support for batched sharegpt tokenization to skip bad data

.github/CONTRIBUTING.md

@@ -15,18 +15,18 @@ First of all, thank you for your interest in contributing to axolotl! We appreci
   - [Commit Messages](#commit-messages)
 - [Additional Resources](#additional-resources)
 
-## Code of Conduct
+## Code of Conductcode
 
 All contributors are expected to adhere to our [Code of Conduct](CODE_OF_CONDUCT.md). Please read it before participating in the axolotl community.
 
 ## Getting Started
 
-Bugs? Please check for open issue else create a new [Issue](https://github.com/axolotl-ai-cloud/axolotl/issues/new).
+Bugs? Please check for open issue else create a new [Issue](https://github.com/OpenAccess-AI-Collective/axolotl/issues/new).
 
 PRs are **greatly welcome**!
 
 1. Fork the repository and clone it to your local machine.
-2. Set up the development environment by following the instructions in the [README.md](https://github.com/axolotl-ai-cloud/axolotl/tree/main/README.md) file.
+2. Set up the development environment by following the instructions in the [README.md](https://github.com/OpenAccess-AI-Collective/axolotl/tree/main/README.md) file.
 3. Explore the codebase, run tests, and verify that everything works as expected.
 
 Please run below to setup env
@@ -42,11 +42,11 @@ pytest tests/
 
 ### Reporting Bugs
 
-If you encounter a bug or issue while using axolotl, please open a new issue on the [GitHub Issues](https://github.com/axolotl-ai-cloud/axolotl/issues) page. Provide a clear and concise description of the problem, steps to reproduce it, and any relevant error messages or logs.
+If you encounter a bug or issue while using axolotl, please open a new issue on the [GitHub Issues](https://github.com/OpenAccess-AI-Collective/axolotl/issues) page. Provide a clear and concise description of the problem, steps to reproduce it, and any relevant error messages or logs.
 
 ### Suggesting Enhancements
 
-We welcome ideas for improvements and new features. To suggest an enhancement, open a new issue on the [GitHub Issues](https://github.com/axolotl-ai-cloud/axolotl/issues) page. Describe the enhancement in detail, explain the use case, and outline the benefits it would bring to the project.
+We welcome ideas for improvements and new features. To suggest an enhancement, open a new issue on the [GitHub Issues](https://github.com/OpenAccess-AI-Collective/axolotl/issues) page. Describe the enhancement in detail, explain the use case, and outline the benefits it would bring to the project.
 
 ### Submitting Pull Requests
 

.github/FUNDING.yml

@@ -1,13 +1,13 @@
 # These are supported funding model platforms
 
-github: [winglian, OpenAccess-AI-Collective] # Replace with up to 4 GitHub Sponsors-enabled usernames e.g., [user1, user2]
+github: OpenAccess-AI-Collective # Replace with up to 4 GitHub Sponsors-enabled usernames e.g., [user1, user2]
 patreon: # Replace with a single Patreon username
 open_collective: # Replace with a single Open Collective username
-ko_fi: axolotl_ai # Replace with a single Ko-fi username
+ko_fi: # Replace with a single Ko-fi username
 tidelift: # Replace with a single Tidelift platform-name/package-name e.g., npm/babel
 community_bridge: # Replace with a single Community Bridge project-name e.g., cloud-foundry
 liberapay: # Replace with a single Liberapay username
 issuehunt: # Replace with a single IssueHunt username
 otechie: # Replace with a single Otechie username
 lfx_crowdfunding: # Replace with a single LFX Crowdfunding project-name e.g., cloud-foundry
-custom: ['https://quickchart.io/qr?text=bitcoin%3Abc1qxlgwlqwfea5s2cxm42xqsfmwjct0rj8w8ea5np&size=480&centerImageUrl=https%3A%2F%2Fupload.wikimedia.org%2Fwikipedia%2Fcommons%2Fthumb%2F4%2F46%2FBitcoin.svg%2F64px-Bitcoin.svg.png'] # Replace with up to 4 custom sponsorship URLs e.g., ['link1', 'link2']
+custom: # Replace with up to 4 custom sponsorship URLs e.g., ['link1', 'link2']

.github/ISSUE_TEMPLATE/bug-report.yaml

@@ -15,7 +15,7 @@ body:
       label: "Please check that this issue hasn't been reported before."
       description: "The **Label filters** may help make your search more focussed."
       options:
-        - label: "I searched previous [Bug Reports](https://github.com/axolotl-ai-cloud/axolotl/labels/bug) didn't find any similar reports."
+        - label: "I searched previous [Bug Reports](https://github.com/OpenAccess-AI-Collective/axolotl/labels/bug) didn't find any similar reports."
           required: true
 
   - type: textarea
@@ -59,7 +59,6 @@ body:
       label: Config yaml
       description: |
         Please attach the config yaml!
-      render: yaml
 
   - type: textarea
     id: possible-solution

.github/ISSUE_TEMPLATE/config.yml

@@ -1,7 +1,7 @@
 blank_issues_enabled: false
 contact_links:
   - name: Ask a question
-    url: https://github.com/axolotl-ai-cloud/axolotl/discussions/categories/q-a
+    url: https://github.com/OpenAccess-AI-Collective/axolotl/discussions/categories/q-a
     about: Ask questions and discuss with other community members
   - name: Discuss the Project in Discord
     url: https://discord.gg/HhrNrHJPRb

.github/ISSUE_TEMPLATE/docs.yml

@@ -10,7 +10,7 @@ body:
       value: |
         * Ask questions in [Discord](https://discord.gg/HhrNrHJPRb).
         * Before you file an issue read the [Contributing guide](./CONTRIBUTING.md).
-        * Check to make sure someone hasn't already opened a [similar issue](https://github.com/axolotl-ai-cloud/axolotl/issues).
+        * Check to make sure someone hasn't already opened a [similar issue](https://github.com/OpenAccess-AI-Collective/axolotl/issues).
   - type: textarea
     attributes:
       label: What piece of documentation is affected?

.github/ISSUE_TEMPLATE/feature-request.yaml

@@ -8,9 +8,9 @@ body:
       label: "⚠️ Please check that this feature request hasn't been suggested before."
       description: "There are two locations for previous feature requests. Please search in both. Thank you. The **Label filters** may help make your search more focussed."
       options:
-        - label: "I searched previous [Ideas in Discussions](https://github.com/axolotl-ai-cloud/axolotl/discussions/categories/ideas) didn't find any similar feature requests."
+        - label: "I searched previous [Ideas in Discussions](https://github.com/OpenAccess-AI-Collective/axolotl/discussions/categories/ideas) didn't find any similar feature requests."
           required: true
-        - label: "I searched previous [Issues](https://github.com/axolotl-ai-cloud/axolotl/labels/enhancement) didn't find any similar feature requests."
+        - label: "I searched previous [Issues](https://github.com/OpenAccess-AI-Collective/axolotl/labels/enhancement) didn't find any similar feature requests."
           required: true
 
   - type: textarea

.github/PULL_REQUEST_TEMPLATE/pull_request_template_simple.md

@@ -20,8 +20,3 @@
 ## Types of changes
 
 <!--- What types of changes does your code introduce? Put an `x` in all the boxes that apply: -->
-
-## Social Handles (Optional)
-
-<!-- Thanks for submitting a bugfix or enhancement. -->
-<!-- We'd love to show our thanks to you on Twitter & Discord if you provide your handle -->

.github/workflows/base.yml

@@ -3,71 +3,53 @@ name: ci-cd-base
 on:
   push:
     branches:
-      - "main"
-    paths:
-      - 'Dockerfile-base'
-      - '.github/workflows/base.yml'
-  pull_request:
-    paths:
-      - 'Dockerfile-base'
-      - '.github/workflows/base.yml'
-  workflow_dispatch:
+      - "main-base"
+      - "dev-base"
 
 jobs:
   build-base:
-    if: github.repository_owner == 'axolotl-ai-cloud'
+    if: github.repository_owner == 'OpenAccess-AI-Collective'
     # this job needs to be run on self-hosted GPU runners...
-    runs-on: axolotl-gpu-runner
+    runs-on: self-hosted
     strategy:
       fail-fast: false
       matrix:
         include:
-          - cuda: "124"
-            cuda_version: 12.4.1
-            cudnn_version: ""
-            python_version: "3.11"
-            pytorch: 2.4.1
-            torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
-          - cuda: "124"
-            cuda_version: 12.4.1
-            cudnn_version: ""
-            python_version: "3.11"
-            pytorch: 2.5.1
-            torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
-          - cuda: "124"
-            cuda_version: 12.4.1
-            cudnn_version: ""
-            python_version: "3.11"
-            pytorch: 2.6.0
-            torch_cuda_arch_list: "7.0 7.5 8.0 8.6 8.7 8.9 9.0+PTX"
+          - cuda: "118"
+            cuda_version: 11.8.0
+            python_version: "3.9"
+            pytorch: 2.0.1
+            torch_cuda_arch_list: "7.0 7.5 8.0 8.6 9.0+PTX"
+          - cuda: "118"
+            cuda_version: 11.8.0
+            python_version: "3.10"
+            pytorch: 2.0.1
+            torch_cuda_arch_list: "7.0 7.5 8.0 8.6 9.0+PTX"
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v3
       - name: Docker metadata
         id: metadata
-        uses: docker/metadata-action@v5
+        uses: docker/metadata-action@v3
         with:
-          images: |
-            winglian/axolotl-base
-            axolotlai/axolotl-base
+          images: winglian/axolotl-base
       - name: Login to Docker Hub
         uses: docker/login-action@v2
         with:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v2
       - name: Build
         uses: docker/build-push-action@v4
         with:
           context: .
           file: ./docker/Dockerfile-base
           push: ${{ github.event_name != 'pull_request' }}
-          tags: ${{ steps.metadata.outputs.tags }}-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
+          tags: ${{ steps.metadata.outputs.tags }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
           labels: ${{ steps.metadata.outputs.labels }}
           build-args: |
             CUDA_VERSION=${{ matrix.cuda_version }}
-            CUDNN_VERSION=${{ matrix.cudnn_version }}
             CUDA=${{ matrix.cuda }}
             PYTHON_VERSION=${{ matrix.python_version }}
             PYTORCH_VERSION=${{ matrix.pytorch }}

.github/workflows/docs.yml

@@ -1,31 +0,0 @@
-name: Publish Docs
-on:
-  push:
-    branches:
-      - main
-
-permissions:
-    contents: write
-    pages: write
-
-jobs:
-    build-deploy:
-        runs-on: ubuntu-latest
-        steps:
-        - name: Check out repository
-          uses: actions/checkout@v4
-        - name: Set up Quarto
-          uses: quarto-dev/quarto-actions/setup@v2
-        - name: Setup Python
-          uses: actions/setup-python@v5
-          with:
-            python-version: '3.11'
-        - name: install dependencies
-          run: |
-            python3 -m pip install jupyter
-        - name: Publish to GitHub Pages (and render)
-          uses: quarto-dev/quarto-actions/publish@v2
-          with:
-            target: gh-pages
-          env:
-            GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/lint.yml

@@ -1,24 +0,0 @@
-name: lint
-on:
-  # check on PRs, and manual triggers
-  merge_group:
-  pull_request:
-      paths:
-       - '**.py'
-       - 'requirements.txt'
-       - '.github/workflows/*.yml'
-       - "*.[q]md"
-       - "examples/**/*.y[a]?ml"
-  workflow_dispatch:
-
-jobs:
-  pre-commit:
-    name: pre-commit
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v4
-      - uses: actions/setup-python@v5
-        with:
-          python-version: "3.11"
-          cache: 'pip' # caching pip dependencies
-      - uses: pre-commit/action@v3.0.1

.github/workflows/main.yml

@@ -4,166 +4,94 @@ on:
   push:
     branches:
       - "main"
-    tags:
-      - "v*"
-  workflow_dispatch:
 
 jobs:
   build-axolotl:
-    if: ${{ ! contains(github.event.commits[0].message, '[skip docker]') && github.repository_owner == 'axolotl-ai-cloud' }}
+    if: github.repository_owner == 'OpenAccess-AI-Collective'
+    # this job needs to be run on self-hosted GPU runners...
     strategy:
       fail-fast: false
       matrix:
         include:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.4.1
+          - cuda: 118
+            cuda_version: 11.8.0
+            python_version: "3.9"
+            pytorch: 2.0.1
             axolotl_extras:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.5.1
-            axolotl_extras: vllm
-            is_latest: true
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.6.0
+          - cuda: 118
+            cuda_version: 11.8.0
+            python_version: "3.10"
+            pytorch: 2.0.1
             axolotl_extras:
-    runs-on: axolotl-gpu-runner
+    runs-on: [self-hosted, gpu, docker]
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v3
       - name: Docker metadata
         id: metadata
-        uses: docker/metadata-action@v5
+        uses: docker/metadata-action@v3
         with:
-          images: |
-            winglian/axolotl
-            axolotlai/axolotl
-          tags: |
-            type=ref,event=branch
-            type=pep440,pattern={{version}}
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
+          images: winglian/axolotl
       - name: Login to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@v2
         with:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
-      # guidance for testing before pushing: https://docs.docker.com/build/ci/github-actions/test-before-push/
-      - name: Build and export to Docker
-        uses: docker/build-push-action@v5
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v2
+      - name: Build
+        uses: docker/build-push-action@v4
         with:
           context: .
           build-args: |
-            BASE_TAG=${{ github.ref_type == 'tag' && 'main' || github.ref_name }}-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}
+            BASE_TAG=${{ github.ref_name }}-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}
             CUDA=${{ matrix.cuda }}
-            PYTORCH_VERSION=${{ matrix.pytorch }}
-            AXOLOTL_ARGS=${{ matrix.axolotl_args }}
           file: ./docker/Dockerfile
           push: ${{ github.event_name != 'pull_request' }}
-          tags: |
-            ${{ steps.metadata.outputs.tags }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
-            ${{ steps.metadata.outputs.tags }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}
-            ${{ (matrix.is_latest) && format('{0}-latest', steps.metadata.outputs.tags) || '' }}
+          tags: ${{ steps.metadata.outputs.tags }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
           labels: ${{ steps.metadata.outputs.labels }}
-
-  build-axolotl-cloud:
+  build-axolotl-runpod:
     needs: build-axolotl
-    if: ${{ ! contains(github.event.commits[0].message, '[skip docker]') && github.repository_owner == 'axolotl-ai-cloud' }}
+    if: github.repository_owner == 'OpenAccess-AI-Collective'
     # this job needs to be run on self-hosted GPU runners...
     strategy:
       matrix:
         include:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.4.1
+          - cuda: 118
+            cuda_version: 11.8.0
+            python_version: "3.9"
+            pytorch: 2.0.1
             axolotl_extras:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.5.1
+          - cuda: 118
+            cuda_version: 11.8.0
+            python_version: "3.10"
+            pytorch: 2.0.1
             axolotl_extras:
             is_latest: true
-    runs-on: axolotl-gpu-runner
+    runs-on: [self-hosted, gpu, docker]
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v3
       - name: Docker metadata
         id: metadata
-        uses: docker/metadata-action@v5
+        uses: docker/metadata-action@v3
         with:
-          images: |
-            winglian/axolotl-cloud
-            axolotlai/axolotl-cloud
-          tags: |
-            type=ref,event=branch
-            type=pep440,pattern={{version}}
+          images: winglian/axolotl-runpod
       - name: Login to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@v2
         with:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v2
       - name: Build
-        uses: docker/build-push-action@v5
+        uses: docker/build-push-action@v4
         with:
           context: .
           build-args: |
-            BASE_TAG=${{ github.ref_type == 'tag' && 'main' || github.ref_name }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
+            BASE_TAG=${{ github.ref_name }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
             CUDA=${{ matrix.cuda }}
-          file: ./docker/Dockerfile-cloud
-          push: ${{ github.event_name != 'pull_request' }}
-          tags: |
-             ${{ steps.metadata.outputs.tags }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
-             ${{ (matrix.is_latest) && format('{0}-latest', steps.metadata.outputs.tags) || '' }}
-          labels: ${{ steps.metadata.outputs.labels }}
-
-  build-axolotl-cloud-no-tmux:
-    needs: build-axolotl
-    if: ${{ ! contains(github.event.commits[0].message, '[skip docker]') && github.repository_owner == 'axolotl-ai-cloud' }}
-    # this job needs to be run on self-hosted GPU runners...
-    strategy:
-      matrix:
-        include:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.4.1
-            axolotl_extras:
-    runs-on: axolotl-gpu-runner
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Docker metadata
-        id: metadata
-        uses: docker/metadata-action@v5
-        with:
-          images: |
-            winglian/axolotl-cloud-term
-            axolotlai/axolotl-cloud-term
-          tags: |
-            type=ref,event=branch
-            type=pep440,pattern={{version}}
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKERHUB_USERNAME }}
-          password: ${{ secrets.DOCKERHUB_TOKEN }}
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-      - name: Build
-        uses: docker/build-push-action@v5
-        with:
-          context: .
-          build-args: |
-            BASE_TAG=${{ github.ref_type == 'tag' && 'main' || github.ref_name }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
-            CUDA=${{ matrix.cuda }}
-          file: ./docker/Dockerfile-cloud-no-tmux
+          file: ./docker/Dockerfile-runpod
           push: ${{ github.event_name != 'pull_request' }}
           tags: |
              ${{ steps.metadata.outputs.tags }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}

.github/workflows/multi-gpu-e2e.yml

@@ -1,69 +0,0 @@
-name: docker-multigpu-tests-biweekly
-
-on:
-  pull_request:
-    paths:
-      - 'tests/e2e/multigpu/*.py'
-  workflow_dispatch:
-  schedule:
-    - cron: '0 0 * * 1,4'  # Runs at 00:00 UTC every monday & thursday
-
-# Cancel jobs on the same ref if a new one is triggered
-concurrency:
-  group: ${{ github.workflow }}-${{ github.ref }}
-  cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
-
-jobs:
-  test-axolotl-multigpu:
-    if: ${{ ! contains(github.event.commits[0].message, '[skip e2e]') && github.repository_owner == 'axolotl-ai-cloud' }}
-    strategy:
-      fail-fast: false
-      matrix:
-        include:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.4.1
-            axolotl_extras:  # no vllm support for 2.4.1
-            num_gpus: 2
-            nightly_build: "true"
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.5.1
-            axolotl_extras: vllm
-            num_gpus: 2
-            nightly_build: "true"
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.6.0
-            # awaiting vllm#12721
-            axolotl_extras:
-            num_gpus: 2
-            nightly_build: "true"
-    runs-on: [self-hosted, modal]
-    timeout-minutes: 120
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Install Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.11"
-      - name: Install Modal
-        run: |
-          python -m pip install --upgrade pip
-          pip install modal==0.71.8 jinja2
-      - name: Update env vars
-        run: |
-          echo "BASE_TAG=main-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
-          echo "PYTORCH_VERSION=${{ matrix.pytorch}}" >> $GITHUB_ENV
-          echo "AXOLOTL_ARGS=${{ matrix.axolotl_args}}" >> $GITHUB_ENV
-          echo "AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}" >> $GITHUB_ENV
-          echo "CUDA=${{ matrix.cuda }}" >> $GITHUB_ENV
-          echo "N_GPUS=${{ matrix.num_gpus }}" >> $GITHUB_ENV
-          echo "NIGHTLY_BUILD=${{ matrix.nightly_build }}" >> $GITHUB_ENV
-      - name: Run tests job on Modal
-        run: |
-          modal run cicd.multigpu

.github/workflows/nightlies.yml

@@ -1,114 +0,0 @@
-name: docker-nightlies
-
-on:
-  workflow_dispatch:
-  schedule:
-    - cron: '0 0 * * *'  # Runs at 00:00 UTC every day
-
-jobs:
-  build-axolotl:
-    if: ${{ ! contains(github.event.commits[0].message, '[skip docker]') && github.repository_owner == 'axolotl-ai-cloud' }}
-    strategy:
-      fail-fast: false
-      matrix:
-        include:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.4.1
-            axolotl_extras:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.5.1
-            axolotl_extras:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.6.0
-            axolotl_extras:
-    runs-on: axolotl-gpu-runner
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Docker metadata
-        id: metadata
-        uses: docker/metadata-action@v5
-        with:
-          images: |
-            winglian/axolotl
-            axolotlai/axolotl
-          tags: |
-            type=raw,value={{ branch }}-{{ date 'YYYYMMDD' }}
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKERHUB_USERNAME }}
-          password: ${{ secrets.DOCKERHUB_TOKEN }}
-      # guidance for testing before pushing: https://docs.docker.com/build/ci/github-actions/test-before-push/
-      - name: Build and export to Docker
-        uses: docker/build-push-action@v5
-        with:
-          context: .
-          build-args: |
-            BASE_TAG=${{ github.ref_name }}-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}
-            CUDA=${{ matrix.cuda }}
-            PYTORCH_VERSION=${{ matrix.pytorch }}
-            AXOLOTL_ARGS=${{ matrix.axolotl_args }}
-          file: ./docker/Dockerfile
-          push: ${{ github.event_name != 'pull_request' }}
-          tags: |
-            ${{ steps.metadata.outputs.tags }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
-          labels: ${{ steps.metadata.outputs.labels }}
-
-  build-axolotl-cloud:
-    needs: build-axolotl
-    if: ${{ ! contains(github.event.commits[0].message, '[skip docker]') && github.repository_owner == 'axolotl-ai-cloud' }}
-    # this job needs to be run on self-hosted GPU runners...
-    strategy:
-      matrix:
-        include:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.4.1
-            axolotl_extras:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.5.1
-            axolotl_extras:
-    runs-on: axolotl-gpu-runner
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Docker metadata
-        id: metadata
-        uses: docker/metadata-action@v5
-        with:
-          images: |
-            winglian/axolotl-cloud
-            axolotlai/axolotl-cloud
-          tags: |
-            type=raw,value={{ branch }}-{{ date 'YYYYMMDD' }}
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKERHUB_USERNAME }}
-          password: ${{ secrets.DOCKERHUB_TOKEN }}
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-      - name: Build
-        uses: docker/build-push-action@v5
-        with:
-          context: .
-          build-args: |
-            BASE_TAG=${{ github.ref_name }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
-            CUDA=${{ matrix.cuda }}
-          file: ./docker/Dockerfile-cloud
-          push: ${{ github.event_name != 'pull_request' }}
-          tags: |
-             ${{ steps.metadata.outputs.tags }}-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}${{ matrix.axolotl_extras != '' && '-' || '' }}${{ matrix.axolotl_extras }}
-          labels: ${{ steps.metadata.outputs.labels }}

.github/workflows/pypi.yml

@@ -3,27 +3,12 @@ name: publish pypi
 on:
   push:
     tags:
-      - 'v*'
-  workflow_dispatch:
+      - '*'
 
 jobs:
-  setup_release:
-    name: Create Release
-    runs-on: ubuntu-latest
-    permissions:
-      contents: write
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Create release
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        run: gh release create "$GITHUB_REF_NAME" --generate-notes
   pypi-publish:
     name: Upload release to PyPI
     runs-on: ubuntu-latest
-    needs: [setup_release]
     environment:
       name: pypi
       url: https://pypi.org/p/axolotl
@@ -31,30 +16,30 @@ jobs:
       id-token: write  # IMPORTANT: this permission is mandatory for trusted publishing
     steps:
       - name: Check out repository code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v3
 
       - name: Setup Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v4
         with:
-          python-version: "3.11"
+          python-version: "3.10"
 
       - name: Install dependencies
         run: |
-          pip3 install wheel packaging
-          pip3 install --no-build-isolation -e .
-          pip3 install -r requirements-dev.txt -r requirements-tests.txt
+          pip3 install wheel
+          pip3 install -e .
+          pip3 install -r requirements-tests.txt
 
       - name: Extract tag name
         id: tag
         run: echo ::set-output name=TAG_NAME::$(echo $GITHUB_REF | cut -d / -f 3)
 
       - name: Update version in setup.py
-        run: |
+        run: >-
           sed -i -E 's/version="([0-9.]+)",/version="${{ steps.tag.outputs.TAG_NAME }}",/g' setup.py
 
-      - name: Build a source dist
-        run: |
-          python setup.py sdist
+      - name: Build a binary wheel
+        run: >-
+          python setup.py sdist bdist_wheel
 
       - name: Publish package distributions to PyPI
         uses: pypa/gh-action-pypi-publish@release/v1

.github/workflows/tests-nightly.yml

@@ -1,139 +0,0 @@
-name: Tests Nightly against upstream main
-on:
-  workflow_dispatch:
-  schedule:
-    - cron: '0 0 * * *'  # Runs at 00:00 UTC every day
-
-jobs:
-  pre-commit:
-    name: pre-commit
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v4
-      - uses: actions/setup-python@v5
-        with:
-          python-version: "3.11"
-          cache: 'pip' # caching pip dependencies
-      - uses: pre-commit/action@v3.0.1
-        env:
-          SKIP: no-commit-to-branch
-
-  pytest:
-    name: PyTest
-    runs-on: ubuntu-latest
-    strategy:
-      fail-fast: false
-      max-parallel: 2
-      matrix:
-        python_version: ["3.11"]
-        pytorch_version: ["2.4.1", "2.5.1", "2.6.0"]
-    timeout-minutes: 20
-
-    steps:
-      - name: Check out repository code
-        uses: actions/checkout@v4
-
-      - name: Setup Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: ${{ matrix.python_version }}
-          cache: 'pip' # caching pip dependencies
-
-      - name: upgrade pip
-        run: |
-          pip3 install --upgrade pip
-          pip3 install --upgrade packaging setuptools wheel
-
-      - name: Install PyTorch
-        run: |
-          pip3 install torch==${{ matrix.pytorch_version }} --index-url https://download.pytorch.org/whl/cpu
-
-      - name: Update requirements.txt
-        run: |
-          sed -i 's#^transformers.*#transformers @ git+https://github.com/huggingface/transformers.git@main#' requirements.txt
-          sed -i 's#^peft.*#peft @ git+https://github.com/huggingface/peft.git@main#' requirements.txt
-          sed -i 's#^accelerate.*#accelerate @ git+https://github.com/huggingface/accelerate.git@main#' requirements.txt
-          sed -i 's#^trl.*#trl @ git+https://github.com/huggingface/trl.git@main#' requirements.txt
-          sed -i 's#^datasets.*#datasets @ git+https://github.com/huggingface/datasets.git@main#' requirements.txt
-
-      - name: Install dependencies
-        run: |
-          pip3 install --upgrade pip
-          pip3 install --upgrade packaging
-          pip3 install --no-build-isolation -U -e .
-          python scripts/unsloth_install.py | sh
-          python scripts/cutcrossentropy_install.py | sh
-          pip3 install -r requirements-dev.txt -r requirements-tests.txt
-
-      - name: Make sure PyTorch version wasn't clobbered
-        run: |
-          python -c "import torch; assert '${{ matrix.pytorch_version }}' in torch.__version__"
-
-      - name: Ensure axolotl CLI was installed
-        run: |
-          axolotl --help
-
-      - name: Run tests
-        run: |
-          pytest -n8 --dist loadfile --ignore=tests/e2e/ --ignore=tests/patched/ tests/
-          pytest tests/patched/
-
-      - name: cleanup pip cache
-        run: |
-          find "$(pip cache dir)/http-v2" -type f -mtime +14 -exec rm {} \;
-
-  docker-e2e-tests:
-    if: github.repository_owner == 'axolotl-ai-cloud'
-    # this job needs to be run on self-hosted GPU runners...
-    runs-on: [self-hosted, modal]
-    timeout-minutes: 60
-    needs: [pre-commit, pytest]
-
-    strategy:
-      fail-fast: false
-      matrix:
-        include:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.4.1
-            num_gpus: 1
-            axolotl_extras:
-            nightly_build: "true"
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.5.1
-            num_gpus: 1
-            axolotl_extras:
-            nightly_build: "true"
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.6.0
-            num_gpus: 1
-            axolotl_extras:
-            nightly_build: "true"
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Install Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.11"
-      - name: Install Modal
-        run: |
-          python -m pip install --upgrade pip
-          pip install modal==0.71.8 jinja2
-      - name: Update env vars
-        run: |
-          echo "BASE_TAG=main-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
-          echo "PYTORCH_VERSION=${{ matrix.pytorch}}" >> $GITHUB_ENV
-          echo "AXOLOTL_ARGS=${{ matrix.axolotl_args}}" >> $GITHUB_ENV
-          echo "AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}" >> $GITHUB_ENV
-          echo "CUDA=${{ matrix.cuda }}" >> $GITHUB_ENV
-          echo "N_GPUS=${{ matrix.num_gpus }}" >> $GITHUB_ENV
-          echo "NIGHTLY_BUILD=${{ matrix.nightly_build }}" >> $GITHUB_ENV
-      - name: Run tests job on Modal
-        run: |
-          modal run cicd.tests

.github/workflows/tests.yml

@@ -1,277 +1,80 @@
 name: Tests
 on:
   # check on push/merge to main, PRs, and manual triggers
-  merge_group:
   push:
     branches:
       - "main"
     paths:
       - '**.py'
       - 'requirements.txt'
-      - '.github/workflows/*.yml'
-      - 'requirements-tests.txt'
-      - 'cicd/cicd.sh'
-      - 'cicd/Dockerfile.jinja'
   pull_request:
       paths:
        - '**.py'
        - 'requirements.txt'
-       - '.github/workflows/*.yml'
-       - 'requirements-tests.txt'
-       - 'cicd/cicd.sh'
-       - 'cicd/Dockerfile.jinja'
   workflow_dispatch:
 
-# Cancel jobs on the same ref if a new one is triggered
-concurrency:
-  group: ${{ github.workflow }}-${{ github.ref }}
-  cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
-
 jobs:
   pre-commit:
     name: pre-commit
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v4
-      - uses: actions/setup-python@v5
+      - uses: actions/checkout@v3
+      - uses: actions/setup-python@v4
         with:
-          python-version: "3.11"
+          python-version: "3.9"
           cache: 'pip' # caching pip dependencies
-      - uses: pre-commit/action@v3.0.1
-        env:
-          SKIP: no-commit-to-branch
+      - uses: pre-commit/action@v3.0.0
 
   pytest:
     name: PyTest
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
-      max-parallel: 2
       matrix:
-        python_version: ["3.11"]
-        pytorch_version: ["2.4.1", "2.5.1", "2.6.0"]
-    timeout-minutes: 20
+        python_version: ["3.9", "3.10"]
+    timeout-minutes: 10
 
     steps:
       - name: Check out repository code
-        uses: actions/checkout@v4
-
-      - name: Restore HF cache
-        id: hf-cache-restore
-        uses: actions/cache/restore@v4
-        with:
-          path: |
-            /home/runner/.cache/huggingface/hub/datasets--*
-            /home/runner/.cache/huggingface/hub/models--*
-          key: ${{ runner.os }}-hf-hub-cache-${{ hashFiles('**/conftest.py') }}
+        uses: actions/checkout@v3
 
       - name: Setup Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v4
         with:
           python-version: ${{ matrix.python_version }}
           cache: 'pip' # caching pip dependencies
 
-      - name: upgrade pip
-        run: |
-          pip3 install --upgrade pip
-          pip3 install --upgrade packaging setuptools wheel
-
-      - name: Install PyTorch
-        run: |
-          pip3 install torch==${{ matrix.pytorch_version }}
-
       - name: Install dependencies
         run: |
-          pip3 show torch
-          pip3 install --no-build-isolation -U -e .
-          python scripts/unsloth_install.py | sh
-          python scripts/cutcrossentropy_install.py | sh
-          pip3 install -r requirements-dev.txt -r requirements-tests.txt
-
-      - name: Make sure PyTorch version wasn't clobbered
-        run: |
-          python -c "import torch; assert '${{ matrix.pytorch_version }}' in torch.__version__"
-
-      - name: Ensure axolotl CLI was installed
-        run: |
-          axolotl --help
+          pip3 install -U -e .
+          pip3 install -r requirements-tests.txt
 
       - name: Run tests
         run: |
-          pytest -v -n8 --dist loadfile --ignore=tests/e2e/ --ignore=tests/patched/ tests/
-          pytest -v tests/patched/
+          pytest --ignore=tests/e2e/ tests/
 
-      - name: cleanup pip cache
-        run: |
-          find "$(pip cache dir)/http-v2" -type f -mtime +14 -exec rm {} \;
-
-      - name: Save HF cache
-        id: hf-cache
-        uses: actions/cache/save@v4
-        with:
-          path: |
-            /home/runner/.cache/huggingface/hub/datasets--*
-            /home/runner/.cache/huggingface/hub/models--*
-          key: ${{ steps.hf-cache-restore.outputs.cache-primary-key }}
-
-  pytest-sdist:
-    name: PyTest from Source Dist
-    runs-on: ubuntu-latest
-    strategy:
-      fail-fast: false
-      max-parallel: 1
-      matrix:
-        python_version: ["3.11"]
-        pytorch_version: ["2.4.1", "2.5.1", "2.6.0"]
+  e2e-test:
+    name: E2E Tests
+    runs-on: [self-hosted, gpu]
     timeout-minutes: 20
+    needs: [pre-commit, pytest]
 
     steps:
       - name: Check out repository code
-        uses: actions/checkout@v4
-
-      - name: Restore HF cache
-        id: hf-cache-restore
-        uses: actions/cache/restore@v4
-        with:
-          path: |
-            /home/runner/.cache/huggingface/hub/datasets--*
-            /home/runner/.cache/huggingface/hub/models--*
-          key: ${{ runner.os }}-hf-hub-cache-${{ hashFiles('**/conftest.py') }}
+        uses: actions/checkout@v3
 
