: Report a regression in speed or accuracy due to a Megatron-LM update
+title: "[REGRESSION]"
+labels: ''
+assignees: ''
+
+---
+
+**Describe the regression**
+A clear and concise description of what the regression is.
+
+**To Reproduce**
+Steps to reproduce the behavior. The easier it is to reproduce the faster it will get maintainer attention.
+
+**Previous performance**
+What speed or accuracy did you previously see.
+
+**New performance**
+What speed or accuracy do you see after the update.
+
+**Stack trace/logs**
+If applicable, add the stack trace or logs related to the regression.
+
+**Environment (please complete the following information):**
+ - Previous Megatron-LM commit ID
+ - New Megatron-LM commit ID
+ - Previous PyTorch version
+ - New PyTorch version
+ - Previous CUDA version
+ - New CUDA version
+ - Previous NCCL version
+ - New NCCL version
+
+**Proposed fix**
+If you have a proposal for how to fix the issue state it here or link to a PR.
+
+**Additional context**
+Add any other context about the problem here.

PAI-Megatron-LM-240718/.github/workflows/stale.yml

+# This workflow warns and then closes issues and PRs that have had no activity for a specified amount of time.
+#
+# You can adjust the behavior by modifying this file.
+# For more information, see:
+# https://github.com/actions/stale
+name: Mark stale issues and pull requests
+
+on:
+  schedule:
+  - cron: '15 18 * * *'
+
+jobs:
+  stale:
+
+    runs-on: ubuntu-latest
+    permissions:
+      issues: write
+      pull-requests: write
+
+    steps:
+    - uses: actions/stale@v5
+      with:
+        repo-token: ${{ secrets.GITHUB_TOKEN }}
+        days-before-stale: 60
+        stale-issue-message: 'Marking as stale. No activity in 60 days.'
+        stale-pr-message: 'Marking as stale. No activity in 60 days.'
+        stale-issue-label: 'stale'
+        stale-pr-label: 'stale'
+        remove-stale-when-updated: true
+        operations-per-run: 1000
+        days-before-close: -1

PAI-Megatron-LM-240718/.gitignore

+__pycache__
+*.so
+build
+.coverage_*
+*.egg-info
+*~
+slurm*
+logs
+.vscode
+local/
\ No newline at end of file