       - name: Setup Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v4
         with:
-          python-version: ${{ matrix.python_version }}
-          cache: 'pip' # caching pip dependencies
-
-      - name: upgrade pip
-        run: |
-          pip3 install --upgrade pip
-          pip3 install --upgrade packaging setuptools setuptools_scm build wheel
-
-      - name: Install PyTorch
-        run: |
-          pip3 install torch==${{ matrix.pytorch_version }}
+          python-version: "3.10"
+#          cache: 'pip' # caching pip dependencies
 
       - name: Install dependencies
         run: |
-          pip3 show torch
-          python -m build --no-isolation --sdist
-          pip3 install --no-build-isolation dist/axolotl*.tar.gz
-          python scripts/unsloth_install.py | sh
-          python scripts/cutcrossentropy_install.py | sh
-          pip3 install -r requirements-dev.txt -r requirements-tests.txt
+          pip3 uninstall -y transformers accelerate
+          pip3 install -U -e .[flash-attn]
+          pip3 install -r requirements-tests.txt
 
-      - name: Make sure PyTorch version wasn't clobbered
+      - name: Run e2e tests
         run: |
-          python -c "import torch; assert '${{ matrix.pytorch_version }}' in torch.__version__"
-
-      - name: Ensure axolotl CLI was installed
-        run: |
-          axolotl --help
-
-      - name: Run tests
-        run: |
-          pytest -v -n8 --dist loadfile --ignore=tests/e2e/ --ignore=tests/patched/ tests/
-          pytest -v tests/patched/
-
-      - name: cleanup pip cache
-        run: |
-          find "$(pip cache dir)/http-v2" -type f -mtime +14 -exec rm {} \;
-
-      - name: Save HF cache
-        id: hf-cache
-        uses: actions/cache/save@v4
-        with:
-          path: |
-            /home/runner/.cache/huggingface/hub/datasets--*
-            /home/runner/.cache/huggingface/hub/models--*
-          key: ${{ steps.hf-cache-restore.outputs.cache-primary-key }}
-
-  docker-e2e-tests-1st:
-    if: ${{ ! contains(github.event.commits[0].message, '[skip e2e]') && github.repository_owner == 'axolotl-ai-cloud' }}
-    # this job needs to be run on self-hosted GPU runners...
-    runs-on: [self-hosted, modal]
-    timeout-minutes: 90
-    needs: [pre-commit, pytest, pytest-sdist]
-
-    strategy:
-      fail-fast: false
-      matrix:
-        include:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.5.1
-            num_gpus: 1
-            axolotl_extras: vllm
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Install Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.11"
-      - name: Install Modal
-        run: |
-          python -m pip install --upgrade pip
-          pip install modal==0.71.8 jinja2
-      - name: Update env vars
-        run: |
-          echo "BASE_TAG=main-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
-          echo "PYTORCH_VERSION=${{ matrix.pytorch}}" >> $GITHUB_ENV
-          echo "AXOLOTL_ARGS=${{ matrix.axolotl_args}}" >> $GITHUB_ENV
-          echo "AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}" >> $GITHUB_ENV
-          echo "CUDA=${{ matrix.cuda }}" >> $GITHUB_ENV
-          echo "MODAL_IMAGE_BUILDER_VERSION=2024.10" >> $GITHUB_ENV
-          echo "N_GPUS=${{ matrix.num_gpus }}" >> $GITHUB_ENV
-      - name: Run tests job on Modal
-        run: |
-          modal run cicd.tests
-
-  docker-e2e-tests:
-    if: github.repository_owner == 'axolotl-ai-cloud'
-    # this job needs to be run on self-hosted GPU runners...
-    runs-on: [self-hosted, modal]
-    timeout-minutes: 90
-    needs: [pre-commit, pytest, docker-e2e-tests-1st]
-
-    strategy:
-      fail-fast: false
-      matrix:
-        include:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.4.1
-            num_gpus: 1
-            axolotl_extras:
-          - cuda: 124
-            cuda_version: 12.4.1
-            python_version: "3.11"
-            pytorch: 2.6.0
-            num_gpus: 1
-            axolotl_extras:
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Install Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.11"
-      - name: Install Modal
-        run: |
-          python -m pip install --upgrade pip
-          pip install modal==0.71.8 jinja2
-      - name: Update env vars
-        run: |
-          echo "BASE_TAG=main-base-py${{ matrix.python_version }}-cu${{ matrix.cuda }}-${{ matrix.pytorch }}" >> $GITHUB_ENV
-          echo "PYTORCH_VERSION=${{ matrix.pytorch}}" >> $GITHUB_ENV
-          echo "AXOLOTL_ARGS=${{ matrix.axolotl_args}}" >> $GITHUB_ENV
-          echo "AXOLOTL_EXTRAS=${{ matrix.axolotl_extras}}" >> $GITHUB_ENV
-          echo "CUDA=${{ matrix.cuda }}" >> $GITHUB_ENV
-          echo "MODAL_IMAGE_BUILDER_VERSION=2024.10" >> $GITHUB_ENV
-          echo "N_GPUS=${{ matrix.num_gpus }}" >> $GITHUB_ENV
-      - name: Run tests job on Modal
-        run: |
-          modal run cicd.tests
+          pytest tests/e2e/

.gitignore

@@ -1,9 +1,5 @@
 **/axolotl.egg-info
 configs
-last_run_prepared/
-outputs
-.vscode
-_site/
 
 # Byte-compiled / optimized / DLL files
 __pycache__/
@@ -134,7 +130,6 @@ venv/
 ENV/
 env.bak/
 venv.bak/
-venv3.10/
 
 # Spyder project settings
 .spyderproject
@@ -170,19 +165,3 @@ cython_debug/
 # WandB
 # wandb creates a folder to store logs for training runs
 wandb
-
-# Runs
-lora-out/*
-qlora-out/*
-mlruns/*
-
-/.quarto/
-prepared-datasets/
-submit.sh
-*.out*
-
-typings/
-out/
-
-# vim
-*.swp

.isort.cfg

@@ -1,3 +1,3 @@
 [settings]
 profile=black
-known_third_party=wandb,comet_ml
+known_third_party=wandb

.mypy.ini

@@ -1,5 +1,5 @@
 [mypy]
-plugins = pydantic.mypy
+
 exclude = venv
 
 [mypy-alpaca_lora_4bit.*]
@@ -8,12 +8,6 @@ ignore_missing_imports = True
 [mypy-axolotl.monkeypatch.*]
 ignore_errors = True
 
-[mypy-axolotl.models.mixtral.*]
-ignore_errors = True
-
-[mypy-axolotl.integrations.liger.models.*]
-ignore_errors = True
-
 [mypy-axolotl.models.phi.*]
 ignore_errors = True
 
@@ -35,9 +29,6 @@ ignore_missing_imports = True
 [mypy-bitsandbytes]
 ignore_missing_imports = True
 
-[mypy-requests]
-ignore_missing_imports = True
-
 [mypy-datasets]
 ignore_missing_imports = True
 

.pre-commit-config.yaml

@@ -8,8 +8,6 @@ repos:
     -   id: check-yaml
     -   id: end-of-file-fixer
     -   id: trailing-whitespace
-    -   id: no-commit-to-branch
-        args: ['--branch', 'main']
 -   repo: https://github.com/psf/black
     rev: 23.3.0
     hooks:
@@ -19,11 +17,11 @@ repos:
     hooks:
       - id: isort
 -   repo: https://github.com/PyCQA/flake8
-    rev: 6.1.0
+    rev: 6.0.0
     hooks:
     - id: flake8
 -   repo: https://github.com/PyCQA/pylint
-    rev: v3.3.0
+    rev: v2.17.4
     hooks:
     - id: pylint
 -   repo: https://github.com/pre-commit/mirrors-mypy
@@ -33,7 +31,6 @@ repos:
       additional_dependencies:
         [
             'types-PyYAML',
-            'pydantic>=2.5.3',
         ]
 -   repo: https://github.com/PyCQA/bandit
     rev: 1.7.5

.pylintrc

@@ -1,5 +1,5 @@
 [MASTER]
-init-hook="from pylint.config import find_default_config_files; import sys; sys.path.append(next(find_default_config_files()).parent.as_posix())"
+init-hook="from pylint.config import find_pylintrc; import os, sys; sys.path.append(os.path.dirname(find_pylintrc()))"
 
 [TYPECHECK]
 
@@ -12,4 +12,4 @@ generated-members=numpy.*, torch.*
 disable=missing-function-docstring, line-too-long, import-error,
     too-many-arguments, too-many-locals, too-many-statements, too-many-branches, too-few-public-methods,
     too-many-instance-attributes, fixme, import-outside-toplevel, logging-fstring-interpolation,
-    too-many-positional-arguments, possibly-used-before-assignment
+    too-many-nested-blocks,

.vscode/README.md

@@ -1 +0,0 @@
-See [docs/debugging.md](../docs/debugging.md) for guidance on how to modify these files to debug axolotl with VSCode.

.vscode/launch.json

@@ -1,34 +0,0 @@
-{
-    // Use IntelliSense to learn about possible attributes.
-    // Hover to view descriptions of existing attributes.
-    // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
-    "version": "0.2.0",
-    "configurations": [
-        {
-            "name": "Debug axolotl prompt - sharegpt",
-            "type": "python",
-            "module": "accelerate.commands.launch",
-            "request": "launch",
-            "args": [
-                "-m", "axolotl.cli.train", "dev_sharegpt.yml",
-                // The flags below simplify debugging by overriding the axolotl config
-                // with the debugging tips above.  Modify as needed.
-                "--dataset_processes=1",      // limits data preprocessing to one process
-                "--max_steps=1",              // limits training to just one step
-                "--batch_size=1",             // minimizes batch size
-                "--micro_batch_size=1",       // minimizes batch size
-                "--val_set_size=0",           // disables validation
-                "--sample_packing=False",     // disables sample packing which is necessary for small datasets
-                "--eval_sample_packing=False",// disables sample packing on eval set
-                "--dataset_prepared_path=temp_debug/axolotl_outputs/data", // send data outputs to a temp folder
-                "--output_dir=temp_debug/axolotl_outputs/model" // send model outputs to a temp folder
-                ],
-            "console": "integratedTerminal",      // show output in the integrated terminal
-            "cwd": "${workspaceFolder}/devtools", // set working directory to devtools from the root of the project
-            "justMyCode": true,                   // step through only axolotl code
-            "env": {"CUDA_VISIBLE_DEVICES": "0",  // Since we aren't doing distributed training, we need to limit to one GPU
-                    "HF_HOME": "${workspaceFolder}/devtools/temp_debug/.hf-cache"}, // send HF cache to a temp folder
-            "preLaunchTask": "cleanup-for-dataprep", // delete temp folders (see below)
-        }
-    ]
-}

.vscode/tasks.json

@@ -1,27 +0,0 @@
-//this file is used by launch.json
-{
-  "version": "2.0.0",
-  "tasks": [
-    // this task changes into the devtools directory and deletes the temp_debug/axolotl_outputs folder
-    {
-      "label": "delete-outputs",
-      "type": "shell",
-      "command": "rm -rf temp_debug/axolotl_outputs",
-      "options":{ "cwd": "${workspaceFolder}/devtools"},
-      "problemMatcher": []
-    },
-    // this task changes into the devtools directory and deletes the `temp_debug/.hf-cache/datasets` folder
-    {
-      "label": "delete-temp-hf-dataset-cache",
-      "type": "shell",
-      "command": "rm -rf temp_debug/.hf-cache/datasets",
-      "options":{ "cwd": "${workspaceFolder}/devtools"},
-      "problemMatcher": []
-    },
-      // this task combines the two tasks above
-    {
-     "label": "cleanup-for-dataprep",
-     "dependsOn": ["delete-outputs", "delete-temp-hf-dataset-cache"],
-    }
-  ]
-}

MANIFEST.in

@@ -1,5 +0,0 @@
-include requirements.txt
-include README.md
-include LICENSE
-include src/setuptools_axolotl_dynamic_dependencies.py
-recursive-include axolotl *.py

README.md

@@ -1,151 +1,882 @@
-<p align="center">
-    <picture>
-        <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/887513285d98132142bf5db2a74eb5e0928787f1/image/axolotl_logo_digital_white.svg">
-        <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/887513285d98132142bf5db2a74eb5e0928787f1/image/axolotl_logo_digital_black.svg">
-        <img alt="Axolotl" src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/887513285d98132142bf5db2a74eb5e0928787f1/image/axolotl_logo_digital_black.svg" width="400" height="104" style="max-width: 100%;">
-    </picture>
-</p>
+# Axolotl
 
-<p align="center">
-    <img src="https://img.shields.io/github/license/axolotl-ai-cloud/axolotl.svg?color=blue" alt="GitHub License">
-    <img src="https://github.com/axolotl-ai-cloud/axolotl/actions/workflows/tests.yml/badge.svg" alt="tests">
-    <a href="https://github.com/axolotl-ai-cloud/axolotl/releases"><img src="https://img.shields.io/github/release/axolotl-ai-cloud/axolotl.svg" alt="Releases"></a>
-    <br/>
-    <a href="https://github.com/axolotl-ai-cloud/axolotl/graphs/contributors"><img src="https://img.shields.io/github/contributors-anon/axolotl-ai-cloud/axolotl?color=yellow&style=flat-square" alt="contributors" style="height: 20px;"></a>
-    <img src="https://img.shields.io/github/stars/axolotl-ai-cloud/axolotl" alt="GitHub Repo stars">
-    <br/>
-    <a href="https://discord.com/invite/HhrNrHJPRb"><img src="https://img.shields.io/badge/discord-7289da.svg?style=flat-square&logo=discord" alt="discord" style="height: 20px;"></a>
-    <a href="https://twitter.com/axolotl_ai"><img src="https://img.shields.io/twitter/follow/axolotl_ai?style=social" alt="twitter" style="height: 20px;"></a>
-    <br/>
-    <img src="https://github.com/axolotl-ai-cloud/axolotl/actions/workflows/tests-nightly.yml/badge.svg" alt="tests-nightly">
-    <img src="https://github.com/axolotl-ai-cloud/axolotl/actions/workflows/multi-gpu-e2e.yml/badge.svg" alt="multigpu-semi-weekly tests">
-    <a href="https://www.phorm.ai/query?projectId=e315ba4a-4e14-421f-ab05-38a1f9076f25">
-    <img alt="phorm.ai" src="https://img.shields.io/badge/Phorm-Ask_AI-%23F2777A.svg?&logo=data:image/svg+xml;base64,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">
-  </a>
-</p>
-
-Axolotl is a tool designed to streamline post-training for various AI models.
-Post-training refers to any modifications or additional training performed on
-pre-trained models - including full model fine-tuning, parameter-efficient tuning (like
-LoRA and QLoRA), supervised fine-tuning (SFT), instruction tuning, and alignment
-techniques. With support for multiple model architectures and training configurations,
-Axolotl makes it easy to get started with these techniques.
-
-Axolotl is designed to work with YAML config files that contain everything you need to
-preprocess a dataset, train or fine-tune a model, run model inference or evaluation,
-and much more.
+Axolotl is a tool designed to streamline the fine-tuning of various AI models, offering support for multiple configurations and architectures.
 
 Features:
-
 - Train various Huggingface models such as llama, pythia, falcon, mpt
 - Supports fullfinetune, lora, qlora, relora, and gptq
 - Customize configurations using a simple yaml file or CLI overwrite
 - Load different dataset formats, use custom formats, or bring your own tokenized datasets
-- Integrated with [xformers](https://github.com/facebookresearch/xformers), flash attention, [liger kernel](https://github.com/linkedin/Liger-Kernel), rope scaling, and multipacking
+- Integrated with xformer, flash attention, rope scaling, and multipacking
 - Works with single GPU or multiple GPUs via FSDP or Deepspeed
 - Easily run with Docker locally or on the cloud
-- Log results and optionally checkpoints to wandb, mlflow or Comet
+- Log results and optionally checkpoints to wandb
 - And more!
 
-## 🚀 Quick Start
 
-**Requirements**:
-- NVIDIA GPU (Ampere or newer for `bf16` and Flash Attention) or AMD GPU
-- Python 3.11
-- PyTorch ≥2.4.1
+<table>
+<tr>
+<td>
 
-### Installation
+## Table of Contents
+- [Introduction](#axolotl)
+- [Supported Features](#axolotl-supports)
+- [Quickstart](#quickstart-)
+- [Installation](#installation)
+  - [Docker Installation](#environment)
+  - [Conda/Pip venv Installation](#condapip-venv)
+  - [LambdaLabs Installation](#lambdalabs)
+- [Dataset](#dataset)
+  - [How to Add Custom Prompts](#how-to-add-custom-prompts)
+  - [How to Use Custom Pretokenized Dataset](#how-to-use-your-custom-pretokenized-dataset)
+- [Config](#config)
+  - [Train](#train)
+  - [Training w/ Deepspeed](#training-with-deepspeed)
+  - [Inference](#inference)
+  - [Merge LORA to Base](#merge-lora-to-base)
+- [Common Errors](#common-errors-)
+- [Need Help?](#need-help-)
+- [Badge](#badge-)
+- [Community Showcase](#community-showcase)
+- [Contributing](#contributing-)
 
-```shell
-pip3 install --no-build-isolation axolotl[flash-attn,deepspeed]
+</td>
+<td>
 
-# Download example axolotl configs, deepspeed configs
-axolotl fetch examples
-axolotl fetch deepspeed_configs  # OPTIONAL
+<div align="center">
+  <img src="image/axolotl.png" alt="axolotl" width="160">
+  <div>
+    <p>
+      <b>Axolotl provides a unified repository for fine-tuning <br />a variety of AI models with ease</b>
+    </p>
+    <p>
+      Go ahead and axolotl questions!!
+    </p>
+    <img src="https://github.com/OpenAccess-AI-Collective/axolotl/actions/workflows/pre-commit.yml/badge.svg?branch=main" alt="pre-commit">
+    <img alt="PyTest Status" src="https://github.com/OpenAccess-AI-Collective/axolotl/actions/workflows/tests.yml/badge.svg?branch=main">
+  </div>
+</div>
+
+</td>
+</tr>
+</table>
+
+## Axolotl supports
+
+|          | fp16/fp32 | lora | qlora | gptq | gptq w/flash attn | flash attn | xformers attn |
+|----------|:----------|:-----|-------|------|-------------------|------------|--------------|
+| llama    | ✅         | ✅    | ✅     | ✅             | ✅                 | ✅          | ✅            |
+| Pythia   | ✅         | ✅    | ✅     | ❌             | ❌                 | ❌          | ❓            |
+| cerebras | ✅         | ✅    | ✅     | ❌             | ❌                 | ❌          | ❓            |
+| btlm     | ✅         | ✅    | ✅     | ❌             | ❌                 | ❌          | ❓            |
+| mpt      | ✅         | ❌    | ❓     | ❌             | ❌                 | ❌          | ❓            |
+| falcon   | ✅         | ✅    | ✅     | ❌             | ❌                 | ❌          | ❓            |
+| gpt-j    | ✅         | ✅    | ✅     | ❌             | ❌                 | ❓          | ❓            |
+| XGen     | ✅         | ❓    | ✅     | ❓             | ❓                 | ❓          | ✅            |
+| phi      | ✅         | ✅    | ✅     | ❓             | ❓                 | ❓          | ❓            |
+
+
+## Quickstart ⚡
+
+Get started with Axolotl in just a few steps! This quickstart guide will walk you through setting up and running a basic fine-tuning task.
+
+**Requirements**: Python >=3.9 and Pytorch >=2.0.
+
+```bash
+git clone https://github.com/OpenAccess-AI-Collective/axolotl
+cd axolotl
+
+pip3 install packaging
+pip3 install -e '.[flash-attn,deepspeed]'
+pip3 install -U git+https://github.com/huggingface/peft.git
+
+# finetune lora
+accelerate launch -m axolotl.cli.train examples/openllama-3b/lora.yml
+
+# inference
+accelerate launch -m axolotl.cli.inference examples/openllama-3b/lora.yml \
+    --lora_model_dir="./lora-out"
 ```
 
-Other installation approaches are described [here](https://axolotl-ai-cloud.github.io/axolotl/docs/installation.html).
+## Installation
 
-### Your First Fine-tune
+### Environment
 
-```shell
-# Fetch axolotl examples
-axolotl fetch examples
+- Docker
+  ```bash
+  docker run --gpus '"all"' --rm -it winglian/axolotl:main-py3.10-cu118-2.0.1
+  ```
+  - `winglian/axolotl-runpod:main-latest`: for runpod or use this [direct link](https://runpod.io/gsc?template=v2ickqhz9s&ref=6i7fkpdz)
 
-# Or, specify a custom path
-axolotl fetch examples --dest path/to/folder
+  Or run on the current files for development:
 
-# Train a model using LoRA
-axolotl train examples/llama-3/lora-1b.yml
+  ```sh
+  docker compose up -d
+  ```
+
+- Conda/Pip venv
+  1. Install python >=**3.9**
+
+  2. Install pytorch stable https://pytorch.org/get-started/locally/
+
+  3. Install axolotl along with python dependencies
+        ```bash
+        pip3 install packaging
+        pip3 install -e '.[flash-attn,deepspeed]'
+        ```
+  4. (Optional) Login to Huggingface to use gated models/datasets.
+        ```bash
+        huggingface-cli login
+        ```
+        Get the token at huggingface.co/settings/tokens
+
+- LambdaLabs
+  <details>
+
+  <summary>Click to Expand</summary>
+
+  1. Install python
+  ```bash
+  sudo apt update
+  sudo apt install -y python3.9
+
+  sudo update-alternatives --install /usr/bin/python python /usr/bin/python3.9 1
+  sudo update-alternatives --config python # pick 3.9 if given option
+  python -V # should be 3.9
+
+  ```
+
+  2. Install pip
+  ```bash
+  wget https://bootstrap.pypa.io/get-pip.py
+  python get-pip.py
+  ```
+
+  3. Install torch
+  ```bash
+  pip3 install -U torch --index-url https://download.pytorch.org/whl/cu118
+  ```
+
+  4. Axolotl
+  ```bash
+  git clone https://github.com/OpenAccess-AI-Collective/axolotl
+  cd axolotl
+
+  pip3 install packaging
+  pip3 install -e '.[flash-attn,deepspeed]'
+  pip3 install protobuf==3.20.3
+  pip3 install -U --ignore-installed requests Pillow psutil scipy
+  ```
+
+  5. Set path
+  ```bash
+  export LD_LIBRARY_PATH=/usr/lib/x86_64-linux-gnu:$LD_LIBRARY_PATH
+  ```
+  </details>
+
+- Windows: Please use WSL or Docker!
+
+### Dataset
+
+Axolotl supports a variety of dataset formats. Below are some of the formats you can use.
+Have dataset(s) in one of the following format (JSONL recommended):
+
+- `alpaca`: instruction; input(optional)
+  ```json
+  {"instruction": "...", "input": "...", "output": "..."}
+  ```
+- `sharegpt`: conversations where `from` is `human`/`gpt`
+  ```json
+  {"conversations": [{"from": "...", "value": "..."}]}
+  ```
+- `completion`: raw corpus
+  ```json
+  {"text": "..."}
+  ```
+
+<details>
+
+<summary>See other formats</summary>
+
+- `jeopardy`: question and answer
+  ```json
+  {"question": "...", "category": "...", "answer": "..."}
+  ```
+- `oasst`: instruction
+  ```json
+  {"INSTRUCTION": "...", "RESPONSE": "..."}
+  ```
+- `gpteacher`: instruction; input(optional)
+  ```json
+  {"instruction": "...", "input": "...", "response": "..."}
+  ```
+- `reflection`: instruction with reflect; input(optional)
+  ```json
+  {"instruction": "...", "input": "...", "output": "...", "reflection": "...", "corrected": "..."}
+  ```
+- `explainchoice`: question, choices, (solution OR explanation)
+  ```json
+  {"question": "...", "choices": ["..."], "solution": "...", "explanation": "..."}
+  ```
+- `concisechoice`: question, choices, (solution OR explanation)
+  ```json
+  {"question": "...", "choices": ["..."], "solution": "...", "explanation": "..."}
+  ```
+- `summarizetldr`: article and summary
+  ```json
+  {"article": "...", "summary": "..."}
+  ```
+- `alpaca_chat`: basic instruct for alpaca chat
+  ```json
+  {"instruction": "...", "input": "...", "response": "..."}
+  ```
+- `alpaca_chat.load_qa`: question and answer for alpaca chat
+  ```json
+  {"question": "...", "answer": "..."}
+  ```
+- `alpaca_chat.load_concise`: question and answer for alpaca chat, for concise answers
+  ```json
+  {"instruction": "...", "input": "...", "response": "..."}
+  ```
+- `alpaca_chat.load_camel_ai`: question and answer for alpaca chat, for load_camel_ai
+  ```json
+  {"message_1": "...", "message_2": "..."}
+  ```
+- `alpaca_w_system.load_open_orca`: support for open orca datasets with included system prompts, instruct
+  ```json
+  {"system_prompt": "...", "question": "...", "response": "..."}
+  ```
+- `context_qa`: in context question answering from an article
+  ```json
+  {"article": "...", "question": "...", "answer": "..."}
+  ```
+- `context_qa.load_v2`: in context question answering (alternate)
+  ```json
+  {"context": "...", "question": "...", "answer": "..."}
+  ```
+- `context_qa.load_404`: in context question answering from an article, with default response for no answer from context
+  ```json
+  {"article": "...", "unanswerable_question": "..."}
+  ```
+- `creative_acr.load_answer`: instruction and revision
+  ```json
+  {"instruction": "...", "revision": "..."}
+  ```
+- `creative_acr.load_critique`: critique
+  ```json
+  {"scores": "...", "critiques": "...", "instruction": "...", "answer": "..."}
+  ```
+- `creative_acr.load_revise`: critique and revise
+  ```json
+  {"scores": "...", "critiques": "...", "instruction": "...", "answer": "...", "revision": "..."}
+  ```
+- `pygmalion`: pygmalion
+  ```json
+  {"conversations": [{"role": "...", "value": "..."}]}
+  ```
+- `metharme`: instruction, adds additional eos tokens
+  ```json
+  {"prompt": "...", "generation": "..."}
+  ```
+- `sharegpt.load_role`: conversations where `role` is used instead of `from`
+  ```json
+  {"conversations": [{"role": "...", "value": "..."}]}
+  ```
+- `sharegpt.load_guanaco`: conversations where `from` is `prompter`/`assistant` instead of default sharegpt
+  ```json
+  {"conversations": [{"from": "...", "value": "..."}]}
+  ```
+- `sharegpt_jokes`: creates a chat where bot is asked to tell a joke, then explain why the joke is funny
+  ```json
+  {"conversations": [{"title": "...", "text": "...", "explanation": "..."}]}
+  ```
+
+</details>
+
+#### How to add custom prompts
+
+Using yaml. Example:
+```yaml
+datasets:
+  - path: repo
+    type:
+      system_prompt: ""
+      no_input_format: |-
+        User: {instruction}<|end_of_turn|>
+        Assistant:
+      format: |-
+        User: {instruction}
+        {input}<|end_of_turn|>
+        Assistant:
 ```
 
-That's it! Check out our [Getting Started Guide](https://axolotl-ai-cloud.github.io/axolotl/docs/getting-started.html) for a more detailed walkthrough.
+Using file:
+1. Add your method to a file in [prompt_strategies](src/axolotl/prompt_strategies). Please see other files as example.
+2. Use your custom file name as the dataset type `<prompt_strategies_file>.load_<load_fn>`.
 
-## ✨ Key Features
+#### How to use your custom pretokenized dataset
 
-- **Multiple Model Support**: Train various models like LLaMA, Mistral, Mixtral, Pythia, and more
-- **Training Methods**: Full fine-tuning, LoRA, QLoRA, and more
-- **Easy Configuration**: Simple YAML files to control your training setup
-- **Performance Optimizations**: Flash Attention, xformers, multi-GPU training
-- **Flexible Dataset Handling**: Use various formats and custom datasets
-- **Cloud Ready**: Run on cloud platforms or local hardware
+- Do not pass a `type:`
+- Columns in Dataset must be exactly `input_ids`, `attention_mask`, `labels`
 
-## 📚 Documentation
 
-- [Installation Options](https://axolotl-ai-cloud.github.io/axolotl/docs/installation.html) - Detailed setup instructions for different environments
-- [Configuration Guide](https://axolotl-ai-cloud.github.io/axolotl/docs/config.html) - Full configuration options and examples
-- [Dataset Guide](https://axolotl-ai-cloud.github.io/axolotl/docs/dataset-formats/) - Supported formats and how to use them
-- [Multi-GPU Training](https://axolotl-ai-cloud.github.io/axolotl/docs/multi-gpu.html)
-- [Multi-Node Training](https://axolotl-ai-cloud.github.io/axolotl/docs/multi-node.html)
-- [Multipacking](https://axolotl-ai-cloud.github.io/axolotl/docs/multipack.html)
-- [FAQ](https://axolotl-ai-cloud.github.io/axolotl/docs/faq.html) - Frequently asked questions
+### Config
 
-## 🤝 Getting Help
+See [examples](examples) for quick start. It is recommended to duplicate and modify to your needs. The most important options are:
 
-- Join our [Discord community](https://discord.gg/HhrNrHJPRb) for support
-- Check out our [Examples](https://github.com/axolotl-ai-cloud/axolotl/tree/main/examples/) directory
-- Read our [Debugging Guide](https://axolotl-ai-cloud.github.io/axolotl/docs/debugging.html)
-- Need dedicated support? Please contact [✉️wing@axolotl.ai](mailto:wing@axolotl.ai) for options
+- model
+  ```yaml
+  base_model: ./llama-7b-hf # local or huggingface repo
+  ```
+  Note: The code will load the right architecture.
 
-## 🌟 Contributing
+- dataset
+  ```yaml
+  sequence_len: 2048 # max token length for prompt
 
-Contributions are welcome! Please see our [Contributing Guide](https://github.com/axolotl-ai-cloud/axolotl/blob/main/.github/CONTRIBUTING.md) for details.
+  # huggingface repo
+  datasets:
+    - path: vicgalle/alpaca-gpt4
+      type: alpaca # format from earlier
 
-## Supported Models
+  # huggingface repo with specific configuration/subset
+  datasets:
+    - path: EleutherAI/pile
+      name: enron_emails
+      type: completion # format from earlier
+      field: text # Optional[str] default: text, field to use for completion data
 
-|             | fp16/fp32 | lora | qlora | gptq | gptq w/flash attn | flash attn | xformers attn |
-|-------------|:----------|:-----|-------|------|-------------------|------------|--------------|
-| llama       | ✅         | ✅    | ✅     | ✅             | ✅                 | ✅          | ✅            |
-| Mistral     | ✅         | ✅    | ✅     | ✅             | ✅                 | ✅          | ✅            |
-| Mixtral-MoE | ✅         | ✅    | ✅     | ❓             | ❓                 | ❓          | ❓            |
-| Mixtral8X22 | ✅         | ✅    | ✅     | ❓             | ❓                 | ❓          | ❓            |
-| Pythia      | ✅         | ✅    | ✅     | ❌             | ❌                 | ❌          | ❓            |
-| cerebras    | ✅         | ✅    | ✅     | ❌             | ❌                 | ❌          | ❓            |
-| btlm        | ✅         | ✅    | ✅     | ❌             | ❌                 | ❌          | ❓            |
-| mpt         | ✅         | ❌    | ❓     | ❌             | ❌                 | ❌          | ❓            |
-| falcon      | ✅         | ✅    | ✅     | ❌             | ❌                 | ❌          | ❓            |
-| gpt-j       | ✅         | ✅    | ✅     | ❌             | ❌                 | ❓          | ❓            |
-| XGen        | ✅         | ❓    | ✅     | ❓             | ❓                 | ❓          | ✅            |
-| phi         | ✅         | ✅    | ✅     | ❓             | ❓                 | ❓          | ❓            |
-| RWKV        | ✅         | ❓    | ❓     | ❓             | ❓                 | ❓          | ❓            |
-| Qwen        | ✅         | ✅    | ✅     | ❓             | ❓                 | ❓          | ❓            |
-| Gemma       | ✅         | ✅    | ✅     | ❓             | ❓                 | ✅          | ❓            |
-| Jamba       | ✅         | ✅    | ✅     | ❓             | ❓                 | ✅          | ❓            |
+  # huggingface repo with multiple named configurations/subsets
+  datasets:
+    - path: bigcode/commitpackft
+      name:
+        - ruby
+        - python
+        - typescript
+      type: ... # unimplemented custom format
 
-✅: supported
-❌: not supported
-❓: untested
+  # local
+  datasets:
+    - path: data.jsonl # or json
+      ds_type: json # see other options below
+      type: alpaca
 
-## ❤️ Sponsors
+  # dataset with splits, but no train split
+  dataset:
+    - path: knowrohit07/know_sql
+      type: context_qa.load_v2
+      train_on_split: validation
+  ```
 
-Thank you to our sponsors who help make Axolotl possible:
+- loading
+  ```yaml
+  load_in_4bit: true
+  load_in_8bit: true
+  bf16: true # require >=ampere
+  fp16: true
+  tf32: true # require >=ampere
+  bfloat16: true # require >=ampere, use instead of bf16 when you don't want AMP (automatic mixed precision)
+  float16: true # use instead of fp16 when you don't want AMP
+  ```
+  Note: Repo does not do 4-bit quantization.
 
-- [Modal](https://www.modal.com?utm_source=github&utm_medium=github&utm_campaign=axolotl) - Modal lets you run
-jobs in the cloud, by just writing a few lines of Python. Customers use Modal to deploy Gen AI models at large scale,
-fine-tune large language models, run protein folding simulations, and much more.
+- lora
+  ```yaml
+  adapter: lora # qlora or leave blank for full finetune
+  lora_r: 8
+  lora_alpha: 16
+  lora_dropout: 0.05
+  lora_target_modules:
+    - q_proj
+    - v_proj
+  ```
 
-Interested in sponsoring? Contact us at [wing@axolotl.ai](mailto:wing@axolotl.ai)
+<details>
 
-## 📜 License
+<summary>All yaml options</summary>
 
-This project is licensed under the Apache 2.0 License - see the [LICENSE](LICENSE) file for details.
+```yaml
+# this is the huggingface model that contains *.pt, *.safetensors, or *.bin files
+# this can also be a relative path to a model on disk
+base_model: ./llama-7b-hf
+# you can specify an ignore pattern if the model repo contains more than 1 model type (*.pt, etc)
+base_model_ignore_patterns:
+# if the base_model repo on hf hub doesn't include configuration .json files,
+# you can set that here, or leave this empty to default to base_model
+base_model_config: ./llama-7b-hf
+# you can specify to choose a specific model revision from huggingface hub
+model_revision:
+# Optional tokenizer configuration override in case you want to use a different tokenizer
+# than the one defined in the base model
+tokenizer_config:
+# If you want to specify the type of model to load, AutoModelForCausalLM is a good choice too
+model_type: AutoModelForCausalLM
+# Corresponding tokenizer for the model AutoTokenizer is a good choice
+tokenizer_type: AutoTokenizer
+# Trust remote code for untrusted source
+trust_remote_code:
+# use_fast option for tokenizer loading from_pretrained, default to True
+tokenizer_use_fast:
+# Whether to use the legacy tokenizer setting, defaults to True
+tokenizer_legacy:
+# resize the model embeddings when new tokens are added to multiples of 32
+# this is reported to improve training speed on some models
+resize_token_embeddings_to_32x:
+
+# used to identify which the model is based on
+is_falcon_derived_model:
+is_llama_derived_model:
+is_mistral_derived_model:
+
+# whether you are training a 4-bit GPTQ quantized model
+gptq: true
+gptq_groupsize: 128 # group size
+gptq_model_v1: false # v1 or v2
+
+# this will attempt to quantize the model down to 8 bits and use adam 8 bit optimizer
+load_in_8bit: true
+# use bitsandbytes 4 bit
+load_in_4bit:
+
+# Use CUDA bf16
+bf16: true # bool or 'full' for `bf16_full_eval`. require >=ampere
+# Use CUDA fp16
+fp16: true
+# Use CUDA tf32
+tf32: true # require >=ampere
+
+# No AMP (automatic mixed precision)
+bfloat16: true # require >=ampere
+float16: true
+
+# a list of one or more datasets to finetune the model with
+datasets:
+  # hf dataset repo | "json" for local dataset, make sure to fill data_files
+  - path: vicgalle/alpaca-gpt4
+  # The type of prompt to use for training. [alpaca, sharegpt, gpteacher, oasst, reflection]
+    type: alpaca # format | format:<prompt_style> (chat/instruct) | <prompt_strategies>.load_<load_fn>
+    ds_type: # Optional[str] (json|arrow|parquet|text|csv) defines the datatype when path is a file
+    data_files: # Optional[str] path to source data files
+    shards: # Optional[int] number of shards to split data into
+    name: # Optional[str] name of dataset configuration to load
+    conversation:  # Optional[str] fastchat conversation type, only used with type: sharegpt
+
+  # custom user prompt
+  - path: repo
+    type:
+      # the below are defaults. only set what's needed.
+      system_prompt: ""
+      field_system: system
+      field_instruction: instruction
+      field_output: input
+
+      # customizable to be single line or multi-line
+      system_format: "{system}"
+      # 'format' can include {input}
+      format: |-
+        User: {instruction} {input}
+        Assistant:
+      # 'no_input_format' cannot include {input}
+      no_input_format: "{instruction} "
+
+      # for completions datsets, uses the provided field if not `text`
+      field:
+
+# axolotl attempts to save the dataset as an arrow after packing the data together so
+# subsequent training attempts load faster, relative path
+dataset_prepared_path: data/last_run_prepared
+# push prepared dataset to hub
+push_dataset_to_hub: # repo path
+# The maximum number of processes to use while preprocessing your input dataset. This defaults to `os.cpu_count()`
+# if not set.
+dataset_processes: # defaults to os.cpu_count() if not set
+# push checkpoints to hub
+hub_model_id: # repo path to push finetuned model
+# how to push checkpoints to hub
+# https://huggingface.co/docs/transformers/v4.31.0/en/main_classes/trainer#transformers.TrainingArguments.hub_strategy
+hub_strategy:
+# whether to use hf `use_auth_token` for loading datasets. Useful for fetching private datasets
+# required to be true when used in combination with `push_dataset_to_hub`
+hf_use_auth_token: # boolean
+# How much of the dataset to set aside as evaluation. 1 = 100%, 0.50 = 50%, etc. 0 for no eval.
+val_set_size: 0.04
+# Num shards for whole dataset
+dataset_shard_num:
+# Index of shard to use for whole dataset
+dataset_shard_idx:
+
+# the maximum length of an input to train with, this should typically be less than 2048
+# as most models have a token/context limit of 2048
+sequence_len: 2048
+# pad inputs so each step uses constant sized buffers
+# this will reduce memory fragmentation and may prevent OOMs, by re-using memory more efficiently
+pad_to_sequence_len:
+# max sequence length to concatenate training samples together up to
+# inspired by StackLLaMA. see https://huggingface.co/blog/stackllama#supervised-fine-tuning
+# FutureWarning: This will soon be DEPRECATED
+max_packed_sequence_len: 1024
+# use efficient multi-packing with block diagonal attention and per sequence position_ids. Recommend set to 'true'
+sample_packing:
+# set to 'false' if getting errors during eval with sample_packing on.
+eval_sample_packing:
+# you can set these packing optimizations AFTER starting a training at least once.
+# The trainer will provide recommended values for these values.
+sample_packing_eff_est:
+total_num_tokens:
+
+# if you want to use 'lora' or 'qlora' or leave blank to train all parameters in original model
+adapter: lora
+# if you already have a lora model trained that you want to load, put that here
+# lora hyperparameters
+lora_model_dir:
+lora_r: 8
+lora_alpha: 16
+lora_dropout: 0.05
+lora_target_modules:
+  - q_proj
+  - v_proj
+#  - k_proj
+#  - o_proj
+#  - gate_proj
+#  - down_proj
+#  - up_proj
+lora_target_linear: # if true, will target all linear layers
+lora_modules_to_save:
+#  - embed_tokens
+#  - lm_head
+lora_out_dir:
+lora_fan_in_fan_out: false
+
+# ReLoRA configuration
+# must use either 'lora' or 'qlora' adapter, and does not support fsdp or deepspeed
+relora_steps: # number of steps per ReLoRA restart
+relora_warmup_steps: # number of per-restart warmup steps
+relora_cpu_offload: # true to perform lora weight merges on cpu during restarts, for modest gpu memory savings
+
+# wandb configuration if you're using it
+wandb_mode: # "offline" to save run metadata locally and not sync to the server, "disabled" to turn off wandb
+wandb_project: # your wandb project name
+wandb_entity: # a wandb Team name if using a Team
+wandb_watch:
+wandb_run_id: # set the name of your wandb run
+wandb_log_model: # "checkpoint" to log model to wandb Artifacts every `save_steps` or "end" to log only at the end of training
+
+# where to save the finished model to
+output_dir: ./completed-model
+
+# whether to use torch.compile and which backend to use
+torch_compile:  # bool
+torch_compile_backend:  # Optional[str]
+
+# training hyperparameters
+gradient_accumulation_steps: 1
+micro_batch_size: 2
+eval_batch_size:
+num_epochs: 3
+warmup_steps: 100
+learning_rate: 0.00003
+lr_quadratic_warmup:
+logging_steps:
+save_strategy: # set to `no` to skip checkpoint saves
+save_steps: # leave empty to save at each epoch
+eval_steps: # leave empty to eval at each epoch
+save_total_limit: # checkpoints saved at a time
+max_steps:
+
+eval_table_size: # approximate number of predictions sent to wandb depending on batch size. Enabled above 0. Default is 0
+eval_table_max_new_tokens: # total number of tokens generated for predictions sent to wandb. Default is 128
+
+# save model as safetensors (require safetensors package)
+save_safetensors:
+
+# whether to mask out or include the human's prompt from the training labels
+train_on_inputs: false
+# group similarly sized data to minimize padding
+# may be slower to start, as it must download and sort the entire dataset
+# note that training loss may have an oscillating pattern with this enabled
+group_by_length: false
+
+# Whether to use gradient checkpointing https://huggingface.co/docs/transformers/v4.18.0/en/performance#gradient-checkpointing
+gradient_checkpointing: false
+
+# stop training after this many evaluation losses have increased in a row
+# https://huggingface.co/transformers/v4.2.2/_modules/transformers/trainer_callback.html#EarlyStoppingCallback
+early_stopping_patience: 3
+
+# specify a scheduler and kwargs to use with the optimizer
+lr_scheduler: # 'one_cycle' | 'log_sweep' | empty for cosine
+lr_scheduler_kwargs:
+
+# for one_cycle optim
+lr_div_factor: # learning rate div factor
+
+# for log_sweep optim
+log_sweep_min_lr:
+log_sweep_max_lr:
+
+# specify optimizer
+# Valid values are driven by the Transformers OptimizerNames class, see:
+# https://github.com/huggingface/transformers/blob/95b374952dc27d8511541d6f5a4e22c9ec11fb24/src/transformers/training_args.py#L134
+#
+# Note that not all optimizers may be available in your environment, ex: 'adamw_anyprecision' is part of
+# torchdistx, 'adamw_bnb_8bit' is part of bnb.optim.Adam8bit, etc. When in doubt, it is recommended to start with the optimizer used
+# in the examples/ for your model and fine-tuning use case.
+#
+# Valid values for 'optimizer' include:
+# - adamw_hf
+# - adamw_torch
+# - adamw_torch_fused
+# - adamw_torch_xla
+# - adamw_apex_fused
+# - adafactor
+# - adamw_anyprecision
+# - sgd
+# - adagrad
+# - adamw_bnb_8bit
+# - lion_8bit
+# - lion_32bit
+# - paged_adamw_32bit
+# - paged_adamw_8bit
+# - paged_lion_32bit
+# - paged_lion_8bit
+optimizer:
+# specify weight decay
+weight_decay:
+# adamw hyperparams
+adam_beta1:
+adam_beta2:
+adam_epsilon:
+# Gradient clipping max norm
+max_grad_norm:
+
+# whether to bettertransformers
+flash_optimum:
+# whether to use xformers attention patch https://github.com/facebookresearch/xformers:
+xformers_attention:
+# whether to use flash attention patch https://github.com/Dao-AILab/flash-attention:
+flash_attention:
+flash_attn_cross_entropy:  # Whether to use flash-attention cross entropy implementation - advanced use only
+flash_attn_rms_norm:  # Whether to use flash-attention rms norm implementation - advanced use only
+# whether to use scaled-dot-product attention
+# https://pytorch.org/docs/stable/generated/torch.nn.functional.scaled_dot_product_attention.html
+sdp_attention:
+# Landmark attention (only llama)
+landmark_attention:
+# xpos RoPE see https://github.com/kaiokendev/cutoff-len-is-context-len/blob/main/util/xpos_rope_llama_monkey_patch.py
+# llama only
+xpos_rope:
+# RoPE Scaling https://github.com/huggingface/transformers/pull/24653
+rope_scaling:
+  type: # linear | dynamic
+  factor: # float
+
+# resume from a specific checkpoint dir
+resume_from_checkpoint:
+# if resume_from_checkpoint isn't set and you simply want it to start where it left off
+# be careful with this being turned on between different models
+auto_resume_from_checkpoints: false
+
+# don't mess with this, it's here for accelerate and torchrun
+local_rank:
+
+# add or change special tokens
+special_tokens:
+  # bos_token: "<s>"
+  # eos_token: "</s>"
+  # unk_token: "<unk>"
+# add extra tokens
+tokens:
+
+# FSDP
+fsdp:
+fsdp_config:
+
+# Deepspeed config path
+deepspeed:
+
+# Advanced DDP Arguments
+ddp_timeout:
+ddp_bucket_cap_mb:
+ddp_broadcast_buffers:
+
+# Path to torch distx for optim 'adamw_anyprecision'
+torchdistx_path:
+
+# Set to HF dataset for type: 'completion' for streaming instead of pre-tokenize
+pretraining_dataset:
+
+# Debug mode
+debug:
+
+# Seed
+seed:
+
+# Allow overwrite yml config using from cli
+strict:
+```
+
+</details>
+
+### Train
+
+Run
+```bash
+accelerate launch -m axolotl.cli.train your_config.yml
+```
+
+#### Multi-GPU
+
+You can optionally pre-tokenize dataset with the following before finetuning:
+```bash
+CUDA_VISIBLE_DEVICES="" accelerate launch -m axolotl.cli.train your_config.yml --prepare_ds_only
+```
+
+##### Config
+
+- llama FSDP
+```yaml
+fsdp:
+  - full_shard
+  - auto_wrap
+fsdp_config:
+  fsdp_offload_params: true
+  fsdp_state_dict_type: FULL_STATE_DICT
+  fsdp_transformer_layer_cls_to_wrap: LlamaDecoderLayer
+```
+
+##### Weights & Biases Logging
+
+- wandb options
+```yaml
+wandb_mode:
+wandb_project:
+wandb_entity:
+wandb_watch:
+wandb_run_id:
+wandb_log_model:
+```
+
+### Training with Deepspeed
+
+Deepspeed is an optimization suite for multi-gpu systems allowing you to train much larger models than you
+might typically be able to fit into your GPU's VRAM. More information about the various optimization types
+for deepspeed is available at https://huggingface.co/docs/accelerate/main/en/usage_guides/deepspeed#what-is-integrated
+
+We provide several default deepspeed JSON configurations for ZeRO stage 1, 2, and 3.
+
+```shell
+accelerate launch -m axolotl.cli.train examples/llama-2/config.py --deepspeed deepspeed/zero1.json
+```
+
+or
+
+```yaml
+deepspeed: deepspeed/zero1.json
+```
+
+### Inference
+
+Pass the appropriate flag to the train command:
+
+- Pretrained LORA:
+  ```bash
+  python -m axolotl.cli.inference examples/your_config.yml --lora_model_dir="./lora-output-dir"
+  ```
+- Full weights finetune:
+  ```bash
+  python -m axolotl.cli.inference examples/your_config.yml --base_model="./completed-model"
+  ```
+- Full weights finetune w/ a prompt from a text file:
+  ```bash
+  cat /tmp/prompt.txt | python -m axolotl.cli.inference examples/your_config.yml \
+    --base_model="./completed-model" --prompter=None --load_in_8bit=True
+  ```
+
+### Merge LORA to base
+
+Add below flag to train command above
+
+```bash
+python3 -m axolotl.cli.merge_lora examples/your_config.yml --lora_model_dir="./completed-model" --load_in_8bit=False --load_in_4bit=False
+```
+
+If you run out of CUDA memory, you can try to merge in system RAM with
+
+```bash
+CUDA_VISIBLE_DEVICES="" python3 -m axolotl.cli.merge_lora ...
+```
+
+## Common Errors 🧰
+
+> If you encounter a 'Cuda out of memory' error, it means your GPU ran out of memory during the training process. Here's how to resolve it:
+
+Please reduce any below
+  - `micro_batch_size`
+  - `eval_batch_size`
+  - `gradient_accumulation_steps`
+  - `sequence_len`
+
+> `failed (exitcode: -9)`
+
+Usually means your system has run out of system memory.
+Similarly, you should consider reducing the same settings as when you run out of VRAM.
+Additionally, look into upgrading your system RAM which should be simpler than GPU upgrades.
+
+> RuntimeError: expected scalar type Float but found Half
+
+Try set `fp16: true`
+
+> NotImplementedError: No operator found for `memory_efficient_attention_forward` ...
+
+Try to turn off xformers.
+
+> accelerate config missing
+
+It's safe to ignore it.
+
+> NCCL Timeouts during training
+
+See the [NCCL](docs/nccl.md) guide.
+
+## Need help? 🙋♂️
+
+Join our [Discord server](https://discord.gg/HhrNrHJPRb) where we can help you
+
+## Badge ❤🏷️
+
+Building something cool with Axolotl? Consider adding a badge to your model card.
+
+```markdown
+[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
+```
+
+[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
+
+## Community Showcase
+
+Check out some of the projects and models that have been built using Axolotl! Have a model you'd like to add to our Community Showcase? Open a PR with your model.
+
+Open Access AI Collective
+- [Minotaur 13b](https://huggingface.co/openaccess-ai-collective/minotaur-13b)
+- [Manticore 13b](https://huggingface.co/openaccess-ai-collective/manticore-13b)
+- [Hippogriff 30b](https://huggingface.co/openaccess-ai-collective/hippogriff-30b-chat)
+
+PocketDoc Labs
+- [Dan's PersonalityEngine 13b LoRA](https://huggingface.co/PocketDoc/Dans-PersonalityEngine-13b-LoRA)
+
+## Contributing 🤝
+
+Please read the [contributing guide](./.github/CONTRIBUTING.md)
+
+Bugs? Please check the [open issues](https://github.com/OpenAccess-AI-Collective/axolotl/issues/bug) else create a new Issue.
+
+PRs are **greatly welcome**!
+
+Please run below to setup env
+```bash
+pip3 install -r requirements-dev.txt -r requirements-tests.txt
+pre-commit install
+
+# test
+pytest tests/
+```

_quarto.yml

@@ -1,57 +0,0 @@
-project:
-  type: website
-
-website:
-  title: "Axolotl"
-  description: "Fine-tuning"
-  favicon: favicon.jpg
-  navbar:
-    title: Axolotl
-    background: dark
-    pinned: false
-    collapse: false
-    tools:
-    - icon: twitter
-      href: https://twitter.com/axolotl_ai
-    - icon: github
-      href: https://github.com/axolotl-ai-cloud/axolotl/
-    - icon: discord
-      href: https://discord.gg/7m9sfhzaf3
-
-  sidebar:
-      pinned: true
-      collapse-level: 2
-      style: docked
-      contents:
-        - text: Home
-          href: index.qmd
-        - section: "How-To Guides"
-          contents:
-          # TODO Edit folder structure after we have more docs.
-            - docs/getting-started.qmd
-            - docs/installation.qmd
-            - docs/debugging.qmd
-            - docs/inference.qmd
-            - docs/multipack.qmd
-            - docs/fsdp_qlora.qmd
-            - docs/input_output.qmd
-            - docs/rlhf.qmd
-            - docs/nccl.qmd
-            - docs/mac.qmd
-            - docs/multi-gpu.qmd
-            - docs/multi-node.qmd
-            - docs/unsloth.qmd
-            - docs/amd_hpc.qmd
-            - docs/ray-integration.qmd
-        - section: "Dataset Formats"
-          contents: docs/dataset-formats/*
-        - section: "Reference"
-          contents:
-            - docs/config.qmd
-        - docs/faq.qmd
-
-format:
-  html:
-    theme: materia
-    css: styles.css
-    toc: true

cicd/Dockerfile.jinja

@@ -1,51 +0,0 @@
-FROM axolotlai/axolotl-base:{{ BASE_TAG }}
-
-ENV TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6+PTX"
-ENV AXOLOTL_EXTRAS="{{ AXOLOTL_EXTRAS }}"
-ENV AXOLOTL_ARGS="{{ AXOLOTL_ARGS }}"
-ENV CUDA="{{ CUDA }}"
-ENV PYTORCH_VERSION="{{ PYTORCH_VERSION }}"
-ENV GITHUB_REF="{{ GITHUB_REF }}"
-ENV GITHUB_SHA="{{ GITHUB_SHA }}"
-ENV NIGHTLY_BUILD="{{ NIGHTLY_BUILD }}"
-ENV HF_HOME="{{ HF_HOME }}"
-
-RUN apt-get update && \
-    apt-get install -y --allow-change-held-packages vim curl nano libnccl2 libnccl-dev
-
-WORKDIR /workspace
-
-RUN git clone --depth=1 https://github.com/axolotl-ai-cloud/axolotl.git
-
-WORKDIR /workspace/axolotl
-
-RUN git fetch origin +$GITHUB_REF && \
-    git checkout FETCH_HEAD
-
-# If AXOLOTL_EXTRAS is set, append it in brackets
-RUN if [ "$NIGHTLY_BUILD" = "true" ] ; then \
-        sed -i 's#^transformers.*#transformers @ git+https://github.com/huggingface/transformers.git@main#' requirements.txt; \
-        sed -i 's#^peft.*#peft @ git+https://github.com/huggingface/peft.git@main#' requirements.txt; \
-        sed -i 's#^accelerate.*#accelerate @ git+https://github.com/huggingface/accelerate.git@main#' requirements.txt; \
-        sed -i 's#^trl.*#trl @ git+https://github.com/huggingface/trl.git@main#' requirements.txt; \
-        sed -i 's#^datasets.*#datasets @ git+https://github.com/huggingface/datasets.git@main#' requirements.txt; \
-    fi
-
-RUN if [ "$AXOLOTL_EXTRAS" != "" ] ; then \
-        pip install --no-build-isolation -e .[deepspeed,flash-attn,optimizers,ray,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \
-    else \
-        pip install --no-build-isolation -e .[deepspeed,flash-attn,optimizers,ray] $AXOLOTL_ARGS; \
-    fi
-
-RUN python scripts/unsloth_install.py | sh
-RUN python scripts/cutcrossentropy_install.py | sh
-
-# So we can test the Docker image
-RUN pip install -r requirements-dev.txt -r requirements-tests.txt
-
-# fix so that git fetch/pull from remote works
-RUN git config remote.origin.fetch "+refs/heads/*:refs/remotes/origin/*" && \
-    git config --get remote.origin.fetch
-
-# helper for huggingface-login cli
-RUN git config --global credential.helper store

cicd/cicd.sh

@@ -1,11 +0,0 @@
-#!/bin/bash
-set -e
-
-python -c "import torch; assert '$PYTORCH_VERSION' in torch.__version__"
-
-pytest -v --durations=10 -n8 --ignore=tests/e2e/ --ignore=tests/patched/ /workspace/axolotl/tests/
-pytest -v --durations=10 /workspace/axolotl/tests/e2e/patched/lora_kernels  # running these with the other patches causes a failure
-pytest -v --durations=10 --ignore=tests/e2e/patched/lora_kernels /workspace/axolotl/tests/e2e/patched
-pytest -v --durations=10 -n1 /workspace/axolotl/tests/e2e/solo/
-pytest -v --durations=10 /workspace/axolotl/tests/e2e/integrations/
-pytest -v --durations=10 --ignore=tests/e2e/solo/ --ignore=tests/e2e/patched/ --ignore=tests/e2e/multigpu/ --ignore=tests/e2e/integrations/ /workspace/axolotl/tests/e2e/

cicd/multigpu.py

@@ -1,85 +0,0 @@
-"""
- modal application to run axolotl gpu tests in Modal
- """
-# pylint: disable=duplicate-code
-
-import os
-import pathlib
-import tempfile
-
-import jinja2
-import modal
-from jinja2 import select_autoescape
-from modal import App, Image
-
-cicd_path = pathlib.Path(__file__).parent.resolve()
-
-template_loader = jinja2.FileSystemLoader(searchpath=cicd_path)
-template_env = jinja2.Environment(
-    loader=template_loader, autoescape=select_autoescape()
-)
-df_template = template_env.get_template("Dockerfile.jinja")
-
-df_args = {
-    "AXOLOTL_EXTRAS": os.environ.get("AXOLOTL_EXTRAS", ""),
-    "AXOLOTL_ARGS": os.environ.get("AXOLOTL_ARGS", ""),
-    "PYTORCH_VERSION": os.environ.get("PYTORCH_VERSION", "2.4.1"),
-    "BASE_TAG": os.environ.get("BASE_TAG", "main-base-py3.11-cu121-2.4.1"),
-    "CUDA": os.environ.get("CUDA", "121"),
-    "GITHUB_REF": os.environ.get("GITHUB_REF", "refs/heads/main"),
-    "GITHUB_SHA": os.environ.get("GITHUB_SHA", ""),
-    "HF_HOME": "/workspace/data/huggingface-cache/hub",
-}
-
-dockerfile_contents = df_template.render(**df_args)
-
-temp_dir = tempfile.mkdtemp()
-with open(pathlib.Path(temp_dir) / "Dockerfile", "w", encoding="utf-8") as f:
-    f.write(dockerfile_contents)
-
-cicd_image = (
-    Image.from_dockerfile(
-        pathlib.Path(temp_dir) / "Dockerfile",
-        force_build=True,
-        gpu="A10G",
-    )
-    .env(df_args)
-    .pip_install("fastapi==0.110.0", "pydantic==2.6.3")
-)
-
-app = App("Axolotl CI/CD", secrets=[])
-
-hf_cache_volume = modal.Volume.from_name(
-    "axolotl-ci-hf-hub-cache", create_if_missing=True
-)
-VOLUME_CONFIG = {
-    "/workspace/data/huggingface-cache/hub": hf_cache_volume,
-}
-
-N_GPUS = int(os.environ.get("N_GPUS", 2))
-GPU_CONFIG = modal.gpu.H100(count=N_GPUS)
-
-
-def run_cmd(cmd: str, run_folder: str):
-    import subprocess  # nosec
-
-    # Propagate errors from subprocess.
-    if exit_code := subprocess.call(cmd.split(), cwd=run_folder):  # nosec
-        exit(exit_code)  # pylint: disable=consider-using-sys-exit
-
-
-@app.function(
-    image=cicd_image,
-    gpu=GPU_CONFIG,
-    timeout=60 * 60,
-    cpu=8.0,
-    memory=131072 * N_GPUS,
-    volumes=VOLUME_CONFIG,
-)
-def cicd_pytest():
-    run_cmd("./cicd/multigpu.sh", "/workspace/axolotl")
-
-
-@app.local_entrypoint()
-def main():
-    cicd_pytest.remote()

cicd/multigpu.sh

@@ -1,5 +0,0 @@
-#!/bin/bash
-set -e
-
-# only run one test at a time so as not to OOM the GPU
-pytest -v -n2 /workspace/axolotl/tests/e2e/multigpu/

cicd/tests.py

@@ -1,81 +0,0 @@
-"""Modal app to run axolotl GPU tests"""
-# pylint: disable=duplicate-code
-
-import os
-import pathlib
-import tempfile
-
-import jinja2
-import modal
-from jinja2 import select_autoescape
-from modal import App, Image
-
-cicd_path = pathlib.Path(__file__).parent.resolve()
-
-template_loader = jinja2.FileSystemLoader(searchpath=cicd_path)
-template_env = jinja2.Environment(
-    loader=template_loader, autoescape=select_autoescape()
-)
-df_template = template_env.get_template("Dockerfile.jinja")
-
-df_args = {
-    "AXOLOTL_EXTRAS": os.environ.get("AXOLOTL_EXTRAS", ""),
-    "AXOLOTL_ARGS": os.environ.get("AXOLOTL_ARGS", ""),
-    "PYTORCH_VERSION": os.environ.get("PYTORCH_VERSION", "2.4.1"),
-    "BASE_TAG": os.environ.get("BASE_TAG", "main-base-py3.11-cu121-2.4.1"),
-    "CUDA": os.environ.get("CUDA", "121"),
-    "GITHUB_REF": os.environ.get("GITHUB_REF", "refs/heads/main"),
-    "GITHUB_SHA": os.environ.get("GITHUB_SHA", ""),
-    "NIGHTLY_BUILD": os.environ.get("NIGHTLY_BUILD", ""),
-    "HF_HOME": "/workspace/data/huggingface-cache/hub",
-}
-
-dockerfile_contents = df_template.render(**df_args)
-
-temp_dir = tempfile.mkdtemp()
-with open(pathlib.Path(temp_dir) / "Dockerfile", "w", encoding="utf-8") as f:
-    f.write(dockerfile_contents)
-
-cicd_image = Image.from_dockerfile(
-    pathlib.Path(temp_dir) / "Dockerfile",
-    context_mount=None,
-    force_build=True,
-    gpu="A10G",
-).env(df_args)
-
-app = App("Axolotl CI/CD", secrets=[])
-
-hf_cache_volume = modal.Volume.from_name(
-    "axolotl-ci-hf-hub-cache", create_if_missing=True
-)
-VOLUME_CONFIG = {
-    "/workspace/data/huggingface-cache/hub": hf_cache_volume,
-}
-
-N_GPUS = int(os.environ.get("N_GPUS", 1))
-GPU_CONFIG = modal.gpu.L40S(count=N_GPUS)
-
-
-def run_cmd(cmd: str, run_folder: str):
-    import subprocess  # nosec
-
-    # Propagate errors from subprocess.
-    if exit_code := subprocess.call(cmd.split(), cwd=run_folder):  # nosec
-        exit(exit_code)  # pylint: disable=consider-using-sys-exit
-
-
-@app.function(
-    image=cicd_image,
-    gpu=GPU_CONFIG,
-    timeout=60 * 60,
-    cpu=8.0,
-    memory=131072,
-    volumes=VOLUME_CONFIG,
-)
-def cicd_pytest():
-    run_cmd("./cicd/cicd.sh", "/workspace/axolotl")
-
-
-@app.local_entrypoint()
-def main():
-    cicd_pytest.remote()

deepspeed/zero1.json

@@ -15,12 +15,26 @@
     "hysteresis": 2,
     "min_loss_scale": 1
   },
-  "compile": {
-    "disable": false,
-    "backend": "inductor"
+  "optimizer": {
+    "type": "AdamW",
+    "params": {
+      "lr": "auto",
+      "betas": "auto",
+      "eps": "auto",
+      "weight_decay": "auto"
+    }
+  },
+  "scheduler": {
+    "type": "WarmupDecayLR",
+    "params": {
+      "warmup_min_lr": "auto",
+      "warmup_max_lr": "auto",
+      "warmup_num_steps": "auto",
+      "warmup_type": "linear",
+      "total_num_steps": "auto"
+    }
   },
   "gradient_accumulation_steps": "auto",
-  "gradient_clipping": "auto",
   "train_batch_size": "auto",
   "train_micro_batch_size_per_gpu": "auto",
   "wall_clock_breakdown": false

deepspeed/zero2.json

@@ -19,8 +19,26 @@
     "hysteresis": 2,
     "min_loss_scale": 1
   },
+  "optimizer": {
+    "type": "AdamW",
+    "params": {
+      "lr": "auto",
+      "betas": "auto",
+      "eps": "auto",
+      "weight_decay": "auto"
+    }
+  },
+  "scheduler": {
+    "type": "WarmupDecayLR",
+    "params": {
+      "warmup_min_lr": "auto",
+      "warmup_max_lr": "auto",
+      "warmup_num_steps": "auto",
+      "warmup_type": "linear",
+      "total_num_steps": "auto"
+    }
+  },
   "gradient_accumulation_steps": "auto",
-  "gradient_clipping": "auto",
   "train_batch_size": "auto",
   "train_micro_batch_size_per_gpu": "auto",
   "wall_clock_breakdown": false

deepspeed/zero3.json

@@ -1,6 +1,14 @@
 {
   "zero_optimization": {
     "stage": 3,
+    "offload_optimizer": {
+      "device": "cpu",
+      "pin_memory": true
+    },
+    "offload_param": {
+      "device": "cpu",
+      "pin_memory": true
+    },
     "overlap_comm": true,
     "contiguous_gradients": true,
     "sub_group_size": 0,
@@ -23,8 +31,25 @@
     "hysteresis": 2,
     "min_loss_scale": 1
   },
+  "optimizer": {
+    "type": "AdamW",
+    "params": {
+      "lr": "auto",
+      "betas": "auto",
+      "eps": "auto",
+      "weight_decay": "auto"
+    }
+  },
+  "scheduler": {
+    "type": "WarmupLR",
+    "params": {
+      "warmup_min_lr": "auto",
+      "warmup_max_lr": "auto",
+      "warmup_num_steps": "auto",
+      "warmup_type": "linear"
+    }
+  },
   "gradient_accumulation_steps": "auto",
-  "gradient_clipping": "auto",
   "train_batch_size": "auto",
   "train_micro_batch_size_per_gpu": "auto",
   "wall_clock_breakdown": false

deepspeed_configs/zero1.json

@@ -1,23 +0,0 @@
-{
-  "zero_optimization": {
-    "stage": 1,
-    "overlap_comm": true
-  },
-  "bf16": {
-    "enabled": "auto"
-  },
-  "fp16": {
-    "enabled": "auto",
-    "auto_cast": false,
-    "loss_scale": 0,
-    "initial_scale_power": 32,
-    "loss_scale_window": 1000,
-    "hysteresis": 2,
-    "min_loss_scale": 1
-  },
-  "gradient_accumulation_steps": "auto",
-  "gradient_clipping": "auto",
-  "train_batch_size": "auto",
-  "train_micro_batch_size_per_gpu": "auto",
-  "wall_clock_breakdown": false
-}

deepspeed_configs/zero3_bf16.json

@@ -1,22 +0,0 @@
-{
-  "zero_optimization": {
-    "stage": 3,
-    "overlap_comm": true,
-    "contiguous_gradients": true,
-    "sub_group_size": 0,
-    "reduce_bucket_size": "auto",
-    "stage3_prefetch_bucket_size": "auto",
-    "stage3_param_persistence_threshold": "auto",
-    "stage3_max_live_parameters": 0,
-    "stage3_max_reuse_distance": 0,
-    "stage3_gather_16bit_weights_on_model_save": true
-  },
-  "bf16": {
-    "enabled": true
-  },
-  "gradient_accumulation_steps": "auto",
-  "gradient_clipping": "auto",
-  "train_batch_size": "auto",
-  "train_micro_batch_size_per_gpu": "auto",
-  "wall_clock_breakdown": false
-}

deepspeed_configs/zero3_bf16_cpuoffload_all.json

@@ -1,32 +0,0 @@
-{
-  "zero_force_ds_cpu_optimizer": false,
-  "zero_allow_untested_optimizer": true,
-  "zero_optimization": {
-    "stage": 3,
-    "offload_optimizer": {
-      "device": "cpu",
-      "pin_memory": true
-    },
-    "offload_param": {
-      "device": "cpu",
-      "pin_memory": true
-    },
-    "overlap_comm": true,
-    "contiguous_gradients": true,
-    "sub_group_size": 0,
-    "reduce_bucket_size": "auto",
-    "stage3_prefetch_bucket_size": "auto",
-    "stage3_param_persistence_threshold": "auto",
-    "stage3_max_live_parameters": 0,
-    "stage3_max_reuse_distance": 0,
-    "stage3_gather_16bit_weights_on_model_save": true
-  },
-  "bf16": {
-    "enabled": true
-  },
-  "gradient_accumulation_steps": "auto",
-  "gradient_clipping": "auto",
-  "train_batch_size": "auto",
-  "train_micro_batch_size_per_gpu": "auto",
-  "wall_clock_breakdown": false
-}

deepspeed_configs/zero3_bf16_cpuoffload_params.json

@@ -1,28 +0,0 @@
-{
-  "zero_force_ds_cpu_optimizer": false,
-  "zero_allow_untested_optimizer": true,
-  "zero_optimization": {
-    "stage": 3,
-    "offload_param": {
-      "device": "cpu",
-      "pin_memory": true
-    },
-    "overlap_comm": true,
-    "contiguous_gradients": true,
-    "sub_group_size": 0,
-    "reduce_bucket_size": "auto",
-    "stage3_prefetch_bucket_size": "auto",
-    "stage3_param_persistence_threshold": "auto",
-    "stage3_max_live_parameters": 0,
-    "stage3_max_reuse_distance": 0,
-    "stage3_gather_16bit_weights_on_model_save": true
-  },
-  "bf16": {
-    "enabled": true
-  },
-  "gradient_accumulation_steps": "auto",
-  "gradient_clipping": "auto",
-  "train_batch_size": "auto",
-  "train_micro_batch_size_per_gpu": "auto",
-  "wall_clock_breakdown": false
-}

devtools/README.md

@@ -1 +0,0 @@
-This directory contains example config files that might be useful for debugging. Please see [docs/debugging.qmd](../docs/debugging.qmd) for more information.