PAI-Megatron-LM-240718/.gitlab-ci.yml

+workflow:
+  rules:
+    - if: $CI_PIPELINE_SOURCE == "schedule"
+      variables:
+        FUNCTIONAL_TEST: "yes"
+    - if: $CI_PIPELINE_SOURCE == "web"
+    - if: $CI_COMMIT_BRANCH == $CI_DEFAULT_BRANCH
+      variables:
+        FUNCTIONAL_TEST: "no"
+    - if: $CI_COMMIT_BRANCH =~ /^core_r/
+      variables:
+        FUNCTIONAL_TEST: "no"
+    - if: $CI_MERGE_REQUEST_LABELS =~ /Run tests and nightly/
+      variables:
+        FUNCTIONAL_TEST: "yes"
+        SLURM_CLUSTER: dgxa100_dracooci
+        SCOPE: mr-and-nightly
+    - if: $CI_MERGE_REQUEST_LABELS =~ /Run tests/
+      variables:
+        FUNCTIONAL_TEST: "yes"
+        SLURM_CLUSTER: dgxa100_dracooci
+        SCOPE: mr
+    - if: $CI_PIPELINE_SOURCE == "merge_request_event"
+      variables:
+        FUNCTIONAL_TEST: "no"
+    - when: never
+  auto_cancel:
+    on_new_commit: interruptible
+
+stages:
+  - build
+  - unit_tests
+  - functional_tests
+
+default:
+  interruptible: true
+
+variables:
+  FUNCTIONAL_TEST: "yes"
+  SCOPE:
+    value: "mr"
+    options:
+      - "mr"
+      - "nightly"
+      - "mr-and-nightly"
+      - "weekly"
+      - "release"
+    description: "Testsuite to run"
+  SLURM_CLUSTER:
+    value: "dgxa100_dracooci"
+    options:
+      - "dgxa100_dracooci"
+      - "dgxh100_eos"
+    description: '"dgxa100_dracooci" for OCI-IAD, "dgxh100_eos" for EOS'
+  # CI wide variables
+  CI_MCORE_IMAGE: gitlab-master.nvidia.com:5005/adlr/megatron-lm/mcore_ci
+  CI_NEMO_IMAGE: gitlab-master.nvidia.com:5005/adlr/megatron-lm/nemo_ci
+  LINTING_IMAGE: gitlab-master.nvidia.com:5005/adlr/megatron-lm/mcore_linting
+
+metadata:
+  image: python:3.10
+  stage: .pre
+  tags:
+    - os/linux
+  script:
+    - set -x
+    - env
+    - JET_CUSTOM_FILTER="type == 'basic'"
+    - |
+      if [[ $SLURM_CLUSTER == dgxh100_eos ]]; then
+        JET_CI_BRANCH=mcore/eos
+        JET_CUSTOM_FILTER="$JET_CUSTOM_FILTER and 'dgx_h100' in spec.platforms"
+      elif [[ $SLURM_CLUSTER == dgxa100_dracooci ]]; then
+        JET_CI_BRANCH=mcore/draco-oci
+        JET_CUSTOM_FILTER="$JET_CUSTOM_FILTER and 'dgx_a100' in spec.platforms"
+      fi
+    - |
+      if [[ $SCOPE == mr ]]; then
+        JET_CUSTOM_FILTER="$JET_CUSTOM_FILTER and 'mr' in spec.scope"
+      elif [[ $SCOPE == nightly ]]; then
+        JET_CUSTOM_FILTER="$JET_CUSTOM_FILTER and 'nightly' in spec.scope"
+      elif [[ $SCOPE == mr-and-nightly ]]; then
+        JET_CUSTOM_FILTER="$JET_CUSTOM_FILTER and ('mr' in spec.scope or 'nightly' in spec.scope)"
+      elif [[ $SCOPE == weekly ]]; then
+        JET_CUSTOM_FILTER="$JET_CUSTOM_FILTER and 'weekly' in spec.scope"
+      elif [[ $SCOPE == release ]]; then
+        JET_CUSTOM_FILTER="$JET_CUSTOM_FILTER and 'release' in spec.scope"
+      fi
+    - |
+      if [[ "$JET_CUSTOM_FILTER" == "type == 'basic'" ]]; then
+        JET_CUSTOM_FILTER="False"
+      fi
+    - echo "JET_CI_BRANCH=$JET_CI_BRANCH" | tee -a build.env
+    - echo "JET_CUSTOM_FILTER=$JET_CUSTOM_FILTER" | tee -a build.env
+  artifacts:
+    reports:
+      dotenv: build.env
+  rules:
+    - if: '$FUNCTIONAL_TEST == "yes"'
+
+ppp_capacity_statistics:
+  tags: [mcore-ssh-agent]
+  stage: .pre
+  script:
+    - |
+      set -x
+
+      ALL_USER=$(sshare -aP | grep coreai_dlalgo_mcore | tail -n +2 | awk -F '|' '{print $2}' | tr '\n' ',')
+
+      # Get the current year, month, and day
+      YEAR=$(date +%Y)
+      MONTH=$(date +%m)
+      DAY=$([[ $(date +%-d) -le 15 ]] && echo "01" || echo "15")
+      TIMESTAMP="${YEAR}-${MONTH}-${DAY}T00:00:01"
+
+      CLUSTER_ID=$(curl "${RESOURCE_ENDPOINT}/api/v1/clusters" \
+        -H "accept: application/json, text/plain, */*" \
+        -H "accept-language: en-US,en;q=0.9" \
+        -H "authorization: Bearer $CSRG_API_KEY" | jq '.[] | select(.name == "draco-oci-iad") | .id' | tr -d '"')
+
+      INITIATIVE_ITEM_ID=$(curl "${RESOURCE_ENDPOINT}/api/v1/initiative-items" \
+        -H "accept: application/json, text/plain, */*" \
+        -H "accept-language: en-US,en;q=0.9" \
+        -H "authorization: Bearer $CSRG_API_KEY" | jq '.[] | select(.name == "coreai_dlalgo_mcore") | .id' | tr -d '"')
+
+      QUOTA=$(curl "${RESOURCE_ENDPOINT}/api/v1/capacity-requests" \
+        -H "accept: application/json, text/plain, */*" \
+        -H "accept-language: en-US,en;q=0.9" \
+        -H "authorization: Bearer $CSRG_API_KEY" | jq --arg CLUSTER_ID $CLUSTER_ID --arg INITIATIVE_ITEM_ID $INITIATIVE_ITEM_ID '[.[] | select(.clusterId == $CLUSTER_ID and .initiativeItemId == $INITIATIVE_ITEM_ID)] | to_entries | [last] | .[0].value.quantity')
+
+      USED_CAPA=$(sacct \
+        -u ${ALL_USER} \
+        --partition batch_block1,batch_block3,batch_block4 \
+        --truncate \
+        -A coreai_dlalgo_mcore \
+        -S ${TIMESTAMP} \
+        -X \
+        --format JobID,JobName%20,Partition,AllocNodes,ElapsedRaw \
+        -p \
+        -n \
+      | awk -F "|" '{{sum+=$4*$5}} END {{print sum*8/3600}}')
+      TOTAL_CAPA=$(( $QUOTA*24*30 ))
+
+      USAGE=$(echo "$USED_CAPA $TOTAL_CAPA" | awk '{print (1 - $1/$2)*100}')%
+
+      echo "Usage left: $USAGE"
+      echo "Disclaimer: Please be careful with this number. Usage does not imply
+        what we are guaranteed to get a slot, SLURM scheduling is more complicated
+        than that. The number is rather a proxy to the FairShare that determines
+        our job-scheduling-priority.
+
+        Most important take-away of this number is to get a sense how much much
+        we are eating up our budget such that we can discuss this with capacity planning.
+        "
+
+build_image:
+  tags:
+    - mcore-docker-node
+  image: docker:26.1.4-dind
+  needs: []  # May start ASAP
+  stage: build
+  timeout: 45m
+  parallel:
+    matrix:
+      - IMAGE: CI_MCORE_IMAGE
+        FILE: Dockerfile.ci
+        BASE_IMAGE: nvcr.io/nvidia/pytorch:24.01-py3
+      - IMAGE: CI_NEMO_IMAGE
+        FILE: Dockerfile.ci
+        BASE_IMAGE: nvcr.io/nvidian/nemo:nightly
+      - IMAGE: LINTING_IMAGE
+        FILE: Dockerfile.linting
+        BASE_IMAGE: python:3.10
+  before_script:
+    - echo "$NGC_API_KEY" | docker login nvcr.io -u '$oauthtoken' --password-stdin
+    - echo "$CI_REGISTRY_PASSWORD" | docker login $CI_REGISTRY -u $CI_REGISTRY_USER --password-stdin
+  script:
+    - |
+      set -x
+      eval "IMAGE=\$$IMAGE"
+
+      OLD_IMAGES=$(docker image ls --format "{{.ID}} {{.Repository}}:{{.Tag}}" \
+                    | grep -v 'nvcr.io/nvidia/pytorch:24.01-py3' \
+                    | grep -v 'gitlab-master.nvidia.com:5005/adlr/megatron-lm/mcore_ci:buildcache' \
+                    | grep -v 'gitlab-master.nvidia.com:5005/adlr/megatron-lm/mcore_nemo:buildcache' \
+                    | grep -v 'gitlab-master.nvidia.com:5005/adlr/megatron-lm/mcore_linting:buildcache' \