devtools/dev_chat_template.yml

@@ -1,48 +0,0 @@
-# Example config for debugging the chat_template prompt format
-base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
-model_type: LlamaForCausalLM
-tokenizer_type: LlamaTokenizer
-
-load_in_8bit: true
-load_in_4bit: false
-
-datasets:
-  - path: fozziethebeat/alpaca_messages_2k_test
-    type: chat_template
-    shards: 10
-val_set_size: 0
-output_dir: temp_debug/axolotl_outputs/model
-dataset_prepared_path: temp_debug/axolotl_outputs/data
-dataset_processes: 1
-
-sequence_len: 4096
-sample_packing: false
-pad_to_sequence_len: true
-
-adapter: lora
-lora_model_dir:
-lora_r: 32
-lora_alpha: 16
-lora_dropout: 0.05
-lora_target_linear: true
-lora_fan_in_fan_out:
-
-micro_batch_size: 1
-num_epochs: 1
-max_steps: 10
-optimizer: adamw_bnb_8bit
-lr_scheduler: cosine
-learning_rate: 0.0002
-
-train_on_inputs: false
-group_by_length: false
-bf16: false
-fp16: true
-tf32: false
-
-gradient_checkpointing: true
-logging_steps: 1
-flash_attention: true
-
-warmup_steps: 10
-weight_decay: 0.0

docker/Dockerfile

@@ -1,36 +1,27 @@
 ARG BASE_TAG=main-base
-FROM axolotlai/axolotl-base:$BASE_TAG
+FROM winglian/axolotl-base:$BASE_TAG
 
 ARG TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6+PTX"
 ARG AXOLOTL_EXTRAS=""
-ARG AXOLOTL_ARGS=""
 ARG CUDA="118"
-ARG PYTORCH_VERSION="2.1.2"
-
-ENV PYTORCH_VERSION=$PYTORCH_VERSION
+ENV BNB_CUDA_VERSION=$CUDA
 
 RUN apt-get update && \
-    apt-get install -y --allow-change-held-packages vim curl nano libnccl2 libnccl-dev rsync s3fs
+    apt-get install -y vim curl
 
 WORKDIR /workspace
 
-RUN git clone --depth=1 https://github.com/axolotl-ai-cloud/axolotl.git
+RUN git clone --depth=1 https://github.com/OpenAccess-AI-Collective/axolotl.git
 
 WORKDIR /workspace/axolotl
 
 # If AXOLOTL_EXTRAS is set, append it in brackets
 RUN if [ "$AXOLOTL_EXTRAS" != "" ] ; then \
-        pip install --no-build-isolation -e .[deepspeed,flash-attn,optimizers,ray,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \
+        pip install -e .[flash-attn,$AXOLOTL_EXTRAS]; \
     else \
-        pip install --no-build-isolation -e .[deepspeed,flash-attn,optimizers,ray] $AXOLOTL_ARGS; \
+        pip install -e .[flash-attn]; \
     fi
 
-RUN python scripts/unsloth_install.py | sh
-RUN python scripts/cutcrossentropy_install.py | sh
-
-# So we can test the Docker image
-RUN pip install pytest
-
 # fix so that git fetch/pull from remote works
 RUN git config remote.origin.fetch "+refs/heads/*:refs/remotes/origin/*" && \
     git config --get remote.origin.fetch

docker/Dockerfile-base

@@ -3,20 +3,18 @@ ARG CUDNN_VERSION="8"
 ARG UBUNTU_VERSION="22.04"
 ARG MAX_JOBS=4
 
-FROM nvidia/cuda:$CUDA_VERSION-cudnn$CUDNN_VERSION-devel-ubuntu$UBUNTU_VERSION AS base-builder
+FROM nvidia/cuda:$CUDA_VERSION-cudnn$CUDNN_VERSION-devel-ubuntu$UBUNTU_VERSION as base-builder
 
 ENV PATH="/root/miniconda3/bin:${PATH}"
 
-ARG PYTHON_VERSION="3.10"
-ARG PYTORCH_VERSION="2.1.2"
+ARG PYTHON_VERSION="3.9"
+ARG PYTORCH_VERSION="2.0.1"
 ARG CUDA="118"
-ARG TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6 9.0+PTX"
 
 ENV PYTHON_VERSION=$PYTHON_VERSION
-ENV TORCH_CUDA_ARCH_LIST=$TORCH_CUDA_ARCH_LIST
 
 RUN apt-get update \
-    && apt-get install -y wget git build-essential ninja-build git-lfs libaio-dev pkg-config && rm -rf /var/lib/apt/lists/* \
+    && apt-get install -y wget git build-essential ninja-build git-lfs libaio-dev && rm -rf /var/lib/apt/lists/*
     && wget \
     https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh \
     && mkdir /root/.conda \
@@ -29,11 +27,52 @@ ENV PATH="/root/miniconda3/envs/py${PYTHON_VERSION}/bin:${PATH}"
 WORKDIR /workspace
 
 RUN python3 -m pip install --upgrade pip && pip3 install packaging && \
-    python3 -m pip install --no-cache-dir -U torch==${PYTORCH_VERSION}+cu${CUDA} --extra-index-url https://download.pytorch.org/whl/cu$CUDA && \
-    python3 -m pip install --no-cache-dir "causal_conv1d @ git+https://github.com/Dao-AILab/causal-conv1d.git@main" && \
-    python3 -m pip install --no-cache-dir "mamba_ssm @ git+https://github.com/state-spaces/mamba.git@main"
+    python3 -m pip install --no-cache-dir -U torch==${PYTORCH_VERSION}+cu${CUDA} --extra-index-url https://download.pytorch.org/whl/cu$CUDA
 
-RUN git lfs install --skip-repo && \
-    pip3 install awscli && \
+FROM base-builder AS deepspeed-builder
+
+ARG TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6 9.0+PTX"
+
+WORKDIR /workspace
+
+RUN git clone https://github.com/microsoft/DeepSpeed.git && \
+    cd DeepSpeed && \
+    MAX_CONCURRENCY=8 DS_BUILD_SPARSE_ATTN=0 DS_BUILD_OPS=1 DS_BUILD_EVOFORMER_ATTN=0 python3 setup.py bdist_wheel
+
+FROM base-builder AS bnb-builder
+
+WORKDIR /workspace
+ARG CUDA="118"
+ENV CUDA=$CUDA
+ARG MAX_JOBS="-1"
+ENV MAX_JOBS=$MAX_JOBS
+
+RUN git clone https://github.com/TimDettmers/bitsandbytes.git && \
+    cd bitsandbytes && \
+    CUDA_VERSION=$CUDA make cuda11x && \
+    python setup.py bdist_wheel
+
+FROM base-builder
+
+ARG TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6 9.0+PTX"
+ENV TORCH_CUDA_ARCH_LIST=$TORCH_CUDA_ARCH_LIST
+
+# recompile apex
+RUN python3 -m pip uninstall -y apex
+RUN git clone https://github.com/NVIDIA/apex
+RUN cd apex && python3 -m pip install -v --disable-pip-version-check --no-cache-dir --no-build-isolation --config-settings "--build-option=--cpp_ext" --config-settings "--build-option=--cuda_ext" ./
+
+RUN mkdir -p /workspace/builds
+COPY --from=bnb-builder /workspace/bitsandbytes /workspace/builds/bitsandbytes
+
+RUN mkdir -p /workspace/wheels/bitsandbytes
+COPY --from=deepspeed-builder /workspace/DeepSpeed/dist/deepspeed-*.whl wheels
+COPY --from=bnb-builder /workspace/bitsandbytes/dist/bitsandbytes-*.whl wheels
+COPY --from=bnb-builder /workspace/bitsandbytes/bitsandbytes/libbitsandbytes*.so wheels/bitsandbytes
+
+RUN pip3 install wheels/deepspeed-*.whl
+RUN cd /workspace/builds/bitsandbytes && python3 setup.py install
+RUN git lfs install --skip-repo
+RUN pip3 install awscli && \
     # The base image ships with `pydantic==1.8.2` which is not working
     pip3 install -U --no-cache-dir pydantic==1.10.10

docker/Dockerfile-cloud

@@ -1,27 +0,0 @@
-ARG BASE_TAG=main
-FROM axolotlai/axolotl:$BASE_TAG
-
-ENV HF_DATASETS_CACHE="/workspace/data/huggingface-cache/datasets"
-ENV HF_HUB_CACHE="/workspace/data/huggingface-cache/hub"
-ENV HF_HOME="/workspace/data/huggingface-cache/hub"
-ENV HF_HUB_ENABLE_HF_TRANSFER="1"
-
-EXPOSE 8888
-EXPOSE 22
-
-COPY scripts/cloud-entrypoint.sh /root/cloud-entrypoint.sh
-COPY scripts/motd /etc/motd
-
-RUN pip install jupyterlab notebook ipywidgets && \
-    jupyter lab clean
-RUN apt install --yes --no-install-recommends openssh-server tmux && \
-    mkdir -p ~/.ssh && \
-    chmod 700 ~/.ssh && \
-    printf "\n[[ -z \"\$TMUX\"  ]] && { tmux attach-session -t ssh_tmux || tmux new-session -s ssh_tmux; exit; }\n" >> ~/.bashrc && \
-    printf "[ ! -z \"\$TERM\" -a -r /etc/motd ] && cat /etc/motd\n" >> ~/.bashrc && \
-    chmod +x /workspace/axolotl/scripts/cloud-entrypoint.sh && \
-    chmod +x /root/cloud-entrypoint.sh && \
-    echo 'set-option -g history-limit 5000' >> ~/.tmux.conf
-
-ENTRYPOINT ["/root/cloud-entrypoint.sh"]
-CMD ["sleep", "infinity"]

docker/Dockerfile-cloud-no-tmux

@@ -1,26 +0,0 @@
-ARG BASE_TAG=main
-FROM axolotlai/axolotl:$BASE_TAG
-
-ENV HF_DATASETS_CACHE="/workspace/data/huggingface-cache/datasets"
-ENV HF_HUB_CACHE="/workspace/data/huggingface-cache/hub"
-ENV HF_HOME="/workspace/data/huggingface-cache/hub"
-ENV HF_HUB_ENABLE_HF_TRANSFER="1"
-
-EXPOSE 8888
-EXPOSE 22
-
-COPY scripts/cloud-entrypoint-term.sh /root/cloud-entrypoint.sh
-COPY scripts/motd /etc/motd
-
-RUN pip install jupyterlab notebook ipywidgets && \
-    jupyter lab clean
-RUN apt install --yes --no-install-recommends openssh-server tmux sudo && \
-    pip3 install -U --no-cache-dir grpcio ray[default]==2.9.3 && \
-    mkdir -p ~/.ssh && \
-    chmod 700 ~/.ssh && \
-    printf "[ ! -z \"\$TERM\" -a -r /etc/motd ] && cat /etc/motd\n" >> ~/.bashrc && \
-    chmod +x /workspace/axolotl/scripts/cloud-entrypoint.sh && \
-    chmod +x /root/cloud-entrypoint.sh
-
-ENTRYPOINT ["/root/cloud-entrypoint.sh"]
-CMD ["sleep", "infinity"]

docker/Dockerfile-runpod

@@ -0,0 +1,18 @@
+ARG BASE_TAG=main
+FROM winglian/axolotl:$BASE_TAG
+
+ENV HF_DATASETS_CACHE="/workspace/data/huggingface-cache/datasets"
+ENV HUGGINGFACE_HUB_CACHE="/workspace/data/huggingface-cache/hub"
+ENV TRANSFORMERS_CACHE="/workspace/data/huggingface-cache/hub"
+
+COPY scripts/runpod-entrypoint.sh /root/runpod-entrypoint.sh
+
+RUN apt install --yes --no-install-recommends openssh-server tmux && \
+    mkdir -p ~/.ssh && \
+    chmod 700 ~/.ssh && \
+    printf "\n[[ -z \"\$TMUX\"  ]] && { tmux attach-session -t ssh_tmux || tmux new-session -s ssh_tmux; exit; }\n" >> ~/.bashrc && \
+    chmod +x /workspace/axolotl/scripts/runpod-entrypoint.sh && \
+    chmod +x /root/runpod-entrypoint.sh
+
+ENTRYPOINT ["/root/runpod-entrypoint.sh"]
+CMD ["sleep", "infinity"]

docker/Dockerfile-tests

@@ -1,40 +0,0 @@
-ARG BASE_TAG=main-base
-FROM axolotlai/axolotl-base:$BASE_TAG
-
-ARG TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6+PTX"
-ARG AXOLOTL_EXTRAS=""
-ARG AXOLOTL_ARGS=""
-ARG CUDA="118"
-ARG PYTORCH_VERSION="2.1.2"
-ARG GITHUB_REF="main"
-
-ENV PYTORCH_VERSION=$PYTORCH_VERSION
-
-RUN apt-get update && \
-    apt-get install -y --allow-change-held-packages vim curl nano libnccl2 libnccl-dev
-
-WORKDIR /workspace
-
-RUN git clone --depth=1 https://github.com/axolotl-ai-cloud/axolotl.git
-
-WORKDIR /workspace/axolotl
-
-RUN git fetch origin +$GITHUB_REF && \
-    git checkout FETCH_HEAD
-
-# If AXOLOTL_EXTRAS is set, append it in brackets
-RUN if [ "$AXOLOTL_EXTRAS" != "" ] ; then \
-        pip install --no-build-isolation -e .[deepspeed,flash-attn,mamba-ssm,$AXOLOTL_EXTRAS] $AXOLOTL_ARGS; \
-    else \
-        pip install --no-build-isolation -e .[deepspeed,flash-attn,mamba-ssm] $AXOLOTL_ARGS; \
-    fi
-
-# So we can test the Docker image
-RUN pip install pytest
-
-# fix so that git fetch/pull from remote works
-RUN git config remote.origin.fetch "+refs/heads/*:refs/remotes/origin/*" && \
-    git config --get remote.origin.fetch
-
-# helper for huggingface-login cli
-RUN git config --global credential.helper store

docs/.gitignore

@@ -1,2 +0,0 @@
-/.quarto/
-_site/

docs/amd_hpc.qmd

@@ -1,108 +0,0 @@
----
-title: Training with AMD GPUs on HPC Systems
-description: A comprehensive guide for using Axolotl on distributed systems with AMD GPUs
----
-
-This guide provides step-by-step instructions for installing and configuring Axolotl on a High-Performance Computing (HPC) environment equipped with AMD GPUs.
-
-## Setup
-
-### 1. Install Python
-
-We recommend using Miniforge, a minimal conda-based Python distribution:
-
-```bash
-curl -L -O "https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-$(uname)-$(uname -m).sh"
-bash Miniforge3-$(uname)-$(uname -m).sh
-```
-
-### 2. Configure Python Environment
-Add Python to your PATH and ensure it's available at login:
-
-```bash
-echo 'export PATH=~/miniforge3/bin:$PATH' >> ~/.bashrc
-echo 'if [ -f ~/.bashrc ]; then . ~/.bashrc; fi' >> ~/.bash_profile
-```
-
-### 3. Load AMD GPU Software
-
-Load the ROCm module:
-
-```bash
-module load rocm/5.7.1
-```
-
-Note: The specific module name and version may vary depending on your HPC system. Consult your system documentation for the correct module name.
-
-### 4. Install PyTorch
-
-Install PyTorch with ROCm support:
-
-```bash
-pip install -U torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm5.7 --force-reinstall
-```
-
-### 5. Install Flash Attention
-
-Clone and install the Flash Attention repository:
-
-```bash
-git clone --recursive https://github.com/ROCmSoftwarePlatform/flash-attention.git
-export GPU_ARCHS="gfx90a"
-cd flash-attention
-export PYTHON_SITE_PACKAGES=$(python -c 'import site; print(site.getsitepackages()[0])')
-patch "${PYTHON_SITE_PACKAGES}/torch/utils/hipify/hipify_python.py" hipify_patch.patch
-pip install --no-build-isolation .
-```
-
-### 6. Install Axolotl
-
-Clone and install Axolotl:
-
-```bash
-git clone https://github.com/axolotl-ai-cloud/axolotl
-cd axolotl
-pip install packaging ninja
-pip install --no-build-isolation -e .
-```
-
-### 7. Apply xformers Workaround
-
-xformers appears to be incompatible with ROCm. Apply the following workarounds:
- - Edit $HOME/packages/axolotl/src/axolotl/monkeypatch/llama_attn_hijack_flash.py modifying the code to always return `False` for SwiGLU availability from xformers.
- - Edit $HOME/miniforge3/lib/python3.10/site-packages/xformers/ops/swiglu_op.py replacing the "SwiGLU" function with a pass statement.
-
-### 8. Prepare Job Submission Script
-
-Create a script for job submission using your HPC's particular software (e.g. Slurm, PBS). Include necessary environment setup and the command to run Axolotl training. If the compute node(s) do(es) not have internet access, it is recommended to include
-
-```bash
-export TRANSFORMERS_OFFLINE=1
-export HF_DATASETS_OFFLINE=1
-```
-
-### 9. Download Base Model
-
-Download a base model using the Hugging Face CLI:
-
-```bash
-huggingface-cli download meta-llama/Meta-Llama-3.1-8B --local-dir ~/hfdata/llama3.1-8B
-```
-
-### 10. Create Axolotl Configuration
-
-Create an Axolotl configuration file (YAML format) tailored to your specific training requirements and dataset. Use FSDP for multi-node training.
-
-Note: Deepspeed did not work at the time of testing. However, if anyone managed to get it working, please let us know.
-
-### 11. Preprocess Data
-
-Run preprocessing on the login node:
-
-```bash
-CUDA_VISIBLE_DEVICES="" python -m axolotl.cli.preprocess /path/to/your/config.yaml
-```
-
-### 12. Train
-
-You are now ready to submit your previously prepared job script. 🚂

docs/batch_vs_grad.qmd

@@ -1,59 +0,0 @@
----
-title: Batch size vs Gradient accumulation
-description: Understanding of batch size and gradient accumulation steps
----
-
-Gradient accumulation means accumulating gradients over several mini-batches and updating the model weights afterward. When the samples in each batch are diverse, this technique doesn't significantly impact learning.
-
-This method allows for effective training with larger effective batch sizes without needing proportionally larger memory. Here's why:
-
-1. **Memory Consumption with Batch Size**: The primary reason increasing the batch size impacts memory is due to the storage requirements for intermediate activations. When you forward propagate a batch through a network, you have to store the activations at each layer for each sample in the batch, because these activations are used during backpropagation to compute gradients. Therefore, larger batches mean more activations, leading to greater GPU memory consumption.
-
-2. **Gradient Accumulation**: With gradient accumulation, you're effectively simulating a larger batch size by accumulating gradients over several smaller batches (or micro-batches). However, at any given time, you're only forward and backward propagating a micro-batch. This means you only store activations for the micro-batch, not the full accumulated batch. As a result, you can simulate the effect of a larger batch size without the memory cost of storing activations for a large batch.
-
-**Example 1:**
-Micro batch size: 3
-Gradient accumulation steps: 2
-Number of GPUs: 3
-Total batch size = 3 * 2 * 3 = 18
-
-```
-| GPU 1          | GPU 2          | GPU 3          |
-|----------------|----------------|----------------|
-| S1, S2, S3     | S4, S5, S6     | S7, S8, S9     |
-| e1, e2, e3     | e4, e5, e6     | e7, e8, e9     |
-|----------------|----------------|----------------|
-| → (accumulate) | → (accumulate) | → (accumulate) |
-|----------------|----------------|----------------|
-| S10, S11, S12  | S13, S14, S15  | S16, S17, S18  |
-| e10, e11, e12  | e13, e14, e15  | e16, e17, e18  |
-|----------------|----------------|----------------|
-| → (apply)      | → (apply)      | → (apply)      |
-
-Accumulated gradient for the weight w1 after the second iteration (considering all GPUs):
-Total gradient for w1 = e1 + e2 + e3 + e4 + e5 + e6 + e7 + e8 + e9 + e10 + e11 + e12 + e13 + e14 + e15 + e16 + e17 + e18
-
-Weight update for w1:
-w1_new = w1_old - learning rate x (Total gradient for w1 / 18)
-```
-
-**Example 2:**
-Micro batch size: 2
-Gradient accumulation steps: 1
-Number of GPUs: 3
-Total batch size = 2 * 1 * 3 = 6
-
-```
-| GPU 1     | GPU 2     | GPU 3     |
-|-----------|-----------|-----------|
-| S1, S2    | S3, S4    | S5, S6    |
-| e1, e2    | e3, e4    | e5, e6    |
-|-----------|-----------|-----------|
-| → (apply) | → (apply) | → (apply) |
-
-Accumulated gradient for the weight w1 (considering all GPUs):
-Total gradient for w1 = e1 + e2 + e3 + e4 + e5 + e6
-
-Weight update for w1:
-w1_new = w1_old - learning rate × (Total gradient for w1 / 6)
-```

docs/cli.qmd

@@ -1,256 +0,0 @@
-# Axolotl CLI Documentation
-
-The Axolotl CLI provides a streamlined interface for training and fine-tuning large language models. This guide covers
-the CLI commands, their usage, and common examples.
-
-### Table of Contents
-
-- Basic Commands
-- Command Reference
-  - fetch
-  - preprocess
-  - train
-  - inference
-  - merge-lora
-  - merge-sharded-fsdp-weights
-  - evaluate
-  - lm-eval
-- Legacy CLI Usage
-- Remote Compute with Modal Cloud
-  - Cloud Configuration
-  - Running on Modal Cloud
-  - Cloud Configuration Options
-
-
-### Basic Commands
-
-All Axolotl commands follow this general structure:
-
-```bash
-axolotl <command> [config.yml] [options]
-```
-
-The config file can be local or a URL to a raw YAML file.
-
-### Command Reference
-
-#### fetch
-
-Downloads example configurations and deepspeed configs to your local machine.
-
-```bash
-# Get example YAML files
-axolotl fetch examples
-
-# Get deepspeed config files
-axolotl fetch deepspeed_configs
-
-# Specify custom destination
-axolotl fetch examples --dest path/to/folder
-```
-
-#### preprocess
-
-Preprocesses and tokenizes your dataset before training. This is recommended for large datasets.
-
-```bash
-# Basic preprocessing
-axolotl preprocess config.yml
-
-# Preprocessing with one GPU
-CUDA_VISIBLE_DEVICES="0" axolotl preprocess config.yml
-
-# Debug mode to see processed examples
-axolotl preprocess config.yml --debug
-
-# Debug with limited examples
-axolotl preprocess config.yml --debug --debug-num-examples 5
-```
-
-Configuration options:
-
-```yaml
-dataset_prepared_path: Local folder for saving preprocessed data
-push_dataset_to_hub: HuggingFace repo to push preprocessed data (optional)
-```
-
-#### train
-
-Trains or fine-tunes a model using the configuration specified in your YAML file.
-
-```bash
-# Basic training
-axolotl train config.yml
-
-# Train and set/override specific options
-axolotl train config.yml \
-    --learning-rate 1e-4 \
-    --micro-batch-size 2 \
-    --num-epochs 3
-
-# Training without accelerate
-axolotl train config.yml --no-accelerate
-
-# Resume training from checkpoint
-axolotl train config.yml --resume-from-checkpoint path/to/checkpoint
-```
-
-#### inference
-
-Runs inference using your trained model in either CLI or Gradio interface mode.
-
-```bash
-# CLI inference with LoRA
-axolotl inference config.yml --lora-model-dir="./outputs/lora-out"
-
-# CLI inference with full model
-axolotl inference config.yml --base-model="./completed-model"
-
-# Gradio web interface
-axolotl inference config.yml --gradio \
-    --lora-model-dir="./outputs/lora-out"
-
-# Inference with input from file
-cat prompt.txt | axolotl inference config.yml \
-    --base-model="./completed-model"
-```
-
-#### merge-lora
-
-Merges trained LoRA adapters into the base model.
-
-```bash
-# Basic merge
-axolotl merge-lora config.yml
-
-# Specify LoRA directory (usually used with checkpoints)
-axolotl merge-lora config.yml --lora-model-dir="./lora-output/checkpoint-100"
-
-# Merge using CPU (if out of GPU memory)
-CUDA_VISIBLE_DEVICES="" axolotl merge-lora config.yml
-```
-
-Configuration options:
-
-```yaml
-gpu_memory_limit: Limit GPU memory usage
-lora_on_cpu: Load LoRA weights on CPU
-```
-
-#### merge-sharded-fsdp-weights
-
-Merges sharded FSDP model checkpoints into a single combined checkpoint.
-
-```bash
-# Basic merge
-axolotl merge-sharded-fsdp-weights config.yml
-```
-
-#### evaluate
-
-Evaluates a model's performance using metrics specified in the config.
-
-```bash
-# Basic evaluation
-axolotl evaluate config.yml
-```
-
-#### lm-eval
-
-Runs LM Evaluation Harness on your model.
-
-```bash
-# Basic evaluation
-axolotl lm-eval config.yml
-
-# Evaluate specific tasks
-axolotl lm-eval config.yml --tasks arc_challenge,hellaswag
-```
-
-Configuration options:
-
-```yaml
-lm_eval_tasks: List of tasks to evaluate
-lm_eval_batch_size: Batch size for evaluation
-output_dir: Directory to save evaluation results
-```
-
-### Legacy CLI Usage
-
-While the new Click-based CLI is preferred, Axolotl still supports the legacy module-based CLI:
-
-```bash
-# Preprocess
-python -m axolotl.cli.preprocess config.yml
-
-# Train
-accelerate launch -m axolotl.cli.train config.yml
-
-# Inference
-accelerate launch -m axolotl.cli.inference config.yml \
-    --lora_model_dir="./outputs/lora-out"
-
-# Gradio interface
-accelerate launch -m axolotl.cli.inference config.yml \
-    --lora_model_dir="./outputs/lora-out" --gradio
-```
-
-### Remote Compute with Modal Cloud
-
-Axolotl supports running training and inference workloads on Modal cloud infrastructure. This is configured using a
-cloud YAML file alongside your regular Axolotl config.
-
-#### Cloud Configuration
-
-Create a cloud config YAML with your Modal settings:
-
-```yaml
-# cloud_config.yml
-provider: modal
-gpu: a100  # Supported: l40s, a100-40gb, a100-80gb, a10g, h100, t4, l4
-gpu_count: 1    # Number of GPUs to use
-timeout: 86400  # Maximum runtime in seconds (24 hours)
-branch: main    # Git branch to use (optional)
-
-volumes:        # Persistent storage volumes
-  - name: axolotl-cache
-    mount: /workspace/cache
-
-env:            # Environment variables
-  - WANDB_API_KEY
-  - HF_TOKEN
-```
-
-#### Running on Modal Cloud
-
-Commands that support the --cloud flag:
-
-```bash
-# Preprocess on cloud
-axolotl preprocess config.yml --cloud cloud_config.yml
-
-# Train on cloud
-axolotl train config.yml --cloud cloud_config.yml
-
-# Train without accelerate on cloud
-axolotl train config.yml --cloud cloud_config.yml --no-accelerate
-
-# Run lm-eval on cloud
-axolotl lm-eval config.yml --cloud cloud_config.yml
-```
-
-#### Cloud Configuration Options
-
-```yaml
-provider: compute provider, currently only `modal` is supported
-gpu: GPU type to use
-gpu_count: Number of GPUs (default: 1)
-memory: RAM in GB (default: 128)
-timeout: Maximum runtime in seconds
-timeout_preprocess: Preprocessing timeout
-branch: Git branch to use
-docker_tag: Custom Docker image tag
-volumes: List of persistent storage volumes
-env: Environment variables to pass
-secrets: Secrets to inject
-```

docs/config.qmd

@@ -1,573 +0,0 @@
----
-title: Config options
-description: A complete list of all configuration options.
----
-
-```yaml
-# This is the huggingface model that contains *.pt, *.safetensors, or *.bin files
-# This can also be a relative path to a model on disk
-base_model: ./llama-7b-hf
-# You can specify an ignore pattern if the model repo contains more than 1 model type (*.pt, etc)
-base_model_ignore_patterns:
-# If the base_model repo on hf hub doesn't include configuration .json files,
-# You can set that here, or leave this empty to default to base_model
-base_model_config: ./llama-7b-hf
-# You can specify to choose a specific model revision from huggingface hub
-revision_of_model:
-# Optional tokenizer configuration path in case you want to use a different tokenizer
-# than the one defined in the base model
-tokenizer_config:
-# If you want to specify the type of model to load, AutoModelForCausalLM is a good choice too
-model_type: AutoModelForCausalLM
-# Corresponding tokenizer for the model AutoTokenizer is a good choice
-tokenizer_type: AutoTokenizer
-# Trust remote code for untrusted source
-trust_remote_code:
-# use_fast option for tokenizer loading from_pretrained, default to True
-tokenizer_use_fast:
-# Whether to use the legacy tokenizer setting, defaults to True
-tokenizer_legacy:
-# Resize the model embeddings when new tokens are added to multiples of 32
-# This is reported to improve training speed on some models
-resize_token_embeddings_to_32x:
-
-# (Internal use only)
-# Used to identify which the model is based on
-is_falcon_derived_model:
-is_llama_derived_model:
-is_qwen_derived_model:
-# Please note that if you set this to true, `padding_side` will be set to "left" by default
-is_mistral_derived_model:
-
-# optional overrides to the base model configuration
-overrides_of_model_config:
-  # RoPE Scaling https://github.com/huggingface/transformers/pull/24653
-  rope_scaling:
-    type: # linear | dynamic
-    factor: # float
-
-# optional overrides the base model loading from_pretrained
-overrides_of_model_kwargs:
-  # use_cache: False
-
-# optional overrides to the bnb 4bit quantization configuration
-# https://huggingface.co/docs/transformers/main/main_classes/quantization#transformers.BitsAndBytesConfig
-bnb_config_kwargs:
-  # These are default values
-  llm_int8_has_fp16_weight: false
-  bnb_4bit_quant_type: nf4
-  bnb_4bit_use_double_quant: true
-
-
-# Whether you are training a 4-bit GPTQ quantized model
-gptq: true
-
-# This will attempt to quantize the model down to 8 bits and use adam 8 bit optimizer
-load_in_8bit: true
-# Use bitsandbytes 4 bit
-load_in_4bit:
-
-# Use CUDA bf16
-bf16: true # bool or 'full' for `bf16_full_eval`. require >=ampere
-# Use CUDA fp16
-fp16: true
-# Use CUDA tf32
-tf32: true # require >=ampere
-
-# No AMP (automatic mixed precision)
-bfloat16: true # require >=ampere
-float16: true
-
-# Limit the memory for all available GPUs to this amount (if an integer, expressed in gigabytes); default: unset
-gpu_memory_limit: 20GiB
-# Do the LoRA/PEFT loading on CPU -- this is required if the base model is so large it takes up most or all of the available GPU VRAM, e.g. during a model and LoRA merge
-lora_on_cpu: true
-
-# A list of one or more datasets to finetune the model with
-datasets:
-  # HuggingFace dataset repo | s3://,gs:// path | "json" for local dataset, make sure to fill data_files
-  - path: vicgalle/alpaca-gpt4
-    # The type of prompt to use for training. [alpaca, gpteacher, oasst, reflection]
-    type: alpaca # format | format:<prompt_style> (chat/instruct) | <prompt_strategies>.load_<load_fn>
-    ds_type: # Optional[str] (json|arrow|parquet|text|csv) defines the datatype when path is a file
-    data_files: # Optional[str] path to source data files
-
-    shards: # Optional[int] split dataset into N pieces (use with shards_idx)
-    shards_idx: # Optional[int] = 0 the index of sharded dataset to use
-
-    preprocess_shards: # Optional[int] process dataset in N sequential chunks for memory efficiency (exclusive with `shards`)
-
-    name: # Optional[str] name of dataset configuration to load
-    train_on_split: train # Optional[str] name of dataset split to load from
-    revision: # Optional[str] The specific revision of the dataset to use when loading from the Hugging Face Hub. This can be a commit hash, tag, or branch name. If not specified, the latest version will be used. This parameter is ignored for local datasets.
-    trust_remote_code: # Optional[bool] Trust remote code for untrusted source
-
-  # Custom user instruction prompt
-  - path: repo
-    type:
-      # The below are defaults. only set what's needed if you use a different column name.
-      system_prompt: ""
-      system_format: "{system}"
-      field_system: system
-      field_instruction: instruction
-      field_input: input
-      field_output: output
-
-      # Customizable to be single line or multi-line
-      # Use {instruction}/{input} as key to be replaced
-      # 'format' can include {input}
-      format: |-
-        User: {instruction} {input}
-        Assistant:
-      # 'no_input_format' cannot include {input}
-      no_input_format: "{instruction} "
-
-      # For `completion` datsets only, uses the provided field instead of `text` column
-      field:
-
-  # Using chat template
-  - path: ...
-    # Set type to `chat_template` to use this strategy
-    type: chat_template
-    # Specify the name of the chat template to use
-    # The name of the chat template to use for training, following values are supported:
-    # - tokenizer_default: Uses the chat template that is available in the tokenizer_config.json. If the chat template is not available in the tokenizer, it will raise an error. This is the default.
-    # - alpaca/inst/chatml/gemma/cohere/llama3/phi_3/deepseek_v2/jamba: These chat templates are available in the axolotl codebase at src/axolotl/utils/chat_templates.py
-    # - tokenizer_default_fallback_*: where * is the name of the chat template to fallback to if the tokenizer does not have a chat template else default to tokenizer. E.g. tokenizer_default_fallback_chatml.
-    # - jinja: Uses a custom jinja template for the chat template. The custom jinja template should be provided in the chat_template_jinja field.
-    chat_template: tokenizer_default
-
-    # Custom jinja chat template. Used only if `chat_template: jinja` or empty.
-    chat_template_jinja:
-
-    # Key containing the messages (default: "messages")
-    field_messages: messages
-
-    # Mapping of properties from the input dataset to the chat template.
-    # (default: message_property_mappings={'role':'role', 'content':'content'})
-    # If a property exists in the template but not in this mapping, the system will attempt
-    # to load it directly from the message using the property name as the key.
-    # Example: In the mapping below, 'from' is loaded from input dataset and used as 'role',
-    # while 'value' is loaded and used as 'content' in the chat template.
-    message_property_mappings:
-      role: from
-      content: value
-      # ...
-
-    message_property_mappings:
-
-    # Optional[Dict[str, List]]. Roles mapping in the messages. The default is:
-    roles:
-      user: ["human", "user"]
-      assistant: ["gpt", "assistant"]
-      system: ["system"]
-      tool: ["tool"]
-
-    # IMPORTANT: The following fields determine which parts of the conversation to train on.
-    # Priority order: message_field_training > message_field_training_detail > train_on_inputs or role in roles_to_train
-    # See examples at `docs/dataset-formats/conversation.qmd`
-    # Note: If the below 4 fields are empty, defaults to training only on the last message.
-
-    # Optional[List[str]]. Roles to train on. The tokens from these roles will be considered for the loss.
-    roles_to_train: ["assistant"]  # default
-    # Optional[str]. Which EOS tokens to train on in the conversation. Possible values are:
-    # - all: train on all EOS tokens
-    # - turn (default): train on the EOS token at the end of each trainable turn
-    # - last: train on the last EOS token in the conversation
-    train_on_eos: last
-    # The key in the message turn that indicates via boolean whether tokens of a turn should be considered for training. Useful to selectively train on certain turns besides the `roles_to_train`.
-    message_field_training: training
-    # The key in the message turn that contains the training details. Useful to selectively train on certain tokens in a turn.
-    # The value of the key is a List[Dict] containing `begin_offset` (start character index in content), `end_offset` (end character index in content), and `train` (boolean whether to train).
-    message_field_training_detail: train_detail
-
-
-# If false, the datasets will not be shuffled and will keep their original order in `datasets`.
-# The same applies to the `test_datasets` option and the `pretraining_dataset` option. Default is true.
-shuffle_merged_datasets: true
-
-Deduplicates datasets and test_datasets with identical entries.
-dataset_exact_deduplication: true
-
-# A list of one or more datasets to eval the model with.
-# You can use either test_datasets, or val_set_size, but not both.
-test_datasets:
-  - path: /workspace/data/eval.jsonl
-    ds_type: json
-    # You need to specify a split. For "json" datasets the default split is called "train".
-    split: train
-    type: completion
-    data_files:
-      - /workspace/data/eval.jsonl
-
-# use RL training: 'dpo', 'ipo', 'kto'
-rl:
-# whether to perform weighting if doing DPO training. Boolean.
-dpo_use_weighting:
-
-# reward modelling: `True` or `False`
-reward_model:
-
-# process reward modelling: `True` or `False`
-process_reward_model:
-
-# The name of the chat template to use for training, following values are supported:
-# - tokenizer_default: Uses the chat template that is available in the tokenizer_config.json. If the chat template is not available in the tokenizer, it will raise an error. This is the default value.
-# - alpaca/inst/chatml/gemma/cohere/llama3/phi_3/deepseek_v2/jamba: These chat templates are available in the axolotl codebase at src/axolotl/utils/chat_templates.py
-# - tokenizer_default_fallback_*: where * is the name of the chat template to fallback to. E.g. tokenizer_default_fallback_chatml. This is useful when the chat template is not available in the tokenizer.
-# - jinja: Uses a custom jinja template for the chat template. The custom jinja template should be provided in the chat_template_jinja field.
-# The selected chat template will be saved to the tokenizer_config.json for easier inferencing
-# Note: It is recommended to set train_on_inputs to true when using a chat template that is different from the model's default chat template.
-chat_template: tokenizer_default
-# custom jinja template for chat template. This will be only used if chat_template is set to `jinja` or `null` (in which case chat_template is automatically set to `jinja`). Default is null.
-chat_template_jinja: null
-# Changes the default system message
-default_system_message: You are a helpful assistant. Please give a long and detailed answer. # Currently only supports chatml.
-# Axolotl attempts to save the dataset as an arrow after packing the data together so
-# subsequent training attempts load faster, relative path
-dataset_prepared_path: data/last_run_prepared
-# Push prepared dataset to hub
-push_dataset_to_hub: # repo path
-# The maximum number of processes to use while preprocessing your input dataset. This defaults to `os.cpu_count()`
-# if not set.
-dataset_processes: # defaults to os.cpu_count() if not set
-# Keep dataset in memory while preprocessing
-# Only needed if cached dataset is taking too much storage
-dataset_keep_in_memory:
-# push checkpoints to hub
-hub_model_id: # private repo path to push finetuned model
-# how to push checkpoints to hub
-# https://huggingface.co/docs/transformers/v4.31.0/en/main_classes/trainer#transformers.TrainingArguments.hub_strategy
-hub_strategy:
-# Whether to use hf `use_auth_token` for loading datasets. Useful for fetching private datasets
-# Required to be true when used in combination with `push_dataset_to_hub`
-hf_use_auth_token: # boolean
-# How much of the dataset to set aside as evaluation. 1 = 100%, 0.50 = 50%, etc. 0 for no eval.
-val_set_size: 0.04
-# Num shards for whole dataset
-dataset_shard_num:
-# Index of shard to use for whole dataset
-dataset_shard_idx:
-
-# The maximum length of an input to train with, this should typically be less than 2048
-# as most models have a token/context limit of 2048
-sequence_len: 2048
-# Pad inputs so each step uses constant sized buffers
-# This will reduce memory fragmentation and may prevent OOMs, by re-using memory more efficiently
-pad_to_sequence_len:
-# Use efficient multi-packing with block diagonal attention and per sequence position_ids. Recommend set to 'true'
-sample_packing:
-# Set to 'false' if getting errors during eval with sample_packing on.
-eval_sample_packing:
-# You can set these packing optimizations AFTER starting a training at least once.
-# The trainer will provide recommended values for these values.
-sample_packing_eff_est:
-total_num_tokens:
-# Increasing the following values helps with packing, but usually only slightly (<%1.)
-# The number of samples packed at a time.
-sample_packing_group_size: 100000
-# The number of samples which can be packed into one sequence. Increase if using a large sequence_len with many short samples.
-sample_packing_bin_size: 200
-# whether to concatenate samples during pretraining
-pretraining_sample_concatenation:
-
-# Use batch flattening for speedups when not using sample_packing
-batch_flattening:
-
-# Passed through to transformers when loading the model when launched without accelerate
-# Use `sequential` when training w/ model parallelism to limit memory
-device_map:
-# Defines the max memory usage per gpu on the system. Passed through to transformers when loading the model.
-max_memory:
-
-# If you want to use 'lora' or 'qlora' or leave blank to train all parameters in original model
-adapter: lora
-# If you already have a lora model trained that you want to load, put that here.
-# This means after training, if you want to test the model, you should set this to the value of `output_dir`.
-# Note that if you merge an adapter to the base model, a new subdirectory `merged` will be created under the `output_dir`.
-lora_model_dir:
-
-# LoRA hyperparameters
-# For more details about the following options, see:
-# https://www.anyscale.com/blog/fine-tuning-llms-lora-or-full-parameter-an-in-depth-analysis-with-llama-2
-lora_r: 8
-lora_alpha: 16
-lora_dropout: 0.05
-lora_target_modules:
-  - q_proj
-  - v_proj
-#  - k_proj
-#  - o_proj
-#  - gate_proj
-#  - down_proj
-#  - up_proj
-lora_target_linear: # If true, will target all linear modules
-peft_layers_to_transform: # The layer indices to transform, otherwise, apply to all layers
-
-# If you added new tokens to the tokenizer, you may need to save some LoRA modules because they need to know the new tokens.
-# For LLaMA and Mistral, you need to save `embed_tokens` and `lm_head`. It may vary for other models.
-# `embed_tokens` converts tokens to embeddings, and `lm_head` converts embeddings to token probabilities.
-# https://github.com/huggingface/peft/issues/334#issuecomment-1561727994
-lora_modules_to_save:
-#  - embed_tokens
-#  - lm_head
-
-lora_fan_in_fan_out: false
-
-# Apply custom LoRA autograd functions and activation function Triton kernels for
-# speed and memory savings
-# See: https://axolotl-ai-cloud.github.io/axolotl/docs/lora_optims.html
-lora_mlp_kernel: true
-lora_qkv_kernel: true
-lora_o_kernel: true
-
-# LoRA+ hyperparameters
-# For more details about the following options, see:
-# https://arxiv.org/abs/2402.12354  and `src/axolotl/core/train_builder.py`
-loraplus_lr_ratio: # loraplus learning rate ratio lr_B / lr_A. Recommended value is 2^4.
-loraplus_lr_embedding: #  loraplus learning rate for lora embedding layers. Default value is 1e-6.
-
-peft:
-  # Configuration options for loftq initialization for LoRA
-  # https://huggingface.co/docs/peft/developer_guides/quantization#loftq-initialization
-  loftq_config:
-    loftq_bits:  # typically 4 bits
-
-# ReLoRA configuration
-# Must use either 'lora' or 'qlora' adapter, and does not support fsdp or deepspeed
-relora_steps: # Number of steps per ReLoRA restart
-relora_warmup_steps: # Number of per-restart warmup steps
-relora_anneal_steps: # Number of anneal steps for each relora cycle
-relora_prune_ratio: # threshold for optimizer magnitude when pruning
-relora_cpu_offload: # True to perform lora weight merges on cpu during restarts, for modest gpu memory savings
-
-# wandb configuration if you're using it
-# Make sure your `WANDB_API_KEY` environment variable is set (recommended) or you login to wandb with `wandb login`.
-wandb_mode: # "offline" to save run metadata locally and not sync to the server, "disabled" to turn off wandb
-wandb_project: # Your wandb project name
-wandb_entity: # A wandb Team name if using a Team
-wandb_watch:
-wandb_name: # Set the name of your wandb run
-wandb_run_id: # Set the ID of your wandb run
-wandb_log_model: # "checkpoint" to log model to wandb Artifacts every `save_steps` or "end" to log only at the end of training
-
-# mlflow configuration if you're using it
-mlflow_tracking_uri: # URI to mlflow
-mlflow_experiment_name: # Your experiment name
-mlflow_run_name: # Your run name
-hf_mlflow_log_artifacts:  # set to true to copy each saved checkpoint on each save to mlflow artifact registry
-
-# Comet configuration if you're using it
-# Make sure your `COMET_API_KEY` environment variable is set (recommended) or you login to Comet with `comet login`.
-# Check out our documentation for more details https://www.comet.com/docs/v2/api-and-sdk/python-sdk/reference/Experiment-Creation/#comet_ml.start
-use_comet: # Enable or disable Comet integration.
-comet_api_key: # API key for Comet. Recommended to set via `comet login`.
-comet_workspace: # Workspace name in Comet. Defaults to the user's default workspace.
-comet_project_name: # Project name in Comet. Defaults to Uncategorized.
-comet_experiment_key: # Identifier for the experiment. Used to append data to an existing experiment or control the key of new experiments. Default to a random key.
-comet_mode: # Create a new experiment ("create") or log to an existing one ("get"). Default ("get_or_create") auto-selects based on configuration.
-comet_online: # Set to True to log data to Comet server, or False for offline storage. Default is True.
-comet_experiment_config: # Dictionary for additional configuration settings, see the doc for more details.
-
-# Tensorboard
-use_tensorboard: # Optional[bool]
-
-# Where to save the full-finetuned model to
-output_dir: ./completed-model
-
-# Whether to use torch.compile and which backend to use
-# setting to `auto` will enable torch compile when torch>=2.5.1
-torch_compile:  # Optional[Union[Literal["auto"], bool]]
-torch_compile_backend:  # Optional[str]
-
-# Training hyperparameters
-
-# If greater than 1, backpropagation will be skipped and the gradients will be accumulated for the given number of steps.
-gradient_accumulation_steps: 1
-# The number of samples to include in each batch. This is the number of samples sent to each GPU.
-# Batch size per gpu = micro_batch_size * gradient_accumulation_steps
-micro_batch_size: 2
-eval_batch_size:
-num_epochs: 4
-warmup_steps: 100  # cannot use with warmup_ratio
-warmup_ratio: 0.05  # cannot use with warmup_steps
-learning_rate: 0.00003
-lr_quadratic_warmup:
-logging_steps:
-eval_steps: # Leave empty to eval at each epoch, integer for every N steps. float for fraction of total steps
-evals_per_epoch: # number of times per epoch to run evals, mutually exclusive with eval_steps
-eval_strategy: # Set to `"no"` to skip evaluation, `"epoch"` at end of each epoch, leave empty to infer from `eval_steps`.
-save_strategy: # Set to `"no"` to skip checkpoint saves, `"epoch"` at end of each epoch, `"best"` when better result is achieved, leave empty to infer from `save_steps`.
-save_steps: # Leave empty to save at each epoch, integer for every N steps. float for fraction of total steps
-saves_per_epoch: # number of times per epoch to save a checkpoint, mutually exclusive with save_steps
-save_total_limit: # Checkpoints saved at a time
-# Maximum number of iterations to train for. It precedes num_epochs which means that
-# if both are set, num_epochs will not be guaranteed.
-# e.g., when 1 epoch is 1000 steps => `num_epochs: 2` and `max_steps: 100` will train for 100 steps
-max_steps:
-
-# bool of whether to include tokens trainer per second in the training metrics. This iterates over the entire dataset once, so it takes some time.
-include_tokens_per_second:
-
-eval_table_size: # Approximate number of predictions sent to wandb depending on batch size. Enabled above 0. Default is 0
-eval_max_new_tokens: # Total number of tokens generated for predictions sent to wandb. Default is 128
-eval_causal_lm_metrics: # HF evaluate metrics used during evaluation. Default is ["sacrebleu", "comet", "ter", "chrf", "perplexity"]
-
-profiler_steps: # enable the pytorch profiler to capture the first N steps of training to the output_dir.
-                # see https://pytorch.org/blog/understanding-gpu-memory-1/ for more information
-                # snapshots can be visualized @ https://pytorch.org/memory_viz
-
-loss_watchdog_threshold: # High loss value, indicating the learning has broken down (a good estimate is ~2 times the loss at the start of training)
-loss_watchdog_patience: # Number of high-loss steps in a row before the trainer aborts (default: 3)
-
-# Save model as safetensors (require safetensors package)
-save_safetensors:
-
-# Whether to mask out or include the human's prompt from the training labels
-train_on_inputs: false
-# Group similarly sized data to minimize padding.
-# May be slower to start, as it must download and sort the entire dataset.
-# Note that training loss may have an oscillating pattern with this enabled.
-group_by_length: false
-
-# Whether to use gradient checkpointing https://huggingface.co/docs/transformers/v4.18.0/en/performance#gradient-checkpointing
-gradient_checkpointing: false
-# additional kwargs to pass to the trainer for gradient checkpointing
-# gradient_checkpointing_kwargs:
-#   use_reentrant: true
-
-# Stop training after this many evaluation losses have increased in a row
-# https://huggingface.co/transformers/v4.2.2/_modules/transformers/trainer_callback.html#EarlyStoppingCallback
-early_stopping_patience: 3
-
-# Specify a scheduler and kwargs to use with the optimizer
-lr_scheduler: # 'one_cycle' | 'log_sweep' | empty for cosine
-lr_scheduler_kwargs:
-cosine_min_lr_ratio: # decay lr to some percentage of the peak lr, e.g. cosine_min_lr_ratio=0.1 for 10% of peak lr
-cosine_constant_lr_ratio: # freeze lr at some percentage of the step, e.g. cosine_constant_lr_ratio=0.8 means start cosine_min_lr at 80% of training step (https://arxiv.org/pdf/2308.04014.pdf)
-
-# For one_cycle optim
-lr_div_factor: # Learning rate div factor
-
-# Specify optimizer
-# Valid values are driven by the Transformers OptimizerNames class, see:
-# https://github.com/huggingface/transformers/blob/95b374952dc27d8511541d6f5a4e22c9ec11fb24/src/transformers/training_args.py#L134
-#
-# Note that not all optimizers may be available in your environment, ex: 'adamw_anyprecision' is part of
-# torchdistx, 'adamw_bnb_8bit' is part of bnb.optim.Adam8bit, etc. When in doubt, it is recommended to start with the optimizer used
-# in the examples/ for your model and fine-tuning use case.
-#
-# Valid values for 'optimizer' include:
-# - adamw_hf
-# - adamw_torch
-# - adamw_torch_fused
-# - adamw_torch_xla
-# - adamw_apex_fused
-# - adopt_adamw (an EXPERIMENTAL optimizer, only for torch version >= 2.5.1)
-# - adafactor
-# - adamw_anyprecision
-# - sgd
-# - adagrad
-# - adamw_bnb_8bit
-# - lion_8bit
-# - lion_32bit
-# - paged_adamw_32bit
-# - paged_adamw_8bit
-# - paged_lion_32bit
-# - paged_lion_8bit
-# - galore_adamw
-# - galore_adamw_8bit
-# - galore_adafactor
-# - galore_adamw_layerwise
-# - galore_adamw_8bit_layerwise
-# - galore_adafactor_layerwise
-optimizer:
-# Dictionary of arguments to pass to the optimizer
-optim_args:
-# For Galore Optimizers the following optim_args are available
-# rank:  # type: int
-# update_proj_gap  # type: int
-# scale  # type: float
-# proj_type:  # type: str, default = std
-
-# The target modules to optimize, i.e. the module names that you would like to train, right now this is used only for GaLore algorithm
-optim_target_modules:
-# - self_attn  # for llama
-# - mlp
-
-# Specify weight decay
-weight_decay:
-# adamw hyperparams
-adam_beta1:
-adam_beta2:
-adam_epsilon:
-# Gradient clipping max norm
-max_grad_norm:
-
-# Augmentation techniques
-# NEFT https://arxiv.org/abs/2310.05914, set this to a number (paper default is 5) to add noise to embeddings
-# currently only supported on Llama and Mistral
-neftune_noise_alpha:
-
-# Whether to bettertransformers
-flash_optimum:
-# Whether to use xformers attention patch https://github.com/facebookresearch/xformers:
-xformers_attention:
-# Whether to use flash attention patch https://github.com/Dao-AILab/flash-attention:
-flash_attention:
-flash_attn_cross_entropy:  # Whether to use flash-attention cross entropy implementation - advanced use only
-flash_attn_rms_norm:  # Whether to use flash-attention rms norm implementation - advanced use only
-flash_attn_fuse_qkv: # Whether to fuse QKV into a single operation
-flash_attn_fuse_mlp: # Whether to fuse part of the MLP into a single operation
-# Whether to use scaled-dot-product attention
-# https://pytorch.org/docs/stable/generated/torch.nn.functional.scaled_dot_product_attention.html
-sdp_attention:
-# Shifted-sparse attention (only llama) - https://arxiv.org/pdf/2309.12307.pdf
-s2_attention:
-# Resume from a specific checkpoint dir
-resume_from_checkpoint:
-# If resume_from_checkpoint isn't set and you simply want it to start where it left off.
-# Be careful with this being turned on between different models.
-auto_resume_from_checkpoints: false
-
-# Don't mess with this, it's here for accelerate and torchrun
-local_rank:
-
-# Add or change special tokens.
-# If you add tokens here, you don't need to add them to the `tokens` list.
-special_tokens:
-  # bos_token: "<s>"
-  # eos_token: "</s>"
-  # unk_token: "<unk>"
-  # pad_token: "[PAD]"
-
-# Add extra tokens.
-tokens:
-
-# FSDP
-fsdp:
-fsdp_config:
-
-# Deepspeed config path. e.g., deepspeed_configs/zero3.json
-deepspeed:
-
-# Advanced DDP Arguments
-ddp_timeout:
-ddp_bucket_cap_mb:
-ddp_broadcast_buffers:
-
-# Path to torch distx for optim 'adamw_anyprecision'
-torchdistx_path:
-
-# Set to HF dataset for type: 'completion' for streaming instead of pre-tokenize
-pretraining_dataset:
-
-# Debug mode
-debug:
-
-# Seed
-seed:
-
-# Allow overwrite yml config using from cli
-strict:
-```

docs/dataset-formats/conversation.qmd

@@ -1,152 +0,0 @@
----
-title: Conversation
-description: Conversation format for supervised fine-tuning.
-order: 3
----
-
-## sharegpt
-
-IMPORTANT: ShareGPT is deprecated!. Please see [chat_template](#chat_template) section below.
-
-## pygmalion
-
-```{.json filename="data.jsonl"}
-{"conversations": [{"role": "...", "value": "..."}]}
-```
-
-## chat_template
-
-Chat Template strategy uses a jinja2 template that converts a list of messages into a prompt. Support using tokenizer's template, a supported template, or custom jinja2.
-
-```{.json filename="data.jsonl"}
-{"conversations": [{"role": "...", "content": "..."}]}
-```
-
-See [configs](../config.qmd) for full configs and supported templates.
-
-### Migrating from sharegpt
-
-Most configs can be adapted as follows:
-
-```yaml
-# old
-chat_template: chatml
-datasets:
-  - path: ...
-    type: sharegpt
-    conversation: chatml
-
-# new (if using tokenizer's chat_template)
-datasets:
-  - path: ...
-    type: chat_template
-
-    field_messages: conversations
-    message_property_mappings:
-      role: from
-      content: value
-
-# new (if setting a new chat_template like chatml, gemma, etc)
-chat_template: chatml
-datasets:
-  - path: ...
-    type: chat_template
-
-    field_messages: conversations
-    message_property_mappings:
-      role: from
-      content: value
-```
-
-We recommend checking the below examples for other usecases.
-
-### Examples
-
-1. Using the default chat template in the tokenizer_config.json on OpenAI messages format, training on only last message.
-
-```yaml
-datasets:
-  - path: ...
-    type: chat_template
-    roles_to_train:
-    train_on_eos:
-```
-
-2. Using the `gemma` chat template to override the tokenizer_config.json's chat template on OpenAI messages format, training on all assistant messages.
-
-```yaml
-chat_template: gemma # this overwrites the tokenizer's chat_template
-datasets:
-  - path: ...
-    type: chat_template
-    roles_to_train: ["assistant"]  # default value
-```
-
-3. Using the tokenizer_config.json's chat template or `chatml` as fallback if the former's chat template does not exist, on OpenAI messages format, training on all assistant messages.
-
-```yaml
-chat_template: tokenizer_default_fallback_chatml # this overwrites the tokenizer's chat_template
-datasets:
-  - path: ...
-    type: chat_template
-```
-
-4. Using a custom jinja template on OpenAI messages format, training on all assistant messages.
-
-```yaml
-# chat_template: jinja # `jinja` will be implied if the `chat_template_jinja` is set and this field is empty
-chat_template_jinja: "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'system') %}{{'<|system|>' + '\n' + message['content'] + '<|end|>' + '\n'}}{% elif (message['role'] == 'user') %}{{'<|user|>' + '\n' + message['content'] + '<|end|>' + '\n' + '<|assistant|>' + '\n'}}{% elif message['role'] == 'assistant' %}{{message['content'] + '<|end|>' + '\n'}}{% endif %}{% endfor %}"
-
-datasets:
-  - path: ...
-    type: chat_template
-```
-
-5. (Advanced) Using fine-grained control over tokens and turns to train in a conversation
-
-For a data sample that looks like:
-
-```{.json filename="data.jsonl"}
-{
-  "conversations": [
-    {"from": "system", "value": "You are an AI assistant.", "train": false},
-    {"from": "human", "value": "Hello", "train": false},
-    {"from": "assistant", "value": "Hello", "train": true},
-    {"from": "human", "value": "How are you?", "train": true},
-    {
-      "from": "assistant",
-      "value": "I'm doing very well, thank you!",
-      "train_detail": [
-        {"begin_offset": 0, "end_offset": 8, "train": false},
-        {"begin_offset": 9, "end_offset": 18, "train": true},
-        {"begin_offset": 19, "end_offset": 30, "train": false},
-      ],
-    },
-    {
-        "from": "human",
-        "value": "I'm doing very well, thank you!",
-        "train": true,
-    },
-    {"from": "assistant", "value": "Hi there!", "train": true}
-  ]
-}
-```
-
-The configuration would look like:
-
-```yaml
-datasets:
-  - path: ...
-    type: chat_template
-    chat_template: tokenizer_default
-    field_messages: conversations
-    message_property_mappings:
-      role: from
-      content: value
-    roles_to_train: []
-    train_on_eos: turn
-    message_field_training: train
-    message_field_training_detail: train_detail
-```
-
-Tip: It is not necessary to use both `message_field_training` and `message_field_training_detail` at a time.