+                    | grep -v 'nvcr.io/nvidian/nemo:nightly' \
+                    | grep -v 'python:3.10' | awk '{ print $1 }'
+                 )
+      docker rmi $OLD_IMAGES || true
+      docker builder prune  -a --filter "until=24h" -f
+
+      if [[ "$CI_COMMIT_BRANCH" == "$CI_DEFAULT_BRANCH" ]]; then
+        ADDITIONAL_PARAMS="--pull"
+      fi
+
+      docker build \
+        -f $FILE \
+        -t ${IMAGE}:${CI_PIPELINE_ID} \
+        --cache-to type=inline \
+        --cache-from type=registry,ref=${IMAGE}:buildcache \
+        --build-arg FROM_IMAGE_NAME=$BASE_IMAGE \
+        ${ADDITIONAL_PARAMS} .
+
+      docker push ${IMAGE}:${CI_PIPELINE_ID}
+
+      if [[ "$CI_COMMIT_BRANCH" == "$CI_DEFAULT_BRANCH" ]]; then
+        docker tag ${IMAGE}:${CI_PIPELINE_ID} ${IMAGE}:buildcache
+        docker push ${IMAGE}:buildcache
+      fi
+
+      if [[ $CI_COMMIT_BRANCH == core_r* ]]; then
+        docker tag ${IMAGE}:${CI_PIPELINE_ID} ${IMAGE}:v${CI_COMMIT_BRANCH#core_r}-${CI_PIPELINE_ID}
+        docker push ${IMAGE}:v${CI_COMMIT_BRANCH#core_r}-${CI_PIPELINE_ID}
+      fi
+
+.unit_test_common:
+  image: ${CI_MCORE_IMAGE}:${CI_PIPELINE_ID}
+  stage: unit_tests
+  needs: [build_image]
+  tags:
+    - 8xL40S
+  variables:
+    MOE_GROUPED_GEMM: 0 # Set to 1 to enable grouped gemm for MoE
+  retry:
+    max: 2
+    when: job_execution_timeout
+
+unit_tests:
+  extends: [.unit_test_common]
+  script:
+    - torchrun --nproc_per_node=8 -m pytest -x -v -s --cov-report=term --cov-report=html --cov=megatron/core --no-cov-on-fail tests/unit_tests
+  coverage: '/(?i)total.*? (100(?:\.0+)?\%|[1-9]?\d(?:\.\d+)?\%)$/'
+  artifacts:
+    paths:
+      - coverage
+    expire_in: 30 days
+  rules:
+    - if: '$FUNCTIONAL_TEST == "yes" && $CI_PIPELINE_SOURCE == "merge_request_event" && ($CI_MERGE_REQUEST_TARGET_BRANCH_NAME != $CI_DEFAULT_BRANCH && $CI_MERGE_REQUEST_TARGET_BRANCH_NAME !~ /^core_r/)'
+      allow_failure: true
+    - if: '$FUNCTIONAL_TEST == "yes"'
+
+unit_tests-data:
+  extends: [.unit_test_common]
+  script:
+    - torchrun --nproc_per_node=8 -m pytest -x -v -s tests/unit_tests/data
+  rules:
+    - if: '$FUNCTIONAL_TEST == "no" && $CI_PIPELINE_SOURCE == "merge_request_event" && ($CI_MERGE_REQUEST_TARGET_BRANCH_NAME != $CI_DEFAULT_BRANCH && $CI_MERGE_REQUEST_TARGET_BRANCH_NAME !~ /^core_r/)'
+      allow_failure: true
+    - if: '$FUNCTIONAL_TEST == "no"'
+
+unit_tests-dist-checkpointing:
+  extends: [.unit_test_common]
+  script:
+    - torchrun --nproc_per_node=8 -m pytest -x -v -s tests/unit_tests/dist_checkpointing
+  rules:
+    - if: '$FUNCTIONAL_TEST == "no" && $CI_PIPELINE_SOURCE == "merge_request_event" && ($CI_MERGE_REQUEST_TARGET_BRANCH_NAME != $CI_DEFAULT_BRANCH && $CI_MERGE_REQUEST_TARGET_BRANCH_NAME !~ /^core_r/)'
+      allow_failure: true
+    - if: '$FUNCTIONAL_TEST == "no"'
+
+unit_tests-fusions:
+  extends: [.unit_test_common]
+  script:
+    - torchrun --nproc_per_node=8 -m pytest -x -v -s tests/unit_tests/fusions
+  rules:
+    - if: '$FUNCTIONAL_TEST == "no" && $CI_PIPELINE_SOURCE == "merge_request_event" && ($CI_MERGE_REQUEST_TARGET_BRANCH_NAME != $CI_DEFAULT_BRANCH && $CI_MERGE_REQUEST_TARGET_BRANCH_NAME !~ /^core_r/)'
+      allow_failure: true
+    - if: '$FUNCTIONAL_TEST == "no"'
+
+unit_tests-inference:
+  extends: [.unit_test_common]
+  script:
+    - torchrun --nproc_per_node=8 -m pytest -x -v -s tests/unit_tests/inference
+  rules:
+    - if: '$FUNCTIONAL_TEST == "no" && $CI_PIPELINE_SOURCE == "merge_request_event" && ($CI_MERGE_REQUEST_TARGET_BRANCH_NAME != $CI_DEFAULT_BRANCH && $CI_MERGE_REQUEST_TARGET_BRANCH_NAME !~ /^core_r/)'
+      allow_failure: true
+    - if: '$FUNCTIONAL_TEST == "no"'
+
+unit_tests-models:
+  extends: [.unit_test_common]
+  script:
+    - torchrun --nproc_per_node=8 -m pytest -x -v -s tests/unit_tests/models
+  rules:
+    - if: '$FUNCTIONAL_TEST == "no" && $CI_PIPELINE_SOURCE == "merge_request_event" && ($CI_MERGE_REQUEST_TARGET_BRANCH_NAME != $CI_DEFAULT_BRANCH && $CI_MERGE_REQUEST_TARGET_BRANCH_NAME !~ /^core_r/)'
+      allow_failure: true
+    - if: '$FUNCTIONAL_TEST == "no"'
+
+unit_tests-pipeline-parallel:
+  extends: [.unit_test_common]
+  script:
+    - torchrun --nproc_per_node=8 -m pytest -x -v -s tests/unit_tests/pipeline_parallel
+  rules:
+    - if: '$FUNCTIONAL_TEST == "no" && $CI_PIPELINE_SOURCE == "merge_request_event" && ($CI_MERGE_REQUEST_TARGET_BRANCH_NAME != $CI_DEFAULT_BRANCH && $CI_MERGE_REQUEST_TARGET_BRANCH_NAME !~ /^core_r/)'
+      allow_failure: true
+    - if: '$FUNCTIONAL_TEST == "no"'
+
+unit_tests-tensor-parallel:
+  extends: [.unit_test_common]
+  script:
+    - torchrun --nproc_per_node=8 -m pytest -x -v -s tests/unit_tests/tensor_parallel
+  rules:
+    - if: '$FUNCTIONAL_TEST == "no" && $CI_PIPELINE_SOURCE == "merge_request_event" && ($CI_MERGE_REQUEST_TARGET_BRANCH_NAME != $CI_DEFAULT_BRANCH && $CI_MERGE_REQUEST_TARGET_BRANCH_NAME !~ /^core_r/)'
+      allow_failure: true
+    - if: '$FUNCTIONAL_TEST == "no"'
+
+unit_tests-transformer:
+  extends: [.unit_test_common]
+  script:
+    - torchrun --nproc_per_node=8 -m pytest -x -v -s tests/unit_tests/transformer
+  rules:
+    - if: '$FUNCTIONAL_TEST == "no" && $CI_PIPELINE_SOURCE == "merge_request_event" && ($CI_MERGE_REQUEST_TARGET_BRANCH_NAME != $CI_DEFAULT_BRANCH && $CI_MERGE_REQUEST_TARGET_BRANCH_NAME !~ /^core_r/)'
+      allow_failure: true
+    - if: '$FUNCTIONAL_TEST == "no"'
+
+unit_tests-top-py:
+  extends: [.unit_test_common]
+  script:
+    - torchrun --nproc_per_node=8 -m pytest -x -v -s tests/unit_tests/*.py
+  rules:
+    - if: '$FUNCTIONAL_TEST == "no" && $CI_PIPELINE_SOURCE == "merge_request_event" && ($CI_MERGE_REQUEST_TARGET_BRANCH_NAME != $CI_DEFAULT_BRANCH && $CI_MERGE_REQUEST_TARGET_BRANCH_NAME !~ /^core_r/)'
+      allow_failure: true
+    - if: '$FUNCTIONAL_TEST == "no"'
+
+docs_build_test:
+  image: gitlab-master.nvidia.com:5005/adlr/megatron-lm/python-format:0.0.1
+  stage: unit_tests
+  tags:
+    - os/linux
+  script:
+    - cd ..
+    - rm -rf documentation && git clone https://gitlab-ci-token:${CI_JOB_TOKEN}@gitlab-master.nvidia.com/nemo-megatron-core-tme/documentation.git
+    - mv megatron-lm/ documentation/
+    - cd documentation/
+    - ./repo docs
+  allow_failure: true
+  except:
+    - main
+  interruptible: true
+
+formatting:
+  image: ${LINTING_IMAGE}:${CI_PIPELINE_ID}
+  tags:
+    - os/linux
+  stage: unit_tests
+  before_script:
+    - git fetch origin main
+  script:
+    - CHECK_ONLY=true bash tools/autoformat.sh
+
+  rules:
+    - if: '$CI_PIPELINE_SOURCE == "merge_request_event" && ($CI_MERGE_REQUEST_TARGET_BRANCH_NAME != $CI_DEFAULT_BRANCH && $CI_MERGE_REQUEST_TARGET_BRANCH_NAME !~ /^core_r/)'
+      allow_failure: true
+    - when: always
+  interruptible: true
+
+include:
+  - jet-tests.yml