docs/dataset-formats/index.qmd

@@ -1,458 +0,0 @@
----
-title: Dataset Formats
-description: Guide to Dataset Formats in Axolotl
-back-to-top-navigation: true
-toc: true
-toc-depth: 5
----
-
-
-Axolotl is a training framework that aims to make the process convenient yet flexible to users by simply passing a config yaml file.
-
-As there are a lot of available options in Axolotl, this guide aims to provide an simplify the user experience to choosing the proper choice.
-
-Axolotl supports 3 kinds of training methods: pre-training, supervised fine-tuning, and preference-based post-training (e.g. DPO, ORPO, PRMs). Each method has their own dataset format which are described below.
-
-## [Pre-training](pretraining.qmd)
-
-When aiming to train on large corpora of text datasets, pre-training is your go-to choice. Due to the size of these datasets, downloading the entire-datasets before beginning training would be prohibitively time-consuming. Axolotl supports [streaming](https://huggingface.co/docs/datasets/en/stream) to only load batches into memory at a time.
-
-A sample format for a pre-training dataset is as follows:
-
-```json
-{"text": "first row"}
-{"text": "second row"}
-...
-```
-
-It is typically recommended to save your dataset as `.jsonl` due to its flexibility and simplicity.
-
-Axolotl supports loading from a Hugging Face hub repo or from local files.
-
-::: {.callout-important}
-For pre-training only, Axolotl would split texts if it exceeds the context length into multiple smaller prompts.
-:::
-
-### Pre-training from Hugging Face hub datasets
-
-As an example, to train using a Hugging Face dataset `hf_org/name`, you can pass the following config:
-
-```yaml
-pretraining_dataset: hf_org/name
-```
-
-### Pre-training from local dataset files
-
-Given a few corpus files: `A.jsonl`, `B.jsonl`, and `C.jsonl`, your config will look like the below:
-
-```yaml
-pretraining_dataset:
-  - path: json
-    data_files:
-      - A.jsonl
-      - B.jsonl
-      - C.jsonl
-```
-
-While we recommend `.jsonl`, you can also use the other formats (`csv`, `parquet`, `arrow`, `SQL`, `Webdataset`) that are supported by [`Dataset.load_dataset`](https://huggingface.co/docs/datasets/loading#local-and-remote-files)
-
-### Pre-training without streaming
-
-On the rare case that the dataset is small and can be loaded entirely into memory, another approach to running pre-training is to use the `completion` format. This would mean that the entire dataset is pre-tokenized instead of on-demand in streaming.
-
-One benefit of this is that the tokenization can be performed separately on a CPU-only machine, and then transferred to a GPU machine for training to save costs.
-
-From Hugging Face:
-
-```yaml
-datasets:
-  - path: hf_org/name
-    type: completion
-```
-
-From local files (either example works):
-
-```yaml
-datasets:
-  - path: A.jsonl
-    type: completion
-
-  - path: json
-    data_files: ["A.jsonl", "B.jsonl", "C.jsonl"]
-    type: completion
-```
-
-### Pre-training dataset configuration tips
-
-#### Setting max_steps
-
-When using streaming for large datasets, Axolotl does not know in advance how large the dataset is and does not know when to stop.
-
-Therefore, it is necessary to set `max_steps: int` in your config for pre-training to run, so that Axolotl knows when to stop training.
-
-One step is equal to `sequence_len * micro_batch_size * gradient_accumulation_steps * total_num_gpus` tokens.
-
-#### Group_by_length
-
-It is recommended to leave this off if downloading from Hugging Face hub as it would download the entire dataset which can be very large.
-
-## Supervised fine-tuning (SFT)
-
-Supervised fine-tuning is the process of training models to respond to an instruction or chat input.
-
-As there are a wide variety of dataset formats, Axolotl tries to support a majority of the formats available in public datasets.
-
-Axolotl provides four approaches for loading datasets, however, it's easier to work backwards from the dataset you have available to figure out which approach to use.
-
-A flow chart is as follows:
-
-1. Do you already have the dataset tokenized? If yes, check [Pre-Tokenized Dataset](#pre-tokenized-dataset).
-
-2. Do you want to format the dataset yourself and manually choose each section to mask? If yes, check [Template Free Dataset](#template-free-dataset)
-
-3. Is your dataset in a "conversation" format, containing a `list[messages]`? If yes, check [Conversation Dataset](#conversation-dataset)
-
-4. Is your dataset in an "instruct" format, containing `{ instruction, response }`? If yes, check [Instruction Dataset](#instruction-dataset)
-
-If you went through the flow chart and did not find one that matches, it is recommended to preprocess your dataset into one of the above or create a thread on Github Discussion.
-
-::: {.callout-tip}
-You can mix and match within each approach or across approaches to train a model on a variety of datasets.
-:::
-
-### [Pre-Tokenized Dataset](tokenized.qmd)
-
-We suggest this approach when you want to bring your own tokenized dataset.
-
-Axolotl expects the dataset to have three keys:
-- `input_ids`: from tokenizing formatted prompt
-- `attention_mask`: for masking padding. If you don't add padding, it would be equal to `len(input_ids) * [1]`
-- `labels`: this is the same as `input_ids`, however, if you want to mask certain tokens, you would set those indices to `-100`.
-
-::: {.callout-tip}
-Make sure to add BOS/EOS tokens to your prompt and mask it appropriately.
-:::
-
-A config for this would look like:
-
-```yaml
-datasets:
-  - path: A.jsonl
-    type:
-```
-
-::: {.callout-note}
-`type: ` is empty!
-:::
-
-### [Template Free Dataset](template_free.qmd)
-
-We reccomend this approach when you want granular control over the prompt formatting, special tokens, and masking, whilst letting Axolotl handle the tokenization. This is very useful if your dataset has unique prompts that differ across samples and where one single general template wouldn't suffice.
-
-In the example below, you could see that there is no proper structure. At the same time, it's very flexible as there are no constraints on how your prompt can look.
-
-```json
-{
-    "segments": [
-        {
-            "label": true,
-            "text": "<s>Hello\n"
-        },
-        {
-            "label": true,
-            "text": "hi there!. "
-        },
-        {
-            "label": false,
-            "text": "goodbye "
-        },
-        {
-            "label": true,
-            "text": "farewell</s>"
-        }
-    ]
-}
-```
-
-Each prompt must be have a key called `segments` which is a list of `{ text, label }`.
-
-```yaml
-datasets:
-  - path: A.jsonl
-    type: input_output
-```
-
-### [Conversation Dataset](conversation.qmd)
-
-`conversation` messages are a list of messages which usually contain a `role` and `content` key.
-
-::: {.callout-tip}
-Fun fact: Axolotl synonymously refers to "chat" messages as `conversation` messages due to how FastChat initially used this term to build a widely used [fastchat conversation](https://github.com/lm-sys/FastChat/blob/main/fastchat/conversation.py) method for formatting chat messages prior to the creation of `chat_templates`.
-:::
-
-#### What are `chat_templates`?
-
-The current most popular and convenient method for inference is to use `chat_templates` for formatting prompts. Axolotl supports using `chat_templates` for training to ensure that the model performs in the same environment as in inference.
-
-Here's a quick rundown on `chat_template`: A `chat_template` is a Jinja2 template which formats a list of messages into a prompt.
-
-An example of a prompt formatted into a popular template called ChatML can be seen below:
-
-Single prompt (pretty-printed):
-```json
-{
-    "messages": [
-        {
-            "role": "user",
-            "content": "Hi"
-        },
-        {
-            "role": "assistant",
-            "content": "How can I help you?"
-        },
-        {
-            "role": "user",
-            "content": "Can you add 3+5?"
-        },
-        {
-            "role": "assistant",
-            "content": "The answer is 8."
-        }
-    ]
-}
-```
-
-The ChatML template is as follows:
-```jinja2
-{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}
-```
-
-The above prompt formatted into this template will result in:
-
-```
-<|im_start|>user
-Hi<|im_end|>
-<|im_start|>assistant
-How can I help you?<|im_end|>
-<|im_start|>user
-Can you add 3+5?<|im_end|>
-<|im_start|>assistant
-The answer is 8.<|im_end|>
-```
-
-By using delimiters (`<|im_start|>` and `<|im_end|>`), a prompt separates different speakers which helps the model identify which portion belongs to whom.
-
-#### Common Conversation Dataset formats
-
-Older conversation datasets with the following format are colloquially called `sharegpt` datasets.
-
-```json
-{"conversations": [{"from": "...", "value": "..."}]}
-```
-
-Newer conversation datasets usually follow the OpenAI format.
-
-```json
-{"messages": [{"role": "...", "content": "..."}]}
-```
-
-Axolotl supports both as well as allowing customization of any kind of key.
-
-#### [Chat Template Usage](conversation.qmd#chat_template)
-
-To properly use this method, it is important to identify three things:
-
-1. Which `chat_template` would you use?
-
-2. What are the keys in your dataset, and what are the possible roles? For example, in OpenAI format, the keys would be `messages`, `role`, and `content`, respectively, whereas the possible roles are `system`, `user`, and `assistant`.
-
-3. What do you want to mask? For instance, only assistant messages, only last message, or nothing.
-
-##### Choosing a `chat_template`
-
-There are a lot of `chat_templates` out there. Axolotl supports the common ones: [supported chat templates](https://github.com/axolotl-ai-cloud/axolotl/blob/860609392184cf62a7e0ca676658b170e059ce6c/src/axolotl/utils/chat_templates.py#L17). For example, to use ChatML, it would be `chat_template: chatml`.
-
-However, it is also possible to use the already configured template within the tokenizer by specifying `chat_template: tokenizer_default`. If you want a fallback (in case some tokenizer does not have it pre-configured), you can do `chat_template: tokenizer_default_fallback_chatml` to fallback to the ChatML template if a tokenizer template was not found.
-
-One last but powerful approach is to bring your own template. This can be set via:
-
-```yaml
-chat_template_jinja: # your template
-```
-
-##### Setting `chat_template` dataset keys
-
-We currently default to OpenAI format for dataset keys, so if that's your current dataset format, there's nothing to do here.
-
-If your dataset format is different, here are the keys you should check (with their defaults):
-
-```yaml
-datasets:
-    ...
-    field_messages: messages  # this should point to the key containing the list of conversations
-    message_property_mappings:  # this is a mapping from keys in your dataset to keys in chat_template
-      role: role
-      content: content
-```
-
-In some `chat_templates` (e.g. [Gemma](https://huggingface.co/google/gemma-2b-it/blob/main/tokenizer_config.json#L1507)), the roles are hardcoded to `user` and `assistant`. Consequently, you may find it necessary to map the roles in your dataset to these above. We currently have some defaults that should work for common datasets, but if you get a `KeyError`, it would be necessary to add mapping for your roles. Here is an example of how it would look like:
-
-```yaml
-datasets:
-    ...
-    roles:
-      assistant:
-        - gpt
-        - model
-      user:
-        - human
-```
-
-In the example above, all `gpt` and `model` values are converted to `assistant`. All `human` values are converted to `user.`
-
-##### Handling masking
-
-The common use case for `chat_template` is for chat messages, therefore, it is common to mask all non-assistant messages. Assistant messages refer to the bot messages that you want the model to learn on.
-
-To train on all `assistant` messages, you would set the following configs.
-
-```yaml
-datasets:
-    ...
-    roles_to_train: ["assistant"]
-    train_on_eos: "turn"
-```
-
-The `train_on_eos` config means that it would mask all EOS tokens for turns that aren't assistant-turns. The other options are: `all` and `last` to choose which EOS to train on.
-
-Perhaps, you want to train on `assistant` and `narrator` roles, you can simply add `narrator` to the list of `roles_to_train`. You would also need to add it to the mapping of `roles` above.
-
-```yaml
-datasets:
-    ...
-    roles_to_train: ["assistant", "narrator"]
-    roles:
-      assistant:
-        - gpt
-        - model
-      user:
-        - human
-      narrator: ["narrator"]
-```
-
-#### Applying `chat_template`
-
-Once all the above steps are completed, you could combine all these configs together to form a bespoke configuration for your  custom dataset. The final step would be to correctly set the EOS token in your config:
-
-```yaml
-datasets:
-  - path: A.jsonl
-    type: chat_template
-
-    # step 1
-    chat_template: chatml
-
-    # step 2
-    field_messages: messages
-    message_property_mappings:
-      role: role
-      content: content
-
-    roles:
-      assistant:
-        - gpt
-        - model
-        - assistant
-      user:
-        - human
-        - user
-
-    # step 3
-    roles_to_train: ["assistant"]
-    train_on_eos: "turn"
-
-special_tokens:
-  eos_token: <|im_end|>
-```
-
-If this config were to be applied to the sample dataset above, the output would look as such (which can be retrieved via `axolotl preprocess config.yaml --debug`):
-
-```
-<|im_start|>(-100, 128256) user(-100, 882)
-(-100, 198) Hi(-100, 13347) <|im_end|>(-100, 128257)
-(-100, 198) <|im_start|>(-100, 128256) assistant(-100, 78191)
-(-100, 198) How(4438, 4438)  can(649, 649)  I(358, 358)  help(1520, 1520)  you(499, 499) ?(30, 30) <|im_end|>(128257, 128257)
-(-100, 198) <|im_start|>(-100, 128256) user(-100, 882)
-(-100, 198) Can(-100, 6854)  you(-100, 499)  add(-100, 923)  (-100, 220) 3(-100, 18) +(-100, 10) 5(-100, 20) ?(-100, 30) <|im_end|>(-100, 128257)
-(-100, 198) <|im_start|>(-100, 128256) assistant(-100, 78191)
-(-100, 198) The(791, 791)  answer(4320, 4320)  is(374, 374)  (220, 220) 8(23, 23) .(13, 13) <|im_end|>(128257, 128257)
-(-100, 198)
-```
-
-The first number refers to the label, the second refers to the `token_id`. For example, `-100` labels appear on non-assistant portions, meaning that they are masked during. For assistant portions, the label is the same as the `token_id`.
-
-### [Instruction Dataset](inst_tune.qmd)
-
-Instruction datasets are used to train instruction-following models and comprise a prompt, containing an instruction, and a single response. In contrast to chat datasets which may be multi-turn, instruct datasets are typically single-turn.
-
-An example is of a common format called Alpaca:
-```json
-{"instruction": "...", "input": "...", "output": "..."}
-```
-
-Using those keys, a prompt can be built based on it.
-```
-Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
-
-### Instruction:
-{instruction}
-
-### Input:
-{input}
-
-### Response:
-{output}
-```
-
-This can be configured as such:
-```yaml
-datasets:
-  - path: A.jsonl
-    type: alpaca
-```
-
-Axolotl supports many kinds of instruction dataset. All of them can be found here (https://axolotl-ai-cloud.github.io/axolotl/docs/dataset-formats/inst_tune.html) with their respective type and sample row format.
-
-#### Custom Instruct Prompt Format
-
-Due to the myriad possibilities of instruction formats, Axolotl allows customizing your own instruction format without having to dive into the code directly.
-
-In the example below, a sample row is used to output in `mistral_v1` format.
-```json
-{"input": "...", "output": "..."}
-```
-
-```yaml
-datasets:
-  - path: repo
-    type:
-      system_prompt: ""
-
-      field_system:
-      field_instruction: input
-      field_input:
-      field_output: output
-
-      # multi-line example with input
-      format: |-
-        [INST] {instruction} {input} [/INST]
-
-      # single-line example without input
-      no_input_format: "[INST] {instruction} [/INST]"
-```
-
-The config sets that the `field_instruction` is actually named `input`, and the `field_input` is empty as we don't have an `input` in this sample. Generally, `instruction` can be thought as the question to the model, and `input` as the additional information with `output` being the response. It is not necessary to have an `input` nor `system`. In the end, the most important part is to understand what format you want it to look like and how you can customize this to your use case.
-
-## Reinforcement Learning from Human Feedback (RLHF)
-
-As there are multiple RLHF methods with their own dataset requirements. Please see [RLHF datasets](../rlhf.qmd) documentation for more detail.

docs/dataset-formats/inst_tune.qmd

@@ -1,189 +0,0 @@
----
-title: Instruction Tuning
-description: Instruction tuning formats for supervised fine-tuning.
-order: 2
----
-
-## alpaca
-
-instruction; input(optional)
-
-```{.json filename="data.jsonl"}
-{"instruction": "...", "input": "...", "output": "..."}
-```
-
-## jeopardy
-
-question and answer
-
-```{.json filename="data.jsonl"}
-{"question": "...", "category": "...", "answer": "..."}
-```
-
-## oasst
-
-instruction
-
-```{.json filename="data.jsonl"}
-{"INSTRUCTION": "...", "RESPONSE": "..."}
-```
-
-## gpteacher
-
-instruction; input(optional)
-
-```{.json filename="data.jsonl"}
-{"instruction": "...", "input": "...", "response": "..."}
-```
-
-## reflection
-
-instruction with reflect; input(optional)
-
-```{.json filename="data.jsonl"}
-{"instruction": "...", "input": "...", "output": "...", "reflection": "...", "corrected": "..."}
-```
-
-## explainchoice
-
-question, choices, (solution OR explanation)
-
-```{.json filename="data.jsonl"}
-{"question": "...", "choices": ["..."], "solution": "...", "explanation": "..."}
-```
-
-## concisechoice
-
-question, choices, (solution OR explanation)
-
-```{.json filename="data.jsonl"}
-{"question": "...", "choices": ["..."], "solution": "...", "explanation": "..."}
-```
-
-## summarizetldr
-
-article and summary
-
-```{.json filename="data.jsonl"}
-{"article": "...", "summary": "..."}
-```
-
-## alpaca_chat
-
-basic instruct for alpaca chat
-
-```{.json filename="data.jsonl"}
-{"instruction": "...", "input": "...", "response": "..."}
-```
-
-## alpaca_chat.load_qa
-
-question and answer for alpaca chat
-
-```{.json filename="data.jsonl"}
-{"question": "...", "answer": "..."}
-```
-
-## alpaca_chat.load_concise
-
-question and answer for alpaca chat, for concise answers
-
-```{.json filename="data.jsonl"}
-{"instruction": "...", "input": "...", "response": "..."}
-```
-
-## alpaca_chat.load_camel_ai
-
-question and answer for alpaca chat, for load_camel_ai
-
-```{.json filename="data.jsonl"}
-{"message_1": "...", "message_2": "..."}
-```
-
-## alpaca_w_system.load_open_orca
-
-support for open orca datasets with included system prompts, instruct
-
-```{.json filename="data.jsonl"}
-{"system_prompt": "...", "question": "...", "response": "..."}
-```
-
-## context_qa
-
-in context question answering from an article
-
-```{.json filename="data.jsonl"}
-{"article": "...", "question": "...", "answer": "..."}
-```
-
-## context_qa.load_v2
-
-in context question answering (alternate)
-
-```{.json filename="data.jsonl"}
-{"context": "...", "question": "...", "answer": "..."}
-```
-
-## context_qa.load_404
-
-in context question answering from an article, with default response for no answer from context
-
-```{.json filename="data.jsonl"}
-{"article": "...", "unanswerable_question": "..."}
-```
-
-## creative_acr.load_answer
-
-instruction and revision
-
-```{.json filename="data.jsonl"}
-{"instruction": "...", "revision": "..."}
-```
-
-## creative_acr.load_critique
-
-critique
-
-```{.json filename="data.jsonl"}
-{"scores": "...", "critiques": "...", "instruction": "...", "answer": "..."}
-```
-
-## creative_acr.load_revise
-
-critique and revise
-
-```{.json filename="data.jsonl"}
-{"scores": "...", "critiques": "...", "instruction": "...", "answer": "...", "revision": "..."}
-```
-
-## metharme
-
-instruction, adds additional eos tokens
-
-```{.json filename="data.jsonl"}
-{"prompt": "...", "generation": "..."}
-```
-
-## How to add custom prompt format
-
-For a dataset that is preprocessed for instruction purposes:
-
-```{.json filename="data.jsonl"}
-{"input": "...", "output": "..."}
-```
-
-You can use this example in your YAML config:
-
-```{.yaml filename="config.yaml"}
-datasets:
-  - path: repo
-    type:
-      system_prompt: ""
-      field_system: system
-      field_instruction: input
-      field_output: output
-      format: "[INST] {instruction} [/INST]"
-      no_input_format: "[INST] {instruction} [/INST]"
-```
-
-See full config options under [here](../config.qmd).

docs/dataset-formats/pretraining.qmd

@@ -1,33 +0,0 @@
----
-title: Pre-training
-description: Data format for a pre-training completion task.
-order: 1
----
-
-For pretraining, there is no prompt template or roles.  The only required field is `text`:
-
-```{.json filename="data.jsonl"}
-{"text": "first row"}
-{"text": "second row"}
-...
-```
-
-:::{.callout-note}
-
-### Streaming is recommended for large datasets
-
-Axolotl usually loads the entire dataset into memory. This will be challenging for large datasets. Use the following config to enable streaming:
-
-```{.yaml filename="config.yaml"}
-pretraining_dataset:
-  - name:
-    path:
-    split:
-    text_column: # column in dataset with the data, usually `text`
-    type: pretrain
-    trust_remote_code:
-    skip: # number of rows of data to skip over from the beginning
-...
-```
-
-:::

docs/dataset-formats/stepwise_supervised.qmd

@@ -1,26 +0,0 @@
----
-title: Stepwise Supervised Format
-description: Format for datasets with stepwise completions and labels
-order: 3
----
-
-## Stepwise Supervised
-
-The stepwise supervised format is designed for chain-of-thought (COT) reasoning
-datasets where each example contains multiple completion steps and a preference label
-for each step.
-
-### Example
-
-Here's a simple example of a stepwise supervised dataset entry:
-
-```json
-{
-  "prompt": "Which number is larger, 9.8 or 9.11?",
-  "completions": [
-    "The fractional part of 9.8 is 0.8, while the fractional part of 9.11 is 0.11.",
-    "Since 0.11 is greater than 0.8, the number 9.11 is larger than 9.8."
-  ],
-  "labels": [true, false]
-}
-```

docs/dataset-formats/template_free.qmd

@@ -1,7 +0,0 @@
----
-title: Template-Free
-description: Construct prompts without a template.
-order: 4
----
-
-See [these docs](../input_output.qmd).

docs/dataset-formats/tokenized.qmd

@@ -1,28 +0,0 @@
----
-title: Custom Pre-Tokenized Dataset
-description: How to use a custom pre-tokenized dataset.
-order: 5
----
-
-- Pass an empty `type:` in your axolotl config.
-- Columns in Dataset must be exactly `input_ids`, `attention_mask`, `labels`
-- To indicate that a token should be ignored during training, set its corresponding label to `-100`.
-- You must add BOS and EOS, and make sure that you are training on EOS by not setting its label to -100.
-- For pretraining, do not truncate/pad documents to the context window length.
-- For instruction training, documents must be truncated/padded as desired.
-
-Sample config:
-
-```{.yaml filename="config.yml"}
-datasets:
-  - path: /path/to/your/file.jsonl
-    ds_type: json
-    type:
-```
-
-Sample jsonl:
-
-```jsonl
-{"input_ids":[271,299,99],"attention_mask":[1,1,1],"labels":[271,-100,99]}
-{"input_ids":[87,227,8383,12],"attention_mask":[1,1,1,1],"labels":[87,227,8383,12]}
-```

docs/dataset_preprocessing.qmd

@@ -1,35 +0,0 @@
----
-title: Dataset Preprocessing
-description: How datasets are processed
----
-
-Dataset pre-processing is the step where Axolotl takes each dataset you've configured alongside
-the (dataset format)[../dataset-formats/] and prompt strategies to:
- - parse the dataset based on the *dataset format*
- - transform the dataset to how you would interact with the model based on the *prompt strategy*
- - tokenize the dataset based on the configured model & tokenizer
- - shuffle and merge multiple datasets together if using more than one
-
-The processing of the datasets can happen one of two ways:
-
-1. Before kicking off training by calling `python -m axolotl.cli.preprocess /path/to/your.yaml --debug`
-2. When training is started
-
-What are the benefits of pre-processing? When training interactively or for sweeps
-(e.g. you are restarting the trainer often), processing the datasets can oftentimes be frustratingly
-slow. Pre-processing will cache the tokenized/formatted datasets according to a hash of dependent
-training parameters so that it will intelligently pull from its cache when possible.
-
-The path of the cache is controlled by `dataset_prepared_path:` and is often left blank in example
-YAMLs as this leads to a more robust solution that prevents unexpectedly reusing cached data.
-
-If `dataset_prepared_path:` is left empty, when training, the processed dataset will be cached in a
-default path of `./last_run_prepared/`, but will ignore anything already cached there. By explicitly
-setting `dataset_prepared_path: ./last_run_prepared`, the trainer will use whatever pre-processed
-data is in the cache.
-
-What are the edge cases? Let's say you are writing a custom prompt strategy or using a user-defined
-prompt template. Because the trainer cannot readily detect these changes, we cannot change the
-calculated hash value for the pre-processed dataset. If you have `dataset_prepared_path: ...` set
-and change your prompt templating logic, it may not pick up the changes you made and you will be
-training over the old prompt.

docs/debugging.qmd

@@ -1,245 +0,0 @@
----
-title: Debugging
-description: How to debug Axolotl
----
-
-
-This document provides some tips and tricks for debugging Axolotl.  It also provides an example configuration for debugging with VSCode.  A good debugging setup is essential to understanding how Axolotl code works behind the scenes.
-
-## Table of Contents
-
-- [General Tips](#general-tips)
-- [Debugging with VSCode](#debugging-with-vscode)
-    - [Background](#background)
-    - [Configuration](#configuration)
-    - [Customizing your debugger](#customizing-your-debugger)
-    - [Video Tutorial](#video-tutorial)
-- [Debugging With Docker](#debugging-with-docker)
-    - [Setup](#setup)
-    - [Attach To Container](#attach-to-container)
-    - [Video - Attaching To Docker On Remote Host](#video---attaching-to-docker-on-remote-host)
-
-## General Tips
-
-While debugging it's helpful to simplify your test scenario as much as possible.  Here are some tips for doing so:
-
-> [!Important]
-> All of these tips are incorporated into the [example configuration](#configuration) for debugging with VSCode below.
-
-1. **Make sure you are using the latest version of axolotl**:  This project changes often and bugs get fixed fast.  Check your git branch and make sure you have pulled the latest changes from `main`.
-1. **Eliminate concurrency**: Restrict the number of processes to 1 for both training and data preprocessing:
-    - Set `CUDA_VISIBLE_DEVICES` to a single GPU, ex: `export CUDA_VISIBLE_DEVICES=0`.
-    - Set `dataset_processes: 1` in your axolotl config or run the training command with `--dataset_processes=1`.
-2. **Use a small dataset**: Construct or use a small dataset from HF Hub. When using a small dataset, you will often have to make sure `sample_packing: False` and `eval_sample_packing: False` to avoid errors.  If you are in a pinch and don't have time to construct a small dataset but want to use from the HF Hub, you can shard the data (this will still tokenize the entire dataset, but will only use a fraction of the data for training.  For example, to shard the dataset into 20 pieces, add the following to your axolotl config):
-    ```yaml
-    dataset:
-        ...
-        shards: 20
-    ```
-3. **Use a small model**: A good example of a small model is [TinyLlama/TinyLlama-1.1B-Chat-v1.0](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0).
-4. **Minimize iteration time**: Make sure the training loop finishes as fast as possible, with these settings.
-    - `micro_batch_size: 1`
-    - `max_steps: 1`
-    - `val_set_size: 0`
-5. **Clear Caches:** Axolotl caches certain steps and so does the underlying HuggingFace trainer.  You may want to clear some of these caches when debugging.
-    - Data preprocessing: When debugging data preprocessing, which includes prompt template formation, you may want to delete the directory set in `dataset_prepared_path:` in your axolotl config.  If you didn't set this value, the default is `last_run_prepared`.
-    - HF Hub: If you are debugging data preprocessing, you should clear the relevant HF cache [HuggingFace cache](https://huggingface.co/docs/datasets/cache), by deleting the appropriate `~/.cache/huggingface/datasets/...` folder(s).
-    - **The recommended approach is to redirect all outputs and caches to a temporary folder and delete selected subfolders before each run.  This is demonstrated in the example configuration below.**
-
-
-## Debugging with VSCode
-
-### Background
-
-The below example shows how to configure VSCode to debug data preprocessing of the `chat_template` format.  This is the format used when you have the following in your axolotl config:
-
-```yaml
-datasets:
-  - path: <path to your chat_template formatted dataset> # example on HF Hub: fozziethebeat/alpaca_messages_2k_test
-    type: chat_template
-```
-
->[!Important]
-> If you are already familiar with advanced VSCode debugging, you can skip the below explanation and look at the files [.vscode/launch.json](../.vscode/launch.json) and [.vscode/tasks.json](../.vscode/tasks.json) for an example configuration.
-
->[!Tip]
-> If you prefer to watch a video, rather than read, you can skip to the [video tutorial](#video-tutorial) below (but doing both is recommended).
-
-### Setup
-
-Make sure you have an [editable install](https://setuptools.pypa.io/en/latest/userguide/development_mode.html) of Axolotl, which ensures that changes you make to the code are reflected at runtime.  Run the following commands from the root of this project:
-
-```bash
-pip3 install packaging
-pip3 install --no-build-isolation -e '.[flash-attn,deepspeed]'
-```
-
-#### Remote Hosts
-
-If you developing on a remote host, you can easily use VSCode to debug remotely.  To do so, you will need to follow this [remote - SSH guide](https://code.visualstudio.com/docs/remote/ssh).  You can also see the video below on [Docker and Remote SSH debugging](#video---attaching-to-docker-on-remote-host).
-
-
-### Configuration
-
-The easiest way to get started is to modify the [.vscode/launch.json](../.vscode/launch.json) file in this project.  This is just an example configuration, so you may need to modify or copy it to suit your needs.
-
-For example, to mimic the command `cd devtools && CUDA_VISIBLE_DEVICES=0 accelerate launch -m axolotl.cli.train dev_chat_template.yml`, you would use the below configuration[^1].  Note that we add additional flags that override the axolotl config and incorporate the tips above (see the comments). We also set the working directory to `devtools` and set the `env` variable `HF_HOME` to a temporary folder that is later partially deleted.  This is because we want to delete the HF dataset cache before each run in order to ensure that the data preprocessing code is run from scratch.
-
-```jsonc
-// .vscode/launch.json
-{
-    "version": "0.2.0",
-    "configurations": [
-        {
-            "name": "Debug axolotl prompt - chat_template",
-            "type": "python",
-            "module": "accelerate.commands.launch",
-            "request": "launch",
-            "args": [
-                "-m", "axolotl.cli.train", "dev_chat_template.yml",
-                // The flags below simplify debugging by overriding the axolotl config
-                // with the debugging tips above.  Modify as needed.
-                "--dataset_processes=1",      // limits data preprocessing to one process
-                "--max_steps=1",              // limits training to just one step
-                "--batch_size=1",             // minimizes batch size
-                "--micro_batch_size=1",       // minimizes batch size
-                "--val_set_size=0",           // disables validation
-                "--sample_packing=False",     // disables sample packing which is necessary for small datasets
-                "--eval_sample_packing=False",// disables sample packing on eval set
-                "--dataset_prepared_path=temp_debug/axolotl_outputs/data", // send data outputs to a temp folder
-                "--output_dir=temp_debug/axolotl_outputs/model" // send model outputs to a temp folder
-                ],
-            "console": "integratedTerminal",      // show output in the integrated terminal
-            "cwd": "${workspaceFolder}/devtools", // set working directory to devtools from the root of the project
-            "justMyCode": true,                   // step through only axolotl code
-            "env": {"CUDA_VISIBLE_DEVICES": "0",  // Since we aren't doing distributed training, we need to limit to one GPU
-                    "HF_HOME": "${workspaceFolder}/devtools/temp_debug/.hf-cache"}, // send HF cache to a temp folder
-            "preLaunchTask": "cleanup-for-dataprep", // delete temp folders (see below)
-        }
-    ]
-}
-```
-
-**Additional notes about this configuration:**
-
-- The argument `justMyCode` is set to `true` such that you step through only the axolotl code.  If you want to step into dependencies, set this to `false`.
-- The `preLaunchTask`: `cleanup-for-dataprep` is defined in [.vscode/tasks.json](../.vscode/tasks.json) and is used to delete the following folders before debugging, which is essential to ensure that the data pre-processing code is run from scratch:
-    -  `./devtools/temp_debug/axolotl_outputs`
-    - `./devtools/temp_debug/.hf-cache/datasets`
-
->[!Tip]
-> You may not want to delete these folders. For example, if you are debugging model training instead of data pre-processing, you may NOT want to delete the cache or output folders. You may also need to add additional tasks to the `tasks.json` file depending on your use case.
-
-Below is the [./vscode/tasks.json](../.vscode/tasks.json) file that defines the `cleanup-for-dataprep` task.  This task is run before each debugging session when you use the above configuration.  Note how there are two tasks that delete the two folders mentioned above.  The third task `cleanup-for-dataprep` is a composite task that combines the two tasks.  A composite task is necessary because VSCode does not allow you to specify multiple tasks in the `preLaunchTask` argument of the `launch.json` file.
-
-```jsonc
-// .vscode/tasks.json
-// this file is used by launch.json
-{
-    "version": "2.0.0",
-    "tasks": [
-      // this task changes into the devtools directory and deletes the temp_debug/axolotl_outputs folder
-      {
-        "label": "delete-outputs",
-        "type": "shell",
-        "command": "rm -rf temp_debug/axolotl_outputs",
-        "options":{ "cwd": "${workspaceFolder}/devtools"},
-        "problemMatcher": []
-      },
-      // this task changes into the devtools directory and deletes the `temp_debug/.hf-cache/datasets` folder
-      {
-        "label": "delete-temp-hf-dataset-cache",
-        "type": "shell",
-        "command": "rm -rf temp_debug/.hf-cache/datasets",
-        "options":{ "cwd": "${workspaceFolder}/devtools"},
-        "problemMatcher": []
-      },
-        // this task combines the two tasks above
-      {
-       "label": "cleanup-for-dataprep",
-       "dependsOn": ["delete-outputs", "delete-temp-hf-dataset-cache"],
-      }
-    ]
-}
-```
-
-### Customizing your debugger
-
-Your debugging use case may differ from the example above.  The easiest thing to do is to put your own axolotl config in the `devtools` folder and modify the `launch.json` file to use your config.  You may also want to modify the `preLaunchTask` to delete different folders or not delete anything at all.
-
-### Video Tutorial
-
-The following video tutorial walks through the above configuration and demonstrates how to debug with VSCode, (click the image below to watch):
-
-<div style="text-align: center; line-height: 0;">
-
-<a href="https://youtu.be/xUUB11yeMmc" target="_blank"
-title="How to debug Axolotl (for fine tuning LLMs)"><img
-src="https://i.ytimg.com/vi/xUUB11yeMmc/maxresdefault.jpg"
-style="border-radius: 10px; display: block; margin: auto;" width="560" height="315" /></a>
-
-<figcaption style="font-size: smaller;"><a href="https://hamel.dev">Hamel Husain's</a> tutorial: <a href="https://www.youtube.com/watch?v=xUUB11yeMmc">Debugging Axolotl w/VSCode</a></figcaption>
-
-</div>
-<br>
-
-## Debugging With Docker
-
-Using [official Axolotl Docker images](https://hub.docker.com/r/axolotlai/axolotl/tags) is a great way to debug your code, and is a very popular way to use Axolotl.  Attaching VSCode to Docker takes a few more steps.
-
-### Setup
-
-On the host that is running axolotl (ex: if you are using a remote host), clone the axolotl repo and change your current directory to the root:
-
-```bash
-git clone https://github.com/axolotl-ai-cloud/axolotl
-cd axolotl
-```
-
->[!Tip]
-> If you already have axolotl cloned on your host, make sure you have the latest changes and change into the root of the project.
-
-Next, run the desired docker image and mount the current directory. Below is a docker command you can run to do this:[^2]
-
-```bash
-docker run --privileged --gpus '"all"' --shm-size 10g --rm -it --name axolotl --ipc=host --ulimit memlock=-1 --ulimit stack=67108864 --mount type=bind,src="${PWD}",target=/workspace/axolotl -v ${HOME}/.cache/huggingface:/root/.cache/huggingface axolotlai/axolotl:main-py3.10-cu118-2.0.1
-```
-
->[!Tip]
-> To understand which containers are available, see the [Docker section of the README](../README.md#docker) and the [DockerHub repo](https://hub.docker.com/r/axolotlai/axolotl/tags).  For details of how the Docker containers are built, see axolotl's [Docker CI builds](../.github/workflows/main.yml).
-
-You will now be in the container.  Next, perform an editable install of Axolotl:
-
-```bash
-pip3 install packaging
-pip3 install --no-build-isolation -e '.[flash-attn,deepspeed]'
-```
-
-### Attach To Container
-
-Next, if you are using a remote host, [Remote into this host with VSCode](https://code.visualstudio.com/docs/remote/ssh).  If you are using a local host, you can skip this step.
-
-Next, select `Dev Containers: Attach to Running Container...` using the command palette (`CMD + SHIFT + P`) in VSCode.  You will be prompted to select a container to attach to.  Select the container you just created.  You will now be in the container with a working directory that is at the root of the project.  Any changes you make to the code will be reflected both in the container and on the host.
-
-Now you are ready to debug as described above (see [Debugging with VSCode](#debugging-with-vscode)).
-
-### Video - Attaching To Docker On Remote Host
-
-Here is a short video that demonstrates how to attach to a Docker container on a remote host:
-
-<div style="text-align: center; line-height: 0;">
-
-<a href="https://youtu.be/0AuoR7QnHR0" target="_blank"
-title="Debugging Axolotl Part 2: Attaching to Docker on a Remote Host"><img
-src="https://i.ytimg.com/vi/0AuoR7QnHR0/hqdefault.jpg"
-style="border-radius: 10px; display: block; margin: auto;" width="560" height="315" /></a>
-
-<figcaption style="font-size: smaller;"><a href="https://hamel.dev">Hamel Husain's</a> tutorial: <a href="https://youtu.be/0AuoR7QnHR0">Debugging Axolotl Part 2: Attaching to Docker on a Remote Host
-</a></figcaption>
-
-</div>
-<br>
-
-[^1]: The config actually mimics the command `CUDA_VISIBLE_DEVICES=0 python -m accelerate.commands.launch -m axolotl.cli.train devtools/chat_template.yml`, but this is the same thing.
-
-[^2]: Many of the below flags are recommended best practices by Nvidia when using nvidia-container-toolkit.  You can read more about these flags [here](https://docs.nvidia.com/deeplearning/frameworks/user-guide/index.html).

docs/faq.qmd

@@ -1,29 +0,0 @@
----
-title: FAQ
-description: Frequently asked questions
----
-
-
-**Q: The trainer stopped and hasn't progressed in several minutes.**
-
-> A: Usually an issue with the GPUs communicating with each other. See the [NCCL doc](nccl.qmd)
-
-**Q: Exitcode -9**
-
-> A: This usually happens when you run out of system RAM.
-
-**Q: Exitcode -7 while using deepspeed**
-
-> A: Try upgrading deepspeed w: `pip install -U deepspeed`
-
-**Q: AttributeError: 'DummyOptim' object has no attribute 'step'**
-
-> A: You may be using deepspeed with single gpu. Please don't set `deepspeed:` in yaml or cli.
-
-**Q: The codes is stuck on saving preprocessed datasets.**
-
-> A: This is usually an issue with the GPU. This can be resolved through setting the os environment variable `CUDA_VISIBLE_DEVICES=0`. If you are on runpod, this is usually a pod issue. Starting a new pod should take care of it.
-
-**Q: `jinja2.exceptions.UndefinedError: 'dict object' has no attribute 'content' / 'role' / ____`**
-
-> A: This means that the property mapping for the stated attribute does not exist when building `chat_template` prompt. For example, if `no attribute 'content'`, please check you have added the correct mapping for `content` under `message_property_mappings`.

docs/fsdp_qlora.qmd

@@ -1,43 +0,0 @@
----
-title: "FDSP + QLoRA"
-description: Use FSDP with QLoRA to fine-tune large LLMs on consumer GPUs.
-format:
-  html:
-    toc: true
----
-
-## Background
-
-Using FSDP with QLoRA is essential for **fine-tuning larger (70b+ parameter) LLMs on consumer GPUs.**  For example, you can use FSDP + QLoRA to train a 70b model on two 24GB GPUs[^1].
-
-Below, we describe how to use this feature in Axolotl.
-
-## Usage
-
-To enable `QLoRA` with `FSDP`, you need to perform the following steps:
-
-> ![Tip]
-> See the [example config](#example-config) file in addition to reading these instructions.
-
-1. Set `adapter: qlora` in your axolotl config file.
-2. Enable FSDP in your axolotl config, as [described here](https://github.com/axolotl-ai-cloud/axolotl?tab=readme-ov-file#fsdp).
-3. Use one of the supported model types: `llama`, `mistral` or `mixtral`.
-
-## Example Config
-
-[examples/llama-2/qlora-fsdp.yml](../examples/llama-2/qlora-fsdp.yml) contains an example of how to enable QLoRA + FSDP in axolotl.
-
-## References
-
-- [PR #1378](https://github.com/axolotl-ai-cloud/axolotl/pull/1378) enabling QLoRA in FSDP in Axolotl.
-- [Blog Post](https://www.answer.ai/posts/2024-03-06-fsdp-qlora.html) from the [Answer.AI](https://www.answer.ai/) team describing the work that enabled QLoRA in FSDP.
-- Related HuggingFace PRs Enabling FDSP + QLoRA:
-    - Accelerate [PR#2544](https://github.com/huggingface/accelerate/pull/2544 )
-    - Transformers [PR#29587](https://github.com/huggingface/transformers/pull/29587)
-    - TRL [PR#1416](https://github.com/huggingface/trl/pull/1416)
-    - PEFT [PR#1550](https://github.com/huggingface/peft/pull/1550)
-
-
-
-
-[^1]: This was enabled by [this work](https://www.answer.ai/posts/2024-03-06-fsdp-qlora.html) from the Answer.AI team.

docs/getting-started.qmd

@@ -1,155 +0,0 @@
----
-title: "Getting Started with Axolotl"
-format:
-  html:
-    toc: true
-    toc-depth: 3
-    number-sections: true
-execute:
-  enabled: false
----
-
-This guide will walk you through your first model fine-tuning project with Axolotl.
-
-## Quick Example {#sec-quick-example}
-
-Let's start by fine-tuning a small language model using LoRA. This example uses a 1B parameter model to ensure it runs on most GPUs.
-Assuming `axolotl` is installed (if not, see our [Installation Guide](installation.qmd))
-
-1. Download example configs:
-```shell
-axolotl fetch examples
-```
-
-2. Run the training:
-```shell
-axolotl train examples/llama-3/lora-1b.yml
-```
-
-That's it! Let's understand what just happened.
-
-## Understanding the Process {#sec-understanding}
-
-### The Configuration File {#sec-config}
-
-The YAML configuration file controls everything about your training. Here's what (part of) our example config looks like:
-
-```yaml
-base_model: NousResearch/Llama-3.2-1B
-# hub_model_id: username/custom_model_name
-
-datasets:
-  - path: teknium/GPT4-LLM-Cleaned
-    type: alpaca
-dataset_prepared_path: last_run_prepared
-val_set_size: 0.1
-output_dir: ./outputs/lora-out
-
-adapter: lora
-lora_model_dir:
-```
-
-See our [Config options](config.qmd) for more details.
-
-### Training {#sec-training}
-
-When you run `axolotl train`, Axolotl:
-
-1. Downloads the base model
-2. (If specified) applies LoRA adapter layers
-3. Loads and processes the dataset
-4. Runs the training loop
-5. Saves the trained model and / or LoRA weights
-
-## Your First Custom Training {#sec-custom}
-
-Let's modify the example for your own data:
-
-1. Create a new config file `my_training.yml`:
-
-```yaml
-base_model: NousResearch/Nous-Hermes-llama-1b-v1
-adapter: lora
-
-# Training settings
-micro_batch_size: 2
-num_epochs: 3
-learning_rate: 0.0003
-
-# Your dataset
-datasets:
-  - path: my_data.jsonl        # Your local data file
-    type: alpaca               # Or other format
-```
-
-This specific config is for LoRA fine-tuning a model with instruction tuning data using
-the `alpaca` dataset format, which has the following format:
-
-```json
-{
-    "instruction": "Write a description of alpacas.",
-    "input": "",
-    "output": "Alpacas are domesticated South American camelids..."
-}
-```
-
-Please see our [Dataset Formats](dataset-formats) for more dataset formats and how to
-format them.
-
-2. Prepare your JSONL data in the specified format (in this case, the expected `alpaca
-format):
-
-```json
-{"instruction": "Classify this text", "input": "I love this!", "output": "positive"}
-{"instruction": "Classify this text", "input": "Not good at all", "output": "negative"}
-```
-
-Please consult the supported [Dataset Formats](dataset-formats/) for more details.
-
-3. Run the training:
-
-```shell
-axolotl train my_training.yml
-```
-
-## Common Tasks {#sec-common-tasks}
-
-### Testing Your Model {#sec-testing}
-
-After training, test your model:
-
-```shell
-axolotl inference my_training.yml --lora-model-dir="./outputs/lora-out"
-```
-
-### Preprocessing Data {#sec-preprocessing}
-
-For large datasets, preprocess first:
-
-```shell
-axolotl preprocess my_training.yml
-```
-
-### Using a UI {#sec-ui}
-
-Launch a Gradio interface:
-
-```shell
-axolotl inference my_training.yml --lora-model-dir="./outputs/lora-out" --gradio
-```
-
-## Next Steps {#sec-next-steps}
-
-Now that you have the basics, you might want to:
-
-- Try different model architectures
-- Experiment with hyperparameters
-- Use more advanced training methods
-- Scale up to larger models
-
-Check our other guides for details on these topics:
-
-- [Configuration Guide](config.qmd) - Full configuration options
-- [Dataset Formats](dataset-formats) - Working with different data formats
-- [Multi-GPU Training](multi-gpu.qmd)
-- [Multi-Node Training](multi-node.qmd)

docs/inference.qmd

@@ -1,148 +0,0 @@
----
-title: "Inference Guide"
-format:
-  html:
-    toc: true
-    toc-depth: 3
-    number-sections: true
-    code-tools: true
-execute:
-  enabled: false
----
-
-This guide covers how to use your trained models for inference, including model loading, interactive testing, and common troubleshooting steps.
-
-## Quick Start {#sec-quickstart}
-
-### Basic Inference {#sec-basic}
-
-::: {.panel-tabset}
-
-## LoRA Models
-
-```{.bash}
-axolotl inference your_config.yml --lora-model-dir="./lora-output-dir"
-```
-
-## Full Fine-tuned Models
-
-```{.bash}
-axolotl inference your_config.yml --base-model="./completed-model"
-```
-
-:::
-
-## Advanced Usage {#sec-advanced}
-
-### Gradio Interface {#sec-gradio}
-
-Launch an interactive web interface:
-
-```{.bash}
-axolotl inference your_config.yml --gradio
-```
-
-### File-based Prompts {#sec-file-prompts}
-
-Process prompts from a text file:
-
-```{.bash}
-cat /tmp/prompt.txt | axolotl inference your_config.yml \
-  --base-model="./completed-model" --prompter=None
-```
-
-### Memory Optimization {#sec-memory}
-
-For large models or limited memory:
-
-```{.bash}
-axolotl inference your_config.yml --load-in-8bit=True
-```
-
-## Merging LoRA Weights {#sec-merging}
-
-Merge LoRA adapters with the base model:
-
-```{.bash}
-axolotl merge-lora your_config.yml --lora-model-dir="./completed-model"
-```
-
-### Memory Management for Merging {#sec-memory-management}
-
-::: {.panel-tabset}
-
-## Configuration Options
-
-```{.yaml}
-gpu_memory_limit: 20GiB  # Adjust based on your GPU
-lora_on_cpu: true        # Process on CPU if needed
-```
-
-## Force CPU Merging
-
-```{.bash}
-CUDA_VISIBLE_DEVICES="" axolotl merge-lora ...
-```
-
-:::
-
-## Tokenization {#sec-tokenization}
-
-### Common Issues {#sec-tokenization-issues}
-
-::: {.callout-warning}
-Tokenization mismatches between training and inference are a common source of problems.
-:::
-
-To debug:
-
-1. Check training tokenization:
-```{.bash}
-axolotl preprocess your_config.yml --debug
-```
-
-2. Verify inference tokenization by decoding tokens before model input
-
-3. Compare token IDs between training and inference
-
-### Special Tokens {#sec-special-tokens}
-
-Configure special tokens in your YAML:
-
-```{.yaml}
-special_tokens:
-  bos_token: "<s>"
-  eos_token: "</s>"
-  unk_token: "<unk>"
-tokens:
-  - "<|im_start|>"
-  - "<|im_end|>"
-```
-
-## Troubleshooting {#sec-troubleshooting}
-
-### Common Problems {#sec-common-problems}
-
-::: {.panel-tabset}
-
-## Memory Issues
-
-- Use 8-bit loading
-- Reduce batch sizes
-- Try CPU offloading
-
-## Token Issues
-
-- Verify special tokens
-- Check tokenizer settings
-- Compare training and inference preprocessing
-
-## Performance Issues
-
-- Verify model loading
-- Check prompt formatting
-- Ensure temperature/sampling settings
-
-:::
-
-For more details, see our [debugging guide](debugging.qmd).

docs/input_output.qmd

@@ -1,265 +0,0 @@
----
-title: Template-free prompt construction
-description: "Template-free prompt construction with the `input_output` format"
----
-
-<!-- TOC -->
-
-- [Background](#background)
-    - [Masking Inputs](#masking-inputs)
-    - [You may not want prompt templates](#you-may-not-want-prompt-templates)
-    - [The `input_output` format](#the-input_output-format)
-- [Usage](#usage)
-    - [1. Prepare Data](#1-prepare-data)
-    - [2. Use `type: input_output`](#2-use-type-input_output)
-    - [3. Check the prompts](#3-check-the-prompts)
-
-<!-- /TOC -->
-
-<a id="markdown-background" name="background"></a>
-
-## Background
-
-<a id="markdown-masking-inputs" name="masking-inputs"></a>
-
-### Masking Inputs
-
-One of the most popular features of
-[axolotl](https://github.com/axolotl-ai-cloud/axolotl) is
-setting the following configuration value:
-
-
-```yaml
-train_on_inputs: false
-```
-
-If you declare a [dataset formats](https://github.com/axolotl-ai-cloud/axolotl?tab=readme-ov-file#dataset)
-such as `alpaca` or `chatml`, axolotl knows what is an input
-(i.e. human) vs. an output (i.e. the assistant) and masks the input
-labels so that your model can focus on predicting the outputs only.
-
-<a id="markdown-you-may-not-want-prompt-templates" name="you-may-not-want-prompt-templates"></a>
-
-### You may not want prompt templates
-
-However, there are many situations where you don't want to use one of
-these formats or templates. This is because they can:
-
--   Add unnecessary boilerplate to your prompts.
--   Create artifacts like special delimiters `<|im_start|>` that can
-    quickly become footguns if you don't include them correctly at
-    inference time.
--   Enforce a *chat* interface when you do not want one. Sometimes you
-    just want to fine-tune a model to a very specific task and do NOT
-    want multi-turn conversations, roles, etc.
--   Limit you to only certain roles that the template allows.
-
-<a id="markdown-the-inputoutput-format" name="the-inputoutput-format"></a>
-
-### The `input_output` format
-
-You can construct your prompts without a template by using the
-`input_output` format, by setting `type: input_output` in your
-configuration file like this:
-
-**config.yml**
-
-```yaml
-train_on_inputs: false # Mask segments of your data
-datasets:
-  - path: output.jsonl
-    type: input_output  # use template free prompt construction
-```
-
-Unlike `type: completion`, which is also template-free,
-`type: input_output` allows you to mask segments of your text. More
-details on how this works are described below.
-
-<a id="markdown-usage" name="usage"></a>
-
-## Usage
-
-This is how you can use the `input_output` format:
-
-<a id="markdown-1-prepare-data" name="1-prepare-data"></a>
-
-### 1. Prepare Data
-
-To use the `input_output` format, collect your data in the following
-format into a jsonl file (below is the first row from the file
-`output`.jsonl` pretty printed):
-
-```bash
-$ head -n1 output.jsonl | python -m json.tool
-```
-
-:::{.cell-output .cell-output-stdout}
-    {
-        "segments": [
-            {
-                "label": true,
-                "text": "<s>Hello\n"
-            },
-            {
-                "label": true,
-                "text": "hi there!. "
-            },
-            {
-                "label": false,
-                "text": "goodbye "
-            },
-            {
-                "label": true,
-                "text": "farewell</s>"
-            }
-        ]
-    }
-:::
-
-Set `label:false` when you want to mask a segment of text so that the
-model isn't trained on it. Some things to keep in mind:
-
-> [!IMPORTANT]
-> 1.  **EOS, BOS, spaces, newlines etc. are entirely up to you. Axolotl
-    concatenates all the segments as-is.** The tokenizer doesn't add
-    anything additional. Notice how I added spaces, newlines, `<s>`
-    (BOS), and `</s>` (EOS) myself.
-> 2.  Make sure you check the materialized output to validate that the
-    prompt is getting assembled how you like.
-
-<a id="markdown-2-use-type-inputoutput" name="2-use-type-inputoutput"></a>
-
-### 2. Use `type: input_output`
-
-Let's materialize data with our `output.jsonl` file by setting
-`type: input_output` in our axolotl config:
-
-```yaml
-# training_config.yaml
-base_model: mistralai/Mistral-7B-v0.1
-data_seed: 49
-seed: 49
-
-datasets:
-  - path: output.jsonl
-    type: input_output
-val_set_size: 0.1
-
-sequence_len: 896
-sample_packing: false
-
-micro_batch_size: 2
-gradient_accumulation_steps: 3
-eval_batch_size: 2
-num_epochs: 1
-learning_rate: 0.0002
-
-train_on_inputs: false
-special_tokens:
-  bos_token: "<s>"
-  eos_token: "</s>"
-  unk_token: "<unk>"
-```
-
-You can use the following command to materialize your data. The
-`--debug` flag will print the tokens, along with the labels so you can
-verify that the correct items are being ignored:
-
-```bash
-$ python -m axolotl.cli.preprocess training_config.yaml --debug
-
-...
-[2024-03-05 23:36:46,969] [INFO] [axolotl.check_example_labels:35] [PID:607731] [RANK:0] <s>(1, 1) Hello(22557, 22557)
-(13, 13) hi(12014, 12014) there(736, 736) !(28808, 28808) .(28723, 28723) (28705, 28705) good(-100, 1179) bye(-100, 17664) (-100, 28705) fare(19111, 19111) well(5458, 5458) </s>(2, 2)
-
-```
-
-The format is `decoded_token`(`label`, `token_id`), for example,
-`<s>(1, 1)` means that the token is `<s>`, the label is `1` and the
-token_id is `1`. When the label is `-100` then that token is ignored for
-training.
-
-<a id="markdown-3-check-the-prompts" name="3-check-the-prompts"></a>
-
-### 3. Check the prompts
-
-Here is another way to check the materialized output:
-
-```python
-from transformers import AutoTokenizer
-from datasets import load_from_disk
-import yaml
-
-directory = !ls last_run_prepared/
-with open('training_config.yaml', 'r') as f:
-    cfg = yaml.safe_load(f)
-model_id = cfg['base_model']
-tok = AutoTokenizer.from_pretrained(model_id)
-ds = load_from_disk(f'last_run_prepared/{directory[0]}/')
-```
-
-```python
->>> row = ds[0]
->>> print(tok.decode(row['input_ids']))
-<s> Hello
-    hi there!.  goodbye  farewell</s>
-```
-
-We can check that the right tokens are ignored by comparing the labels
-to each token:
-
-```python
-import pandas as pd
-pd.DataFrame([{'token': tok.decode(i), 'label': l, 'id':i} for i,l in
-              zip(row['input_ids'], row['labels'])])
-```
-
-| token | label | id    |
-|-------|-------|-------|
-| 0     | \<s\> | 1     |
-| 1     | Hello | 22557 |
-| 2     | \\n   | 13    |
-| 3     | hi    | 12014 |
-| 4     | there | 736   |
-| 5     | !     | 28808 |
-| 6     | .     | 28723 |
-| 7     |       | 28705 |
-| 8     | good  | -100  |
-| 9     | bye   | -100  |
-| 10    |       | -100  |
-| 11    | fare  | 19111 |
-| 12    | well  | 5458  |
-| 13    | \</s\>| 2     |
-
-
-
-If we look at the input data, the above table seems correct! (The jsonl
-version is repeated below for reference):
-
-
-```bash
-$ head -n1 output.jsonl | python -m json.tool
-```
-
-:::{.cell-output .cell-output-stdout}
-    {
-        "segments": [
-            {
-                "label": true,
-                "text": "<s>Hello\n"
-            },
-            {
-                "label": true,
-                "text": "hi there!. "
-            },
-            {
-                "label": false,
-                "text": "goodbye "
-            },
-            {
-                "label": true,
-                "text": "farewell</s>"
-            }
-        ]
-    }
-:::

docs/installation.qmd

@@ -1,119 +0,0 @@
----
-title: "Installation Guide"
-format:
-  html:
-    toc: true
-    toc-depth: 3
-    number-sections: true
-    code-tools: true
-execute:
-  enabled: false
----
-
-This guide covers all the ways you can install and set up Axolotl for your environment.
-
-## Requirements {#sec-requirements}
-
-- NVIDIA GPU (Ampere architecture or newer for `bf16` and Flash Attention) or AMD GPU
-- Python ≥3.10
-- PyTorch ≥2.4.1
-
-## Installation Methods {#sec-installation-methods}
-
-### PyPI Installation (Recommended) {#sec-pypi}
-
-```{.bash}
-pip3 install --no-build-isolation axolotl[flash-attn,deepspeed]
-```
-
-We use `--no-build-isolation` in order to detect the installed PyTorch version (if
-installed) in order not to clobber it, and so that we set the correct version of
-dependencies that are specific to the PyTorch version or other installed
-co-dependencies.
-
-### Edge/Development Build {#sec-edge-build}
-
-For the latest features between releases:
-
-```{.bash}
-git clone https://github.com/axolotl-ai-cloud/axolotl.git
-cd axolotl
-pip3 install packaging ninja
-pip3 install --no-build-isolation -e '.[flash-attn,deepspeed]'
-```
-
-### Docker {#sec-docker}
-
-```{.bash}
-docker run --gpus '"all"' --rm -it axolotlai/axolotl:main-latest
-```
-
-For development with Docker:
-
-```{.bash}
-docker compose up -d
-```
-
-::: {.callout-tip}
-### Advanced Docker Configuration
-```{.bash}
-docker run --privileged --gpus '"all"' --shm-size 10g --rm -it \
-  --name axolotl --ipc=host \
-  --ulimit memlock=-1 --ulimit stack=67108864 \
-  --mount type=bind,src="${PWD}",target=/workspace/axolotl \
-  -v ${HOME}/.cache/huggingface:/root/.cache/huggingface \
-  axolotlai/axolotl:main-latest
-```
-:::
-
-## Cloud Environments {#sec-cloud}
-
-### Cloud GPU Providers {#sec-cloud-gpu}
-
-For providers supporting Docker:
-
-- Use `axolotlai/axolotl-cloud:main-latest`
-- Available on:
-  - [Latitude.sh](https://latitude.sh/blueprint/989e0e79-3bf6-41ea-a46b-1f246e309d5c)
-  - [JarvisLabs.ai](https://jarvislabs.ai/templates/axolotl)
-  - [RunPod](https://runpod.io/gsc?template=v2ickqhz9s&ref=6i7fkpdz)
-
-### Google Colab {#sec-colab}
-
-Use our [example notebook](../examples/colab-notebooks/colab-axolotl-example.ipynb).
-
-## Platform-Specific Instructions {#sec-platform-specific}
-
-### macOS {#sec-macos}
-
-```{.bash}
-pip3 install --no-build-isolation -e '.'
-```
-
-See @sec-troubleshooting for Mac-specific issues.
-
-### Windows {#sec-windows}
-
-::: {.callout-important}
-We recommend using WSL2 (Windows Subsystem for Linux) or Docker.
-:::
-
-## Environment Managers {#sec-env-managers}
-
-### Conda/Pip venv {#sec-conda}
-
-1. Install Python ≥3.10
-2. Install PyTorch: https://pytorch.org/get-started/locally/
-3. Install Axolotl:
-   ```{.bash}
-   pip3 install packaging
-   pip3 install --no-build-isolation -e '.[flash-attn,deepspeed]'
-   ```
-4. (Optional) Login to Hugging Face:
-   ```{.bash}
-   huggingface-cli login
-   ```
-
-## Troubleshooting {#sec-troubleshooting}
-
-If you encounter installation issues, see our [FAQ](faq.qmd) and [Debugging Guide](debugging.qmd).

docs/lora_optims.qmd

@@ -1,128 +0,0 @@
----
-title: "LoRA Optimizations"
-description: "Custom autograd functions and Triton kernels in Axolotl for optimized
-LoRA fine-tuning"
----
-
-Inspired by [Unsloth](https://github.com/unslothai/unsloth), we've implemented two
-optimizations for LoRA and QLoRA fine-tuning, supporting both single GPU and multi-GPU
-(in the DDP and DeepSpeed settings) training. These include (1) SwiGLU and GEGLU activation function
-Triton kernels, and (2) LoRA MLP and attention custom autograd functions. Our goal was
-to leverage operator fusion and tensor re-use in order to improve speed and reduce
-memory usage during the forward and backward passes of these calculations.
-
-We currently support several common model architectures, including (but not limited to):
-
-- `llama`
-- `mistral`
-- `qwen2`
-- `gemma`
-- `gemma2`
-
-<details>
-
-The set of models we support is currently limited by our attention patching strategy,
-which assumes (and replaces) specific code blocks for query / key / value and output
-projections:
-
-```python
-ORIGINAL_QKV_CODE = """
-    query_states = self.q_proj(hidden_states).view(hidden_shape).transpose(1, 2)
-    key_states = self.k_proj(hidden_states).view(hidden_shape).transpose(1, 2)
-    value_states = self.v_proj(hidden_states).view(hidden_shape).transpose(1, 2)
-""".lstrip(
-    "\n"
-)
-
-ORIGINAL_O_CODE = """
-    attn_output = self.o_proj(attn_output)
-""".lstrip(
-    "\n"
-)
-```
-
-Is replaced with:
-
-```python
-PATCHED_QKV_CODE = """
-    query_states, key_states, value_states = self.apply_qkv(hidden_states)
-    query_states = query_states.view(hidden_shape).transpose(1, 2)
-    key_states = key_states.view(hidden_shape).transpose(1, 2)
-    value_states = value_states.view(hidden_shape).transpose(1, 2)
-""".lstrip(
-    "\n"
-)
-
-PATCHED_O_CODE = """
-    attn_output = self.apply_o(attn_output)
-""".lstrip(
-    "\n"
-)
-```
-
-Where `apply_qkv` and `apply_o` are defined in the `axolotl.kernels.lora` module.
-
-We welcome testing of other model architectures and / or PRs to expand our patching
-logic to be compatible with more of them.
-
-</details>
-
-## Usage
-
-These optimizations can be enabled in your Axolotl config YAML file. The
-`lora_mlp_kernel` option enables the optimized MLP path, while `lora_qkv_kernel` and
-`lora_o_kernel` enable the fused query-key-value projection and optimized output
-projection, respectively.
-
-```yaml
-lora_mlp_kernel: true
-lora_qkv_kernel: true
-lora_o_kernel: true
-```
-
-## Requirements
-
-- One or more NVIDIA or AMD GPUs (in order to use the Triton kernels)
-    - Note: Set `TORCH_ROCM_AOTRITON_ENABLE_EXPERIMENTAL=1` to enable [memory-efficient attention on AMD GPUs](https://github.com/ROCm/aotriton/issues/16#issuecomment-2346675491)
-- Targeted LoRA adapters cannot use Dropout
-    - This may limit model expressivity / cause overfitting
-- Targeted LoRA adapters cannot have bias terms
-    - This may limit model expressivity
-
-Models with pre-existing LoRA adapters that use Dropout or have bias terms may need to
-be re-finetuned without these features in order to be useful.
-
-## Implementation details
-
-### Custom autograd functions
-
-The LoRA MLP autograd function optimizes the entire MLP computation path. It fuses the
-LoRA and base weight computations together and provides a single, efficient backward
-pass for the entire MLP block.
-
-For attention components, similar optimizations are provided through a function that
-handles the query, key, and value projections, and a function that handles the output
-projection. They are designed to work with the existing `transformers` attention
-implementation via some monkey-patching logic.
-
-### Triton kernels
-
-Two activation functions (SwiGLU and GeGLU) are implemented with Triton kernels for
-improved speed and memory performance. These kernels handle both the forward and
-backward passes.
-
-### Integration
-
-The custom autograd functions and Triton kernels are designed to work together. The
-autograd function manages the high-level computation flow and gradient tracking, while
-calling the Triton kernels for the activation function computation. During the backward
-pass, the kernel computes both the activation output and the required gradients, which
-the autograd function then uses to compute the final gradients for the entire
-computation path.
-
-## Future Work
-
-- Support for additional model architectures
-- Support for the FSDP setting
-- Support for dropout and bias
-- Additional operator fusions

docs/lr_groups.qmd

@@ -1,29 +0,0 @@
----
-title: Learning Rate Groups
-description: "Setting different learning rates by module name"
----
-
-## Background
-
-Inspired by LoRA+, Axolotl allows practitioners to specify separate learning rates for each module or groups of
-modules in a model.
-
-## Example
-
-```yaml
-lr_groups:
-  - name: o_proj
-    modules:
-      - self_attn.o_proj.weight
-    lr: 1e-6
-  - name: q_proj
-    modules:
-      - model.layers.2.self_attn.q_proj.weight
-    lr: 1e-5
-
-learning_rate: 2e-5
-```
-
-In this example, we have a default learning rate of 2e-5 across the entire model, but we have a separate learning rate
-of 1e-6 for all the self attention `o_proj` modules across all layers, and a learning are of 1e-5 to the 3rd layer's
-self attention `q_proj` module.

docs/mac.qmd

@@ -1,22 +0,0 @@
----
-title: Mac M-series
-description: Mac M-series support
----
-
-Currently Axolotl on Mac is partially usable, many of the dependencies of Axolotl including Pytorch do not support MPS or have incomplete support.
-
-Current support:
-
-- [x] Support for all models
-- [x] Full training of models
-- [x] LoRA training
-- [x] Sample packing
-- [ ] FP16 and BF16 (awaiting AMP support for MPS in Pytorch)
-- [ ] Tri-dao's flash-attn (until it is supported use spd_attention as an alternative)
-- [ ] xformers
-- [ ] bitsandbytes (meaning no 4/8 bits loading and bnb optimizers)
-- [ ] qlora
-- [ ] DeepSpeed
-
-Untested:
-- FSDP

docs/multi-gpu.qmd

@@ -1,118 +0,0 @@
----
-title: "Multi-GPU Training Guide"
-format:
-  html:
-    toc: true
-    toc-depth: 3
-    number-sections: true
-    code-tools: true
-execute:
-  enabled: false
----
-
-This guide covers advanced training configurations for multi-GPU setups using Axolotl.
-
-## Overview {#sec-overview}
-
-Axolotl supports several methods for multi-GPU training:
-
-- DeepSpeed (recommended)
-- FSDP (Fully Sharded Data Parallel)
-- FSDP + QLoRA
-
-## DeepSpeed {#sec-deepspeed}
-
-DeepSpeed is the recommended approach for multi-GPU training due to its stability and performance. It provides various optimization levels through ZeRO stages.
-
-### Configuration {#sec-deepspeed-config}
-
-Add to your YAML config:
-
-```{.yaml}
-deepspeed: deepspeed_configs/zero1.json
-```
-
-### Usage {#sec-deepspeed-usage}
-
-```{.bash}
-accelerate launch -m axolotl.cli.train examples/llama-2/config.yml --deepspeed deepspeed_configs/zero1.json
-```
-
-### ZeRO Stages {#sec-zero-stages}
-
-We provide default configurations for:
-
-- ZeRO Stage 1 (`zero1.json`)
-- ZeRO Stage 2 (`zero2.json`)
-- ZeRO Stage 3 (`zero3.json`)
-
-Choose based on your memory requirements and performance needs.
-
-## FSDP {#sec-fsdp}
-
-### Basic FSDP Configuration {#sec-fsdp-config}
-
-```{.yaml}
-fsdp:
-  - full_shard
-  - auto_wrap
-fsdp_config:
-  fsdp_offload_params: true
-  fsdp_state_dict_type: FULL_STATE_DICT
-  fsdp_transformer_layer_cls_to_wrap: LlamaDecoderLayer
-```
-
-### FSDP + QLoRA {#sec-fsdp-qlora}
-
-For combining FSDP with QLoRA, see our [dedicated guide](fsdp_qlora.qmd).
-
-## Performance Optimization {#sec-performance}
-
-### Liger Kernel Integration {#sec-liger}
-
-::: {.callout-note}
-Liger Kernel provides efficient Triton kernels for LLM training, offering:
-
-- 20% increase in multi-GPU training throughput
-- 60% reduction in memory usage
-- Compatibility with both FSDP and DeepSpeed
-:::
-
-Configuration:
-
-```{.yaml}
-plugins:
-  - axolotl.integrations.liger.LigerPlugin
-liger_rope: true
-liger_rms_norm: true
-liger_glu_activation: true
-liger_layer_norm: true
-liger_fused_linear_cross_entropy: true
-```
-
-## Troubleshooting {#sec-troubleshooting}
-
-### NCCL Issues {#sec-nccl}
-
-For NCCL-related problems, see our [NCCL troubleshooting guide](nccl.qmd).
-
-### Common Problems {#sec-common-problems}
-
-::: {.panel-tabset}
-
-## Memory Issues
-
-- Reduce `micro_batch_size`
-- Reduce `eval_batch_size`
-- Adjust `gradient_accumulation_steps`
-- Consider using a higher ZeRO stage
-
-## Training Instability
-
-- Start with DeepSpeed ZeRO-2
-- Monitor loss values
-- Check learning rates
-
-:::
-
-For more detailed troubleshooting, see our [debugging guide](debugging.qmd).

docs/multi-node.md

@@ -0,0 +1,45 @@
+# Multi Node
+
+You will need to create a configuration for accelerate, either by using `accelerate config` and follow the instructions or you can use one of the preset below:
+
+~/.cache/huggingface/accelerate/default_config.yaml
+```yaml
+compute_environment: LOCAL_MACHINE
+debug: false
+distributed_type: FSDP
+downcast_bf16: 'no'
+machine_rank: 0 # Set to 0 for the main machine, increment by one for other machines
+main_process_ip: 10.0.0.4 # Set to main machine's IP
+main_process_port: 5000
+main_training_function: main
+mixed_precision: bf16
+num_machines: 2 # Change to the number of machines
+num_processes: 4 # That's the total number of GPUs, (for example: if you have 2 machines with 4 GPU, put 8)
+rdzv_backend: static
+same_network: true
+tpu_env: []
+tpu_use_cluster: false
+tpu_use_sudo: false
+use_cpu: false
+```
+
+Configure your model to use FSDP with for example:
+```yaml
+fsdp:
+  - full_shard
+  - auto_wrap
+fsdp_config:
+  fsdp_offload_params: true
+  fsdp_state_dict_type: FULL_STATE_DICT
+  fsdp_transformer_layer_cls_to_wrap: LlamaDecoderLayer
+```
+
+## Machine configuration
+
+On each machine you need a copy of Axolotl, we suggest using the same commit to ensure compatibility.
+
+You will also need to have the same configuration file for your model on each machine.
+
+On the main machine only, make sure the port you set as `main_process_port` is open in TCP and reachable by other machines.
+
+All you have to do now is launch using accelerate as you would usually do on each machine and voila, the processes will start once you have launched accelerate on every machine.