PAI-Megatron-LM-240718/CODEOWNERS

+[MCORE][3]
+megatron/core/ @shanmugamr @jcasper @eharper @terryk @okoenig
+
+[TESTS]
+tests/ @shanmugamr @terryk @okoenig
+
+[MODELOPT]
+examples/inference/quantization @chenhany @kmorabia

PAI-Megatron-LM-240718/CONTRIBUTING.md

+# Contributing to Megatron-LM
+
+This document outlines the processes and policies for issues and pull requests by non-NVIDIA contributors to the Megatron-LM github repository.
+
+Everyone is welcome to contribute to the project but development of Megatron-LM continues internally at NVIDIA. When contributing it important to ensure that changes are in line with the project direction. Small changes to fix bugs are welcomed and appreciated. If proposing large architectural changes or changes for stylistic reasons open an issue first so we can discuss it.
+
+PRs will first be pulled into NVIDIA's internal Megatron-LM repo and then pushed back out to the open github repo with proper credit given to the committers.
+
+## Issue policy
+
+Please do file any bugs you find, keeping the following in mind:
+
+- If filing a bug, i.e. you have found something that doesn't work as expected, use the BUG template.
+- If you've found a regression in speed or accuracy use the REGRESSION template.
+- If you are requesting a new feature or modification of an existing feature use the ENHANCEMENT template.
+- If opening an issue to ask a question no template is needed but please make your question as clear and concise as possible.
+- One issue per bug. Putting multiple things in the same issue makes both discussion and completion unnecessarily complicated.
+- Your bug is mostly likely to get attention from the development team quickly if we can easily reproduce it.
+- Use proper spelling, grammar, and punctuation.
+- Write in an authoritative and technical tone.
+
+## Code submission policy
+
+Here are some dos & don'ts to try and stick to:
+
+### Do:
+
+- Format new code in a style that is consistent with the file being changed. Megatron-LM doesn't (yet) have a style guide or enforced formatting.
+- Split your changes into separate, atomic commits i.e. A commit per feature or fix.
+- Make sure your commits are rebased on the master branch.
+- Write the commit message subject line in the imperative mood ("Change the default argument for X", not "Changed the default argument for X").
+- Write your commit messages in proper English, with care and punctuation.
+- Check the spelling of your code, comments and commit messages.
+
+### Don't:
+
+- Submit code that's incompatible with the project licence.
+- Touch anything outside the stated scope of the PR. This includes formatting changes to code not relevant to the PR.
+- Iterate excessively on your design across multiple commits.
+- Include commented-out code.
+- Attempt large architectural changes without first opening an issue to discuss.
+
+## Issue and Pull Request Q&A (Updated Jul 2023)
+
+### I've submitted an issue and PR. When can I expect to get some feedback?
+
+Megatron-LM is developed and maintained by a small team of researchers. We will endeavour to read and acknowledge all new issues and PRs within a week. A few rules of thumb:
+- Reproducible bugs/regressions and bug/regression fixes are likely to get the attention of maintainers the quickest.
+- Issues requesting an enhancement may only recieve acknowlegement that they've been read and may be closed with a "wontfix" label if they're not inline with the project direction. If they are acknowledged and remain open you can assume the maintainers agree they're a desirable feature.
+- Support requests, i.e. requests for help running the code, have the lowest priority and will be responded to as maintainer time permits.
+
+### If my issue or PR isn't getting attention, how long should I wait before pinging one of the project maintainers?
+
+One week if there is no acknowledgement of the intial request.
+
+### Who are the project maintainers I should ping?
+
+The corresponding maintainers at this time are @jaredcasper and @jon-barker.
+
+### Is there a policy for issues and PRs that haven't been touched in X days? Should they be closed?
+
+Yes, starting in July 2023 we have a bot that will mark untouched PRs as "stale" after 60 days.
+
+We have a long backlog of issues and PRs dating back 3.5 years. We are trying to triage these now by working backwards. Older issues we believe may still be relevant may recieve a request to re-test them with the latest code. If there's no response they may be closed. Again, if you they should be re-opened then just respond with a comment to that effect.
+
+Thank-you!
\ No newline at end of file

PAI-Megatron-LM-240718/Dockerfile.ci

+# syntax=docker/dockerfile:experimental
+
+ARG FROM_IMAGE_NAME
+FROM $FROM_IMAGE_NAME
+ENV DEBIAN_FRONTEND=noninteractive
+
+RUN sed -i -e 's/^APT/# APT/' -e 's/^DPkg/# DPkg/' \
+      /etc/apt/apt.conf.d/docker-clean
+
+RUN apt-get update && \
+      apt-get install -y --no-install-recommends gettext && \
+      apt-get clean
+
+RUN wget https://github.com/mikefarah/yq/releases/download/v4.44.1/yq_linux_amd64 -O /usr/local/bin/yq && \
+chmod a+x /usr/local/bin/yq
+
+RUN pip3 install --no-cache-dir \
+      einops \
+      flask-restful \
+      nltk \
+      pytest \
+      pytest-cov \
+      pytest_mock \
+      sentencepiece \
+      wrapt \
+      git+https://github.com/fanshiqing/grouped_gemm@v1.1.2 \
+      zarr \
+      tensorstore==0.1.45
+
+##### For Mamba begin #####
+RUN pip uninstall -y triton && \
+    pip install triton==2.1.0
+
+# The causal-conv1d and mamba-ssm packages below are built from scratch here
+# (which takes significant time) because there are no wheels available on PyPI
+# for these relatively newer versions of the packages that are compatible with
+# the older NGC-variant PyTorch version (e.g. version 2.2.0.dev231106) that we
+# are using (in the NGC base container). Generally, if the package is not
+# compatible with the PyTorch version, then it will generate a Python import
+# error. The package authors tend to only release wheels for new versions of
+# these pacakges which are compatible with the versions of regular PyTorch and
+# NGC-variant PyTorch that are newer at the time of release. So, to use newer
+# versions of these packages with relatively older versions of the NGC PyTorch
+# container, we tend to have to build the packages from scratch.
+
+RUN cd /tmp && \
+    pip uninstall -y causal-conv1d && \
+    git clone https://github.com/Dao-AILab/causal-conv1d.git && \
+    cd causal-conv1d && \
+    git checkout v1.2.2.post1 && \
+    CAUSAL_CONV1D_FORCE_BUILD=TRUE pip install . && \
+    cd .. && \
+    rm -rf causal-conv1d
+
+RUN cd /tmp && \
+    pip uninstall -y mamba-ssm && \
+    git clone https://github.com/state-spaces/mamba.git && \
+    cd mamba && \
+    git checkout v2.0.3 && \
+    MAMBA_FORCE_BUILD=TRUE pip install . && \
+    cd .. && \
+    rm -rf mamba
+##### For Mamba end #####
+
+COPY . /workspace/megatron-lm
+
+RUN cp -r /workspace/megatron-lm /opt && \
+    pip install /opt/megatron-lm

PAI-Megatron-LM-240718/Dockerfile.linting

+# syntax=docker/dockerfile:experimental
+
+ARG FROM_IMAGE_NAME
+FROM $FROM_IMAGE_NAME
+ENV DEBIAN_FRONTEND=noninteractive
+
+RUN sed -i -e 's/^APT/# APT/' -e 's/^DPkg/# DPkg/' \
+      /etc/apt/apt.conf.d/docker-clean
+
+
+RUN pip3 install --no-cache-dir \
+      black==24.4.2 \
+      isort
+
+COPY . /opt/megatron-lm
+
+WORKDIR /opt/megatron-lm
\ No newline at end of file

PAI-Megatron-LM-240718/LICENSE

+The following applies to all files unless otherwise noted:
+
+# Copyright (c) 2022, NVIDIA CORPORATION. All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions
+# are met:
+#  * Redistributions of source code must retain the above copyright
+#    notice, this list of conditions and the following disclaimer.
+#  * Redistributions in binary form must reproduce the above copyright
+#    notice, this list of conditions and the following disclaimer in the
+#    documentation and/or other materials provided with the distribution.
+#  * Neither the name of NVIDIA CORPORATION nor the names of its
+#    contributors may be used to endorse or promote products derived
+#    from this software without specific prior written permission.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
+# EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+# PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT OWNER OR
+# CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+# EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
+# PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
+# OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+--
+
+This repository also contains code from Hugging Face Inc., Google Research,
+Facebook (from their Fairseq, Dino, and ParlAI projects), Microsoft (from their
+Swin-Transformer project), Philip Popien, and the Mamba project (Tri Dao and
+Albert Gu). Files from these organizations have notices at the top of each file.
+Below are licenses used in those files, as indicated.
+
+
+--------------------------------------------------------------------------------
+-- LICENSE FOR Facebook, huggingface, Google Research, LLaVA, and Mamba code  --
+
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.
+
+------------- LICENSE FOR various code from Facebook --------------
+
+MIT License
+
+Copyright (c) Facebook, Inc. and its affiliates.
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
+------------- LICENSE FOR Mircrosoft Swin transformer code --------------
+
+MIT License
+
+Copyright (c) Microsoft Corporation.
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE
+
+

PAI-Megatron-LM-240718/MANIFEST.in

+include megatron/core/requirements.txt

PAI-Megatron-LM-240718/docs/llama_mistral.md

+# Llama, Mistral and other Llama-like model support in Megatron-LM
+
+NOTE: Llama-3 and Mistral support in Megatron is currently experimental and we are still evaluting benchmark results to confirm model conversion, training and inference correctness.
+
+The [Llama-2](https://ai.meta.com/llama/) and [Llama-3](https://llama.meta.com/) family of models are an open-source set of pretrained & finetuned (for chat) models that have achieved strong results across a wide set of benchmarks. At their times of release, both Llama-2 and Llama-3 models achieved among the best results for open-source models, and were competitive with leading closed-source models (see https://arxiv.org/pdf/2307.09288.pdf and https://ai.meta.com/blog/meta-llama-3/).
+
+Similarly, [Mistral-7b](https://mistral.ai/news/announcing-mistral-7b/) is an open-source model with pretrained and finetuned (for chat) variants that achieve strong benchmark results.
+
+Architecturally Llama-2, Llama-3 and Mistral-7b are very similar. As such Megatron can support loading checkpoints from all three for inference and finetuning. Converting the checkpoints and loading them is slightly different for each model and is detailed for each below.
+
+# Llama-2
+
+Llama-2 checkpoints can be loaded into Megatron for inference and for finetuning. Loading these checkpoints consists of three steps:
+
+1. Get access to download the checkpoints.
+2. Convert the checkpoints from Meta/Huggingface format to Megatron format.
+3. Setup arguments for launching the model.
+
+The following sections detail these steps. The final section lists benchmark result comparisons between: 1) Llama-2 inference code running the Meta-format checkpoints, and 2) Megatron inference code running the converted checkpoints.
+
+## Contents
+  * [Download Meta or Huggingface checkpoints](#download-meta-or-huggingface-checkpoints)
+  * [Convert checkpoint format](#convert-checkpoint-format)
+    * [Meta format](#meta-format)
+    * [Huggingface format](#huggingface-format)
+  * [Launch model](#launch-model)
+    * [Megatron](#launch-megatron)
+    * [Meta](#launch-meta)
+    * [Huggingface](#launch-hf)
+  * [Benchmark results](#benchmark-results)
+
+## Download Meta or Huggingface checkpoints
+
+Users must first apply for access to download the Llama-2 checkpoints either directly from [Meta](https://ai.meta.com/resources/models-and-libraries/llama-downloads/) or through [Huggingface](https://huggingface.co/docs/transformers/main/model_doc/llama2) (HF). The checkpoints are available in two formats, Meta's native format (available from both the Meta and HF links), and HF's format (available only from HF). Either format can be converted to Megatron, as detailed next.
+
+## Convert checkpoint format
+
+We recommend passing `--dtype bf16` for training or finetuning. Inference can be done in bfloat16 or float16.
+
+### Meta format
+
+The Meta format checkpoints are converted to HF format as an intermediate step before converting to Megatron format. The `transformers` package is required, and must have version >=4.31.0 (e.g., `pip install transformers>=4.31.0`). (**Note**: we have specifically tested with versions `4.31.0` and `4.32.0`; your experience may vary with newer versions.) Assuming the downloaded checkpoints are in `$CHECKPOINT_DIR` (with separate sub-directories for 7B, 13B, 70B, etc.), the following example command can be used to convert from Llama-2 format to HF format in bfloat16:
+
+```
+python tools/checkpoint/convert.py --model-type GPT \
+>   --loader llama_mistral \
+>   --saver megatron \
+>   --checkpoint-type meta \
+>   --model-size llama2-7B \
+>   --load-dir $LLAMA_META_FORMAT_DIR \
+>   --save-dir ${MEGATRON_FORMAT_DIR} \
+>   --tokenizer-model ${TOKENIZER_MODEL} \
+>   --target-tensor-parallel-size ${TP} \
+>   --target-pipeline-parallel-size ${PP} \
+>   --bf16
+```
+
+Valid values for `--model-size` are `llama2-7B`, `llama2-13B`, and `llama2-70B` (for pretrained-only models), and `llama2-7Bf`, `llama2-13Bf`, and `llama2-70Bf` (for chat-finetuned models).
+
+### Huggingface format
+
+The HF checkpoints can be converted to Megatron format by using Megatron's own Llama-2 checkpoint converter for HF format (see script `tools/checkpoint/loader_llama_mistral.py`). One important argument that must be set correctly is the tensor parallel size (`TP`) for each model. The following table shows these values:
+
+| Model size | Tensor parallel size (`TP`) |
+| ---------- | --------------------------- |
+|  7B        | 1                           |
+| 13B        | 2                           |
+| 70B        | 8                           |
+
+Using these values for `TP`, along with the path to the Llama-2 tokenizer model (automatically downloaded with original checkpoint download; see `${TOKENIZER_MODEL}` below), run the following command from the root of your Megatron source code to convert from HF format to Megatron format:
+
+```
+$>: python tools/checkpoint/convert.py \
+ >    --model-type GPT \
+ >    --loader llama_mistral \
+ >    --saver megatron \
+ >    --target-tensor-parallel-size ${TP} \
+ >    --checkpoint-type hf
+ >    --load-dir ${HF_FORMAT_DIR} \
+ >    --save-dir ${MEGATRON_FORMAT_DIR} \
+ >    --tokenizer-model ${TOKENIZER_MODEL}
+```
+
+After this conversion, we are ready to load the checkpoints into a Megatron GPT model.
+
+## Launch model
+
+### Launch Megatron
+
+If loading for either inference or finetuning, use the following arguments:
+
+```
+--tensor-model-parallel-size ${TP} \
+--pipeline-model-parallel-size 1 \
+--seq-length 4096 \
+--max-position-embeddings 4096 \
+--tokenizer-type Llama2Tokenizer \
+--tokenizer-model ${TOKENIZER_MODEL} \
+--load ${CHECKPOINT_DIR} \
+--exit-on-missing-checkpoint \
+--use-checkpoint-args \
+--no-load-optim \
+--no-load-rng \
+--untie-embeddings-and-output-weights \
+--use-rotary-position-embeddings \
+--normalization RMSNorm \
+--no-position-embedding \
+--no-masked-softmax-fusion \
+--attention-softmax-in-fp32
+```
+
+### Launch Meta
+
+Meta checkpoints can be launched with: https://github.com/facebookresearch/llama
+
+### Launch Huggingface
+
+Huggingface checkpoints can be launched with: https://github.com/huggingface/transformers/blob/main/src/transformers/models/llama/modeling_llama.py
+
+## Benchmark results
+
+The tables below list the benchmark comparisons between native Llama-2 (using Meta's checkpoint and Meta's inference code) and Megatron (using a converted HF checkpoint and Megatron's inference code).
+
+The values are the percent error between Megatron and Llama-2, calculated using the formula: `|<llama_score> - <megatron_score>| / <llama_score>`, where the type of score is detailed before each table. Across all tests (80 total per model size), the mean error is 0.15%. The small difference in benchmark scores between the two models is due to minor arithmetic differences in implementation that alter the numerics slightly. Some of the factors that influence this difference include:
+
+- Megatron performs batch matrix multiplications in a couple places, such as within self attention and in SwiGLU, that Llama performs separately.
+- Megatron uses `torch.baddbmm` within self attention, versus Llama using `torch.matmul`.
+- Megatron uses a `sin`/`cos` implementation for rotary position embeddings, versus Llama using a `polar`/`complex` implementation.
+- Llama calls `torch.set_default_dtype(torch.float16)` during initialization, which Megatron does not.
+
+### Big Bench
+
+Score type: multiple choice grade.
+
+| bigbench / standard | 7b | 13b | 70b |
+| -- | -- | -- | -- |
+| date_understanding | 0.29% | 0.13% | 0.12% |
+| general_knowledge | 0.00% | 0.00% | 0.00% |
+| human_organs_senses | 0.00% | 0.00% | 0.00% |
+| intent_recognition | 0.00% | 0.11% | 0.00% |
+| riddle_sense | 0.00% | 0.00% | 0.00% |
+| similarities_abstraction | 0.00% | 0.58% | 0.00% |
+| simple_arithmetic_json_multiple_choice | 0.00% | 0.00% | 0.00% |
+| undo_permutation | 0.19% | 0.19% | 0.18% |
+
+### Multilingual
+
+Score type: multiple choice grade.
+
+| multilingual / xcopa | 7b  | 13b  | 70b |
+| -- | -- | -- | -- |
+| en-template-mGPT-remove-punctuation | 0.08% | 0.00% | 0.00% |
+| et-template-mGPT-remove-punctuation | 0.00% | 0.13% | 0.25% |
+| ht-template-mGPT-remove-punctuation | 0.26% | 0.13% | 0.26% |
+| id-template-mGPT-remove-punctuation | 0.11% | 0.00% | 0.19% |
+| it-template-mGPT-remove-punctuation | 0.00% | 0.10% | 0.09% |
+| qu-template-mGPT-remove-punctuation | 0.00% | 0.00% | 0.27% |
+| sw-template-mGPT-remove-punctuation | 0.14% | 0.13% | 0.13% |
+| th-template-mGPT-remove-punctuation | 0.25% | 0.13% | 0.13% |
+| tr-template-mGPT-remove-punctuation | 0.26% | 0.00% | 0.34% |
+| vi-template-mGPT-remove-punctuation | 0.00% | 0.11% | 0.00% |
+| zh-template-mGPT-remove-punctuation | 0.00% | 0.10% | 0.09% |
+
+### LM Evaluation Harness
+
+Score type: multiple choice grade.
+
+| lm-eval | 7b  | 13b  | 70b |
+| -- | -- | -- | -- |
+| boolq | 0.04% | 0.04% | 0.07% |
+| hellaswag | 0.02% | 0.03% | 0.03% |
+| piqa | 0.00% | 0.00% | 0.07% |
+| winogrande | 0.00% | 0.11% | 0.20% |
+
+### MMLU
+
+Score type: multiple choice grade.
+
+Note: the number in brackets is the number of sub-tasks for each supercategory.
+
+| mmlu | 7b  | 13b  | 70b |
+| -- | -- | -- | -- |
+| stem [18]  | 0.79% | 0.05% | 0.01% |
+| humanities [13]  | 0.19% | 0.01% | 0.02% |
+| other (business, health, misc.) [14]  | 0.08% | 0.06% | 0.12% |
+| social sciences [12]  | 0.37% | 0.21% | 0.01% |
+
+# Llama-3
+
+Llama-3 checkpoints can be loaded into Megatron for inference and for finetuning. Loading these checkpoints consists of several steps:
+
+1. Get access to download the checkpoints (weights and tokenizer).
+2. Clone the llama3 loading code from Meta.
+3. Install the llama package from source.
+4. Convert the checkpoints from Meta/Huggingface format to Megatron format.
+5. Setup arguments for launching the model.
+
+The following sections detail these steps.
+
+## Contents
+  * [Download Meta or Huggingface checkpoints](#download-meta-or-huggingface-checkpoints)
+  * [Install tiktoken](#install-tiktoken)
+  * [Install llama package from Meta](#install-llama-package)
+  * [Convert checkpoint format](#convert-checkpoint-format)
+    * [Meta format](#meta-format)
+    * [Huggingface format](#huggingface-format)
+  * [Launch model](#launch-model)
+    * [Megatron](#launch-megatron)
+    * [Meta](#launch-meta)
+    * [Huggingface](#launch-hf)
+  * [Benchmark results](#benchmark-results)
+
+## Download Meta or Huggingface checkpoints
+
+Users must first apply for access to download the Llama-3 checkpoints either directly from [Meta](https://llama.meta.com/llama-downloads) or through [Huggingface](https://huggingface.co/meta-llama) (HF). The checkpoints are available in two formats, Meta's native format (available from both the Meta and HF links), and HF's format (available only from HF). Either format can be converted to Megatron, as detailed next.
+
+## Install tiktoken
+
+The Llama-3 tokenizer relies on the availability of the `tiktoken` module which can be installed through `pip`.
+
+## Install llama package from Meta
+
+1. In a location outside of the megatron-lm source directory, e.g `~`: `git clone https://github.com/meta-llama/llama3.git`
+2. `cd $LLAMA3_SOURCE_DIR`
+4. `pip install -e .`
+
+## Convert checkpoint format
+
+We recommend passing `--dtype bf16` for training or finetuning. Inference can be done in bfloat16 or float16.
+
+### Meta format
+
+The Meta format checkpoints are converted to HF format as an intermediate step before converting to Megatron format. The `transformers` package is required, and must have version >=4.31.0 (e.g., `pip install transformers>=4.31.0`). (**Note**: we have specifically tested with versions `4.31.0` and `4.32.0`; your experience may vary with newer versions.) Assuming the downloaded checkpoints are in `$CHECKPOINT_DIR` (with separate sub-directories for 8B, 70B, etc.), the following example command can be used to convert from Llama-3 format to HF format in bfloat16:
+
+```
+python tools/checkpoint/convert.py \
+>   --model-type GPT \
+>   --loader llama_mistral \
+>   --saver mcore \
+>   --checkpoint-type meta \
+>   --model-size llama3-8B \
+>   --load-dir $LLAMA_META_FORMAT_DIR \
+>   --save-dir ${MEGATRON_FORMAT_DIR} \
+>   --tokenizer-model ${TOKENIZER_MODEL} \
+>   --target-tensor-parallel-size ${TP} \
+>   --target-pipeline-parallel-size ${PP} \
+>   --bf16
+```
+
+Valid values for `--model_size` are `llama3-8B` and `llama3-70B` (for pretrained-only models), and `llama3-8Bf` and `llama3-70Bf` (for chat-finetuned models).
+
+### Huggingface format
+
+The HF checkpoints can be converted to Megatron format by using Megatron's own Llama-3 checkpoint converter for HF format (see script `tools/checkpoint/loader_llama_mistral.py`). One important argument that must be set correctly is the tensor parallel size (`TP`) for each model. The following table shows these values:
+
+| Model size | Tensor parallel size (`TP`) |
+| ---------- | --------------------------- |
+|  8B        | 1                           |
+| 70B        | 8                           |
+
+Using these values for `TP`, along with the path to the Llama-3 tokenizer model (automatically downloaded with original checkpoint download; see `${TOKENIZER_MODEL}` below), run the following command from the root of your Megatron source code to convert from HF format to Megatron format:
+
+```
+$>: python tools/checkpoint/convert.py \
+ >    --model-type GPT \
+ >    --loader llama_mistral \
+ >    --saver mcore \
+ >    --target-tensor-parallel-size ${TP} \
+ >    --checkpoint-type hf
+ >    --load-dir ${HF_FORMAT_DIR} \
+ >    --save-dir ${MEGATRON_FORMAT_DIR} \
+ >    --tokenizer-model ${TOKENIZER_MODEL}
+ >    --model-size llama3-8B \
+```
+
+Valid values for `--model-size` are `llama3-8B` and `llama3-70B` (for pretrained-only models), and `llama3-8Bf` and `llama3-70Bf` (for chat-finetuned models).
+
+After this conversion, we are ready to load the checkpoints into a Megatron GPT model.
+
+## Launch model
+
+### Launch Megatron
+
+If loading for either inference or finetuning, use the following arguments:
+
+```
+--tensor-model-parallel-size ${TP} \
+--pipeline-model-parallel-size 1 \
+--seq-length 4096 \
+--max-position-embeddings 4096 \
+--tokenizer-type Llama3Tokenizer \
+--tokenizer-model ${TOKENIZER_MODEL} \
+--load ${CHECKPOINT_DIR} \
+--exit-on-missing-checkpoint \
+--use-checkpoint-args \
+--no-load-optim \
+--no-load-rng \
+--untie-embeddings-and-output-weights \
+--normalization RMSNorm \
+--position-embedding-type rope \
+--no-masked-softmax-fusion \
+--attention-softmax-in-fp32
+```
+
+### Launch Meta
+
+Meta checkpoints can be launched with: https://github.com/meta-llama/llama3
+
+### Launch Huggingface
+
+Huggingface checkpoints can be launched by following the instructions here: https://huggingface.co/blog/llama3
+
+## Benchmark results
+
+Llama-3 support in Megatron is currently experimental and we are still carrying out benchmark evaluations.
+
+# Mistral-7b
+
+Megatron currently supports loading the v.03 release of Mistral-7b (which does not use sliding window attention and offers a larger 32768 vocabulary) for inference and finetuning. Loading these checkpoints consists of several steps:
+
+1. Get access to download the checkpoints (weights and tokenizer).
+2. Install the `mistral-common` package
+3. Convert the checkpoints from HuggingFace format to Megatron format.
+4. Setup arguments for launching the model.
+
+The following sections detail these steps.
+
+## Contents
+  * [Download Huggingface checkpoints](#download-huggingface-checkpoints)
+  * [Install mistral-common packgage](#install-mistral-common)
+  * [Convert checkpoint format](#convert-checkpoint-format)
+  * [Launch model](#launch-model)
+  * [Benchmark results](#benchmark-results)
+
+## Download Huggingface checkpoints
+
+Users must first apply for access to download the Mistral-7b checkpoints through [Huggingface](https://huggingface.co/mistralai/Mistral-7B-v0.3) (HF). Megatron does not currently support the v0.1 or v0.2 checkpoints, ensure you download v0.3. Megatron does not currently support using the raw weights directly from [Mistral](https://docs.mistral.ai/getting-started/open_weight_models/).
+
+## Install the mistral-common package
+
+`pip install mistral-common`
+
+## Convert checkpoint format
+
+The HF checkpoints can be converted to Megatron format by using Megatron's own Mistral checkpoint converter for HF format (see script `tools/checkpoint/loader_llama_mistral.py`).
+
+Using the path to the Mistral tokenizer model (downloaded alongside the HF checkpoint), run the following command from the root of your Megatron source code to convert from HF format to mcore format:
+
+```
+$>: python tools/checkpoint/convert.py \
+ >    --model-type GPT \
+ >    --loader llama_mistral \
+ >    --saver mcore \
+ >    --target-tensor-parallel-size ${TP} \
+ >    --checkpoint-type hf \
+ >    --load-dir ${HF_FORMAT_DIR} \
+ >    --save-dir ${MEGATRON_FORMAT_DIR} \
+ >    --tokenizer-model ${TOKENIZER_MODEL} \
+ >    --model-size mistral-7B \
+```
+
+Valid values for `--model-size` are mistral-7B for the pretrained model or mistral-7Bf for the chat fine-tuned model.
+
+After this conversion, we are ready to load the checkpoints into an mcore GPT model.
+
+## Launch model
+
+If loading for either inference or finetuning, use the following arguments:
+
+```
+--tensor-model-parallel-size ${TP} \
+--pipeline-model-parallel-size 1 \
+--seq-length 4096 \
+--max-position-embeddings 4096 \
+--tokenizer-type MistralTokenizer \
+--tokenizer-model ${TOKENIZER_MODEL} \
+--load ${CHECKPOINT_DIR} \
+--exit-on-missing-checkpoint \
+--use-checkpoint-args \
+--no-load-optim \
+--no-load-rng \
+--untie-embeddings-and-output-weights \
+--normalization RMSNorm \
+--position-embedding-type rope \
+--no-masked-softmax-fusion \
+--attention-softmax-in-fp32
+```
+
+## Benchmark results
+
+Mistral-7B support in Megatron is currently experimental and we are still carrying out benchmark evaluations.
+
+# Other Llama-like model support
+
+*Note: Experimental*
+
+Many models such as Yi-34B use the Llama architecture and may be converted from HuggingFace to Megatron using the commands in [Llama3](#llama-3).