docs/multi-node.qmd

@@ -1,88 +0,0 @@
----
-title: Multi Node
-description: How to use Axolotl on multiple machines
----
-
-The below are three ways to train multi-node in Axolotl.
-
-::: {.callout-important}
-Each machine needs a copy of Axolotl, we suggest using the same commit to ensure compatibility.
-
-You will also need to have the same configuration file for your model on each machine.
-
-Make sure the main machine is reachable by other machines.
-:::
-
-# Accelerate
-
-You will need to create a configuration for accelerate, either by using `accelerate config` and follow the instructions or you can use one of the preset below:
-
-~/.cache/huggingface/accelerate/default_config.yaml
-```yaml
-compute_environment: LOCAL_MACHINE
-debug: false
-distributed_type: FSDP
-downcast_bf16: 'no'
-machine_rank: 0 # Set to 0 for the main machine, increment by one for other machines
-main_process_ip: 10.0.0.4 # Set to main machine's IP
-main_process_port: 5000
-main_training_function: main
-mixed_precision: bf16
-num_machines: 2 # Change to the number of machines
-num_processes: 4 # That's the total number of GPUs, (for example: if you have 2 machines with 4 GPU, put 8)
-rdzv_backend: static
-same_network: true
-tpu_env: []
-tpu_use_cluster: false
-tpu_use_sudo: false
-use_cpu: false
-```
-
-Configure your model to use FSDP in the Axolotl yaml. For example:
-```yaml
-fsdp:
-  - full_shard
-  - auto_wrap
-fsdp_config:
-  fsdp_offload_params: true
-  fsdp_state_dict_type: FULL_STATE_DICT
-  fsdp_transformer_layer_cls_to_wrap: LlamaDecoderLayer
-```
-
-All you have to do now is launch using accelerate as you would usually do on each machine and voila, the processes will start once you have launched accelerate on every machine.
-
-# Raytrain
-
-Please see ray train doc [here](ray-integration.qmd).
-
-# Torchrun
-
-If you are using Infiniband, we recommend torchrun to utilize the full bandwidth.
-
-Set the following env (change buffersize/socketname depending on your system):
-
-```yaml
-export NCCL_IB_DISABLE=0
-export NCCL_SOCKET_IFNAME="eth0,en,eth,em,bond"
-export NCCL_BUFFSIZE=2097152
-```
-
-Run the following on each node:
-
-```bash
-torchrun --nnodes $num_nodes --nproc_per_node $gpu_per_node --rdzv_id $rdzv_id --rdzv_backend c10d --rdzv_endpoint "$head_node_ip:$head_node_port" -m axolotl.cli.train config.yaml
-```
-
-Please make sure to substitute the placeholder variables.
-
-- `num_nodes`: Number of nodes (containing GPUs)
-- `gpu_per_node`: Number of gpus per node
-- `head_node_ip`: IP of the head node (make sure other machines can connect to this)
-- `head_node_port`: Port of the head node (make sure other machines can connect to this. Default 29400)
-- `rdzv_id`: A unique job ID that is used by the job across nodes.
-
-::: {.callout-note}
-You need to call `axolotl.cli.train` instead of `axolotl train` as the latter calls accelerate under the hood
-:::
-
-More info on the available configs can be found on the Pytorch docs [here](https://pytorch.org/docs/stable/elastic/run.html)

docs/multimodal.qmd

@@ -1,28 +0,0 @@
-# MultiModal / Vision Language Models (BETA)
-
-### Supported Models
-
-- Mllama, i.e. llama with vision models
-
-### Usage
-
-Currently multimodal support is limited and doesn't have full feature parity. To finetune a multimodal Llama w/ LoRA,
-you'll need to use the following in YAML in combination with the rest of the required hyperparams.
-
-```yaml
-base_model: alpindale/Llama-3.2-11B-Vision-Instruct
-processor_type: AutoProcessor
-skip_prepare_dataset: true
-
-chat_template: llama3_2_vision
-datasets:
-  - path: HuggingFaceH4/llava-instruct-mix-vsft
-    type: chat_template
-    split: train[:1%]
-    field_messages: messages
-remove_unused_columns: false
-sample_packing: false
-
-# only finetune the Language model, leave the vision model and vision tower frozen
-lora_target_modules: 'language_model.model.layers.[\d]+.(mlp|cross_attn|self_attn).(up|down|gate|q|k|v|o)_proj'
-```

docs/multipack.qmd

@@ -1,76 +0,0 @@
----
-title: Multipack (Sample Packing)
-description: Multipack is a technique to pack multiple sequences into a single batch to increase training throughput.
----
-
-## Visualization of Multipack with Flash Attention
-
-Because Flash Attention simply drops the attention mask, we do not need to
-construct a 4d attention mask. We only need to concatenate the sequences into
-a single batch and let flash attention know where each new sequence begins.
-
-
-4k context, bsz =4,
-each character represents 256 tokens
-X represents a padding token
-
-```
-   0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5
-[[ A A A A A A A A A A A ]
-   B B B B B B ]
-   C C C C C C C ]
-   D D D D ]]
-
-[[ E E E E E E E E ]
- [ F F F F ]
- [ G G G ]
- [ H H H H ]]
-
-[[ I I I ]
- [ J J J ]
- [ K K K K K]
- [ L L L ]]
-```
-
-after padding to longest input in each step
-```
-   0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5
-[[ A A A A A A A A A A A ]
-   B B B B B B X X X X X X ]
-   C C C C C C C X X X X ]
-   D D D D X X X X X X X ]]
-
-[[ E E E E E E E E ]
- [ F F F F X X X X ]
- [ G G G X X X X X ]
- [ H H H H X X X X ]]
-
-[[ I I I X X ]
- [ J J J X X ]
- [ K K K K K ]
- [ L L L X X ]]
-```
-
-w packing ( note it's the same effective number of tokens per step, but a true bsz of 1)
-```
-   0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5
-[[ A A A A A A A A A A A B B B B B
-   B C C C C C C C D D D D E E E E
-   E E E E F F F F F G G G H H H H
-   I I I J J J J K K K K K L L L X ]]
-```
-
-cu_seqlens:
-[[ 0, 11, 17, 24, 28, 36, 41 44, 48, 51, 55, 60, 64]]
-
-
-## Multipack without Flash Attention
-
-Multipack can still be achieved without Flash attention, but with lower packing
-efficiency as we are not able to join multiple batches into a single batch due to
-context length limits without flash attention. We can use either Pytorch's Scaled
-Dot Product Attention implementation or native Pytorch attention implementation
-along with [4d attention masks](https://github.com/huggingface/transformers/pull/27539)
-to pack sequences together and avoid cross attention.
-
-<img src="./images/4d-mask.png" alt="axolotl" width="800">

docs/nccl.md

@@ -1,7 +1,4 @@
----
-title: NCCL
-description: Troubleshooting NCCL issues
----
+# NCCL
 
 NVIDIA NCCL is a library to facilitate and optimize multi-GPU communication operations, such as broadcast, all-gather, reduce, all-reduce, etc. Broadly, NCCL configuration is highly environment-specific and is configured via several [environment variables](https://docs.nvidia.com/deeplearning/nccl/user-guide/docs/env.html). A common NCCL-related problem occurs when a long-running operation times out causing the training process to abort:
 

docs/ray-integration.qmd

@@ -1,93 +0,0 @@
----
-title: Ray Train integration
-description: How to use Axolotl with Ray Train
----
-
-Axolotl supports using Ray as an alternative to `accelerate` for orchestrating training. This is especially useful for multi-node training since you only have to setup code and dependencies in a single node and launch training as if you were using a single node.
-
-With the `--use-ray` CLI flag, Axolotl will use Ray Train's [`TorchTrainer`](https://docs.ray.io/en/latest/train/api/doc/ray.train.torch.TorchTrainer.html#ray.train.torch.TorchTrainer) to run training.
-
-## Ray cluster setup
-
-A prerequisite using the Ray Train integration is to setup a Ray cluster on your desired node(s). For a detailed guide on how you can get started with ray clusters, check the official Ray docs here: https://docs.ray.io/en/latest/cluster/getting-started.html
-
-Every Ray cluster has one _head_ node and a set of worker nodes. The head node is just like any other worker node, but it also runs certain special processes related to scheduling and orchestration. Ray-enabled scripts are run on the head node and depending on the resources (number of CPUs, GPUs, etc) they request, will be scheduled to run certain tasks on the worker nodes. For more on key concepts behind a Ray cluster, you can refer this [doc](https://docs.ray.io/en/latest/cluster/key-concepts.html#cluster-key-concepts).
-
-## Sanity check
-
-To run a sanity check on whether your ray cluster is setup properly, execute the following on the head node:
-
-```bash
-ray status
-```
-
-The output should have a summary of your Ray cluster - list of all the nodes in your cluster, the number of CPUs and GPUs in your cluster, etc. For example, if you have a cluster with 1 CPU-only head node and 2 4xL40S worker nodes, the output can look like this:
-
-
-```
-Node status
----------------------------------------------------------------
-Active:
- 1 head
-Idle:
- 2 4xL40S:48CPU-384GB
-Pending:
- (no pending nodes)
-Recent failures:
- (no failures)
-
-Resources
----------------------------------------------------------------
-Usage:
- 0.0/96.0 CPU
- 0.0/8.0 GPU
- 0B/800.00GiB memory
- 0B/229.57GiB object_store_memory
-
-Demands:
- (no resource demands)
-```
-
-You should also be able to see the same on the [Ray dashboard](https://docs.ray.io/en/latest/ray-observability/getting-started.html).
-
-
-## Configuring training with Ray Train
-
-You can find an example configuration at `configs/llama-3/lora-1b-ray.yaml`.
-
-The key parameters to note here are:
-
-```yaml
-...
-use_ray: true
-ray_num_workers: 4
-# optional
-resources_per_worker:
-    GPU: 1
-...
-```
-
-- `use_ray`: This is the flag that enables the Ray Train integration. You can either use the corresponding `--use-ray` flag in the CLI or set `use_ray` in the config file.
-- `ray_num_workers`: This is the number of workers/GPUs to use for training.
-- `resources_per_worker`: This is the Ray [resource request](https://docs.ray.io/en/latest/ray-core/scheduling/resources.html) for each worker. This can be used to request a specific GPU type or a custom resource for each worker. For example, if your ray cluster has GPUs of different types, and you only want to use NVIDIA L40S GPUs, you can do
-
-```yaml
-resources_per_worker:
-    accelerator_type:L40S: 0.001
-```
-
-## Launching training
-
-You can simply run the following command on the head node:
-
-```bash
-axolotl train examples/llama-3/lora-1b-ray.yml --use-ray
-```
-
-This will launch training on the head node and workers will be scheduled automatically by Ray Train to run on the appropriate head or worker nodes.
-
-You can also monitor training progress on the Ray dashboard.
-
-Coming back to the example on a Ray cluster with 1 head node and 2 4xL40S worker nodes, let's say you want to make use of all 8 GPUs. You would be able to just set `ray_num_workers: 8` and run the previous command. The Cluster tab will show the following:
-
-![Ray dashboard](./images/ray-cluster-dashboard.png)

docs/reward_modelling.qmd

@@ -1,47 +0,0 @@
----
-title: "Reward Modelling"
-description: "Reward models are used to guide models towards behaviors which is preferred by humans, by training over large datasets annotated with human preferences. "
----
-
-### Overview
-
-Reward modelling is a technique used to train models to predict the reward or value of a given input. This is particularly useful in reinforcement learning scenarios where the model needs to evaluate the quality of its actions or predictions.
-We support the reward modelling techniques supported by `trl`.
-
-### (Outcome) Reward Models
-
-Outcome reward models are trained using data which contains preference annotations for an entire interaction between the user and model (e.g. rather than per-turn or per-step).
-
-```yaml
-base_model: google/gemma-2-2b
-model_type: AutoModelForSequenceClassification
-num_labels: 1
-tokenizer_type: AutoTokenizer
-
-reward_model: true
-chat_template: gemma
-datasets:
-  - path: argilla/distilabel-intel-orca-dpo-pairs
-    type: bradley_terry.chat_template
-
-val_set_size: 0.1
-eval_steps: 100
-```
-
-### Process Reward Models (PRM)
-
-Process reward models are trained using data which contains preference annotations for each step in a series of interactions. Typically, PRMs are trained to provide reward signals over each step of a reasoning trace and are used for downstream reinforcement learning.
-```yaml
-base_model: Qwen/Qwen2.5-3B
-model_type: AutoModelForTokenClassification
-num_labels: 2
-
-process_reward_model: true
-datasets:
-  - path: trl-lib/math_shepherd
-    type: stepwise_supervised
-    split: train
-
-val_set_size: 0.1
-eval_steps: 100
-```

docs/rlhf.qmd

@@ -1,515 +0,0 @@
----
-title: "RLHF (Beta)"
-description: "Reinforcement Learning from Human Feedback is a method whereby a language model is optimized from data using human feedback."
-back-to-top-navigation: true
-toc: true
-toc-depth: 3
----
-
-# Overview
-
-Reinforcement Learning from Human Feedback is a method whereby a language model is optimized from data using human
-feedback. Various methods include, but not limited to:
-
-- Proximal Policy Optimization (PPO) (not yet supported in axolotl)
-- [Direct Preference Optimization (DPO)](#dpo)
-- [Identity Preference Optimization (IPO)](#ipo)
-- [Kahneman-Tversky Optimization (KTO)](#kto)
-- [Odds Ratio Preference Optimization (ORPO)](#orpo)
-
-
-# RLHF using Axolotl
-
-::: {.callout-important}
-This is a BETA feature and many features are not fully implemented. You are encouraged to open new PRs to improve the integration and functionality.
-:::
-
-We rely on the [TRL](https://github.com/huggingface/trl) library for implementations of various RL training methods, which we wrap around to expose in axolotl. Each method has their own supported ways of loading datasets and prompt formats.
-
-::: {.callout-tip}
-You can find what each method supports by going into `src/axolotl/prompt_strategies/{method}` where `{method}` is one of our supported methods. The `type: ` can be retrieved from `{method}.{function_name}`.
-:::
-
-## DPO
-
-Example config:
-
-```yaml
-rl: dpo
-datasets:
-  - path: Intel/orca_dpo_pairs
-    split: train
-    type: chatml.intel
-  - path: argilla/ultrafeedback-binarized-preferences
-    split: train
-    type: chatml
-```
-
-DPO supports the following types with the following dataset format:
-
-### chatml.argilla
-
-```json
-{
-    "system": "...", // optional
-    "instruction": "...",
-    "chosen_response": "...",
-    "rejected_response": "..."
-}
-```
-
-### chatml.argilla_chat
-
-```json
-{
-    "chosen": [
-        {"role": "user", "content": "..."},
-        {"role": "assistant", "content": "..."}
-    ],
-    "rejected": [
-        {"role": "user", "content": "..."},
-        {"role": "assistant", "content": "..."}
-    ]
-}
-```
-
-### chatml.icr
-
-```json
-{
-    "system": "...", // optional
-    "input": "...",
-    "chosen": "...",
-    "rejected": "..."
-}
-```
-
-### chatml.intel
-
-```json
-{
-    "system": "...", // optional
-    "question": "...",
-    "chosen": "...",
-    "rejected": "..."
-}
-```
-
-### chatml.prompt_pairs
-
-```json
-{
-    "system": "...", // optional
-    "prompt": "...",
-    "chosen": "...",
-    "rejected": "..."
-}
-```
-
-### chatml.ultra
-
-```json
-{
-    "system": "...", // optional
-    "prompt": "...",
-    "chosen": [
-        {"role": "user", "content": "..."},
-        {"role": "assistant", "content": "..."}
-    ],
-    "rejected": [
-        {"role": "user", "content": "..."},
-        {"role": "assistant", "content": "..."}
-    ]
-}
-```
-
-### llama3.argilla
-
-```json
-{
-    "system": "...", // optional
-    "instruction": "...",
-    "chosen_response": "...",
-    "rejected_response": "..."
-}
-```
-
-### llama3.argilla_chat
-
-```json
-{
-    "chosen": [
-        {"role": "user", "content": "..."},
-        {"role": "assistant", "content": "..."}
-    ],
-    "rejected": [
-        {"role": "user", "content": "..."},
-        {"role": "assistant", "content": "..."}
-    ]
-}
-```
-
-### llama3.icr
-
-```json
-{
-    "system": "...", // optional
-    "input": "...",
-    "chosen": "...",
-    "rejected": "..."
-}
-```
-
-### llama3.intel
-
-```json
-{
-    "system": "...", // optional
-    "question": "...",
-    "chosen": "...",
-    "rejected": "..."
-}
-```
-
-### llama3.prompt_pairs
-
-```json
-{
-    "system": "...", // optional
-    "prompt": "...",
-    "chosen": "...",
-    "rejected": "..."
-}
-```
-
-### llama3.ultra
-
-```json
-{
-    "system": "...", // optional
-    "prompt": "...",
-    "chosen": [
-        {"role": "user", "content": "..."},
-        {"role": "assistant", "content": "..."}
-    ],
-    "rejected": [
-        {"role": "user", "content": "..."},
-        {"role": "assistant", "content": "..."}
-    ]
-}
-```
-
-### zephyr.nectar
-
-```json
-{
-    "prompt": "...",
-    "answers": [
-        {
-            "answer": "...",
-            "rank": 1
-        },
-        {
-            "answer": "...",
-            "rank": 2
-        }
-        // ... more answers with ranks
-    ]
-}
-```
-
-### chat_template.default
-
-```yaml
-rl: dpo
-datasets:
-  - path: ...
-    split: train
-    type: chat_template.default
-    field_messages: "messages"
-    field_chosen: "chosen"
-    field_rejected: "rejected"
-    message_property_mappings:
-      role: role
-      content: content
-    roles:
-      user: ["user"]
-      assistant: ["assistant"]
-      system: ["system"]
-```
-
-Sample input format:
-
-```json
-{
-    "messages": [
-        {
-            "role": "system",
-            "content": "..."
-        },
-        {
-            "role": "user",
-            "content": "..."
-        },
-        // ... more messages
-    ],
-    "chosen": {
-        "role": "assistant",
-        "content": "..."
-    },
-    "rejected": {
-        "role": "assistant",
-        "content": "..."
-    }
-}
-```
-
-### user_defined.default
-
-For custom behaviors,
-
-```yaml
-rl: dpo
-datasets:
-  - path: ...
-    split: train
-    type: user_defined.default
-
-    field_prompt: "prompt"
-    field_system: "system"
-    field_chosen: "chosen"
-    field_rejected: "rejected"
-    prompt_format: "{prompt}"
-    chosen_format: "{chosen}"
-    rejected_format: "{rejected}"
-```
-
-The input format is a simple JSON input with customizable fields based on the above config.
-
-```json
-{
-    "system": "...",  // optional
-    "prompt": "...",
-    "chosen": "...",
-    "rejected": "..."
-}
-```
-
-## IPO
-
-As IPO is just DPO with a different loss function, all supported options for DPO works here.
-
-```yaml
-rl: ipo
-```
-
-## ORPO
-
-Paper: https://arxiv.org/abs/2403.07691
-
-```yaml
-rl: orpo
-orpo_alpha: 0.1
-remove_unused_columns: false
-
-chat_template: chatml
-datasets:
-  - path: argilla/ultrafeedback-binarized-preferences-cleaned
-    type: chat_template.argilla
-```
-
-ORPO supports the following types with the following dataset format:
-
-### chat_template.argilla
-
-```json
-{
-    "system": "...",  // optional
-    "prompt": "...",  // if available, will be taken as user message for single-turn instead of from list below
-
-    // chosen/rejected should be same till last content and only even-number of alternating user/assistant turns
-    "chosen": [
-        {"role": "user", "content": "..."},
-        {"role": "assistant", "content": "..."}
-    ],
-    "rejected": [
-        {"role": "user", "content": "..."},
-        {"role": "assistant", "content": "..."}
-    ]
-}
-```
-
-## KTO
-
-```yaml
-rl: kto
-rl_beta: 0.5
-kto_desirable_weight: 0.2
-
-remove_unused_columns: false
-
-datasets:
-  - path: argilla/ultrafeedback-binarized-preferences-cleaned-kto
-    type: llama3.ultra
-    split: train
-
-gradient_checkpointing: true
-gradient_checkpointing_kwargs:
-  use_reentrant: true
-```
-
-KTO supports the following types with the following dataset format:
-
-### chatml.argilla
-
-```json
-{
-    "system": "...", // optional
-    "instruction": "...",
-    "completion": "..."
-}
-```
-
-### chatml.argilla_chat
-
-```json
-{
-    "chosen": [
-        {"role": "user", "content": "..."}
-    ],
-    "completion": [
-        {"role": "assistant", "content": "..."}
-    ]
-}
-```
-
-### chatml.intel
-
-```json
-{
-    "system": "...", // optional
-    "question": "...",
-    "completion": "..."
-}
-```
-
-### chatml.prompt_pairs
-
-```json
-{
-    "system": "...", // optional
-    "prompt": "...",
-    "completion": "..."
-}
-```
-
-### chatml.ultra
-
-```json
-{
-    "system": "...", // optional
-    "prompt": "...",
-    "completion": "..."
-}
-```
-
-### llama3.argilla
-
-```json
-{
-    "system": "...", // optional
-    "instruction": "...",
-    "completion": "..."
-}
-```
-
-### llama3.argilla_chat
-
-```json
-{
-    "completion": [
-        {"role": "user", "content": "..."},
-        {"role": "assistant", "content": "..."}
-    ]
-}
-```
-
-### llama3.intel
-
-```json
-{
-    "system": "...", // optional
-    "question": "...",
-    "completion": "..."
-}
-```
-
-### llama3.prompt_pairs
-
-```json
-{
-    "system": "...", // optional
-    "prompt": "...",
-    "completion": "..."
-}
-```
-
-### llama3.ultra
-
-```json
-{
-    "system": "...", // optional
-    "prompt": "...",
-    "completion": "..."
-}
-```
-
-### user_defined.default
-
-For custom behaviors,
-
-```yaml
-rl: kto
-datasets:
-  - path: ...
-    split: train
-    type: user_defined.default
-
-    field_prompt: "prompt"
-    field_system: "system"
-    field_completion: "completion"
-    field_label: "label"
-    prompt_format: "{prompt}"
-    completion_format: "{completion}"
-```
-
-The input format is a simple JSON input with customizable fields based on the above config.
-
-```json
-{
-    "system": "...",  // optional
-    "prompt": "...",
-    "completion": "...",
-    "label": "..."
-}
-```
-
-## Using local dataset files
-
-```yaml
-datasets:
-  - ds_type: json
-    data_files:
-      - orca_rlhf.jsonl
-    split: train
-    type: chatml.intel
-```
-
-## TRL auto-unwrapping for PEFT
-
-TRL supports auto-unwrapping PEFT models for RL training paradigms which rely on a reference model. This significantly reduces memory pressure as an additional refreference model does not need to be loaded, and reference model log-probabilities can be obtained by disabling PEFT adapters. This is enabled by default. To turn it off, pass the following config:
-
-```yaml
-# load ref model when adapter training.
-rl_adapter_ref_model: true
-```

docs/torchao.qmd

@@ -1,19 +0,0 @@
----
-title: "PyTorch ao"
-description: "Custom data types and layouts for training and inference"
----
-
-### Installation
-
-Stable Release from the PyTorch index
-
-```bash
-pip install torchao --extra-index-url https://download.pytorch.org/whl/cu121 # full options are cpu/cu118/cu121/cu124
-```
-
-
-Nightly release
-
-```bash
-pip install --pre torchao-nightly --index-url https://download.pytorch.org/whl/nightly/cu121 # full options are cpu/cu118/cu121/cu124
-```

docs/unsloth.qmd

@@ -1,47 +0,0 @@
----
-title: "Unsloth"
-description: "Hyper-optimized QLoRA finetuning for single GPUs"
----
-
-### Overview
-
-Unsloth provides hand-written optimized kernels for LLM finetuning that slightly improve speed and VRAM over
-standard industry baselines.
-
-
-### Installation
-
-The following will install the correct unsloth and extras from source.
-
-```bash
-python scripts/unsloth_install.py | sh
-```
-
-### Using unsloth w Axolotl
-
-Axolotl exposes a few configuration options to try out unsloth and get most of the performance gains.
-
-Our unsloth integration is currently limited to the following model architectures:
- - llama
-
-These options are specific to LoRA finetuning and cannot be used for multi-GPU finetuning
-```yaml
-unsloth_lora_mlp: true
-unsloth_lora_qkv: true
-unsloth_lora_o: true
-```
-
-These options are composable and can be used with multi-gpu finetuning
-```yaml
-unsloth_cross_entropy_loss: true
-unsloth_rms_norm: true
-unsloth_rope: true
-```
-
-### Limitations
-
-- Single GPU only; e.g. no multi-gpu support
-- No deepspeed or FSDP support (requires multi-gpu)
-- LoRA + QLoRA support only. No full fine tunes or fp8 support.
-- Limited model architecture support. Llama, Phi, Gemma, Mistral only
-- No MoE support.

examples/cerebras/btlm-ft.yml

@@ -1,10 +1,7 @@
 base_model: cerebras/btlm-3b-8k-base
-# optionally might have model_type or tokenizer_type
+base_model_config: cerebras/btlm-3b-8k-base
 model_type: AutoModelForCausalLM
 tokenizer_type: GPT2Tokenizer
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
-
 trust_remote_code: true
 tokenizer_use_fast: true
 tokenizer_legacy: true
@@ -18,7 +15,7 @@ datasets:
   - path: mhenrichsen/alpaca_2k_test
     type: alpaca
 dataset_prepared_path: last_prepared_run
-val_set_size: 0.05
+val_set_size: 0.01
 
 adapter:
 lora_model_dir:
@@ -39,14 +36,14 @@ lora_fan_in_fan_out:
 wandb_project:
 wandb_entity:
 wandb_watch:
-wandb_name:
+wandb_run_id:
 wandb_log_model:
 
-output_dir: ./outputs/btlm-out
+output_dir: btlm-out
 gradient_accumulation_steps: 1
 micro_batch_size: 1
 num_epochs: 1
-optimizer: adamw_torch_fused
+optimizer: adamw_torch
 adam_beta2: 0.95
 adam_eps: 0.000000001
 max_grad_norm: 1.0
@@ -57,8 +54,8 @@ lr_quadratic_warmup: true
 learning_rate: 0.000085
 train_on_inputs: true
 group_by_length: false
-bf16: auto
-fp16:
+bf16: true
+fp16: false
 tf32: true
 
 gradient_checkpointing: false
@@ -76,8 +73,8 @@ gptq_groupsize:
 gptq_model_v1:
 
 warmup_steps: 32
-evals_per_epoch: 4
-saves_per_epoch: 1
+eval_steps:
+save_steps:
 save_total_limit:
 
 debug:

examples/cerebras/qlora.yml

@@ -1,7 +1,5 @@
 base_model: cerebras/Cerebras-GPT-1.3B
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
-
+base_model_config: cerebras/Cerebras-GPT-1.3B
 load_in_8bit: false
 load_in_4bit: true
 strict: false
@@ -10,10 +8,11 @@ datasets:
   - path: teknium/GPT4-LLM-Cleaned
     type: alpaca
 dataset_prepared_path:
-val_set_size: 0.05
+val_set_size: 0.01
 adapter: qlora
 lora_model_dir:
 sequence_len: 2048
+max_packed_sequence_len: 2048
 lora_r: 16
 lora_alpha: 32
 lora_dropout: 0.05
@@ -26,9 +25,9 @@ lora_fan_in_fan_out:
 wandb_project:
 wandb_entity:
 wandb_watch:
-wandb_name:
+wandb_run_id:
 wandb_log_model:
-output_dir: ./outputs/qlora-out
+output_dir: ./qlora-out
 batch_size: 4
 micro_batch_size: 4
 num_epochs: 2
@@ -38,8 +37,8 @@ lr_scheduler: cosine
 learning_rate: 0.0002
 train_on_inputs: false
 group_by_length: false
-bf16: auto
-fp16:
+bf16: true
+fp16: false
 tf32: true
 gradient_checkpointing: true
 early_stopping_patience:
@@ -51,8 +50,8 @@ flash_attention:
 gptq_groupsize:
 gptq_model_v1:
 warmup_steps: 10
-evals_per_epoch: 4
-saves_per_epoch: 1
+eval_steps: 20
+save_steps:
 debug:
 deepspeed:
 weight_decay: 0.1

examples/cloud/modal.yaml

@@ -1,28 +0,0 @@
-project_name:
-volumes:
-  - name: axolotl-data
-    mount: /workspace/data
-  - name: axolotl-artifacts
-    mount: /workspace/artifacts
-
-# environment variables from local to set as secrets
-secrets:
-  - HF_TOKEN
-  - WANDB_API_KEY
-
-# Which branch of axolotl to use remotely
-branch:
-
-# additional custom commands when building the image
-dockerfile_commands:
-
-gpu: h100
-gpu_count: 1
-
-# Train specific configurations
-memory: 128
-timeout: 86400
-
-# Preprocess specific configurations
-memory_preprocess: 32
-timeout_preprocess: 14400

examples/code-llama/13b/lora.yml

@@ -1,9 +1,8 @@
 base_model: codellama/CodeLlama-13b-hf
-# optionally might have model_type or tokenizer_type
+base_model_config: codellama/CodeLlama-13b-hf
 model_type: LlamaForCausalLM
 tokenizer_type: CodeLlamaTokenizer
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
+is_llama_derived_model: true
 
 load_in_8bit: true
 load_in_4bit: false
@@ -13,8 +12,8 @@ datasets:
   - path: mhenrichsen/alpaca_2k_test
     type: alpaca
 dataset_prepared_path:
-val_set_size: 0.05
-output_dir: ./outputs/lora-out
+val_set_size: 0.01
+output_dir: ./lora-out
 
 sequence_len: 4096
 sample_packing: true
@@ -31,20 +30,20 @@ lora_fan_in_fan_out:
 wandb_project:
 wandb_entity:
 wandb_watch:
-wandb_name:
+wandb_run_id:
 wandb_log_model:
 
 gradient_accumulation_steps: 4
 micro_batch_size: 2
-num_epochs: 4
+num_epochs: 3
 optimizer: adamw_bnb_8bit
 lr_scheduler: cosine
 learning_rate: 0.0002
 
 train_on_inputs: false
 group_by_length: false
-bf16: auto
-fp16:
+bf16: true
+fp16: false
 tf32: false
 
 gradient_checkpointing: true
@@ -54,11 +53,10 @@ local_rank:
 logging_steps: 1
 xformers_attention:
 flash_attention: true
-s2_attention:
 
 warmup_steps: 10
-evals_per_epoch: 4
-saves_per_epoch: 1
+eval_steps: 20
+save_steps:
 debug:
 deepspeed:
 weight_decay: 0.0

examples/code-llama/13b/qlora.yml

@@ -1,9 +1,8 @@
 base_model: codellama/CodeLlama-13b-hf
-# optionally might have model_type or tokenizer_type
+base_model_config: codellama/CodeLlama-13b-hf
 model_type: LlamaForCausalLM
 tokenizer_type: CodeLlamaTokenizer
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
+is_llama_derived_model: true
 
 load_in_8bit: false
 load_in_4bit: true
@@ -13,8 +12,8 @@ datasets:
   - path: mhenrichsen/alpaca_2k_test
     type: alpaca
 dataset_prepared_path:
-val_set_size: 0.05
-output_dir: ./outputs/qlora-out
+val_set_size: 0.01
+output_dir: ./qlora-out
 
 adapter: qlora
 lora_model_dir:
@@ -33,20 +32,20 @@ lora_fan_in_fan_out:
 wandb_project:
 wandb_entity:
 wandb_watch:
-wandb_name:
+wandb_run_id:
 wandb_log_model:
 
 gradient_accumulation_steps: 4
 micro_batch_size: 2
-num_epochs: 4
+num_epochs: 3
 optimizer: paged_adamw_32bit
 lr_scheduler: cosine
 learning_rate: 0.0002
 
 train_on_inputs: false
 group_by_length: false
-bf16: auto
-fp16:
+bf16: true
+fp16: false
 tf32: false
 
 gradient_checkpointing: true
@@ -58,8 +57,8 @@ xformers_attention:
 flash_attention: true
 
 warmup_steps: 10
-evals_per_epoch: 4
-saves_per_epoch: 1
+eval_steps: 20
+save_steps:
 debug:
 deepspeed:
 weight_decay: 0.0

examples/code-llama/34b/lora.yml

@@ -1,9 +1,8 @@
 base_model: codellama/CodeLlama-34b-hf
-# optionally might have model_type or tokenizer_type
+base_model_config: codellama/CodeLlama-34b-hf
 model_type: LlamaForCausalLM
 tokenizer_type: CodeLlamaTokenizer
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
+is_llama_derived_model: true
 
 load_in_8bit: true
 load_in_4bit: false
@@ -13,8 +12,8 @@ datasets:
   - path: mhenrichsen/alpaca_2k_test
     type: alpaca
 dataset_prepared_path:
-val_set_size: 0.05
-output_dir: ./outputs/lora-out
+val_set_size: 0.01
+output_dir: ./lora-out
 
 sequence_len: 4096
 sample_packing: true
@@ -31,20 +30,20 @@ lora_fan_in_fan_out:
 wandb_project:
 wandb_entity:
 wandb_watch:
-wandb_name:
+wandb_run_id:
 wandb_log_model:
 
 gradient_accumulation_steps: 4
 micro_batch_size: 2
-num_epochs: 4
+num_epochs: 3
 optimizer: adamw_bnb_8bit
 lr_scheduler: cosine
 learning_rate: 0.0002
 
 train_on_inputs: false
 group_by_length: false
-bf16: auto
-fp16:
+bf16: true
+fp16: false
 tf32: false
 
 gradient_checkpointing: true
@@ -54,11 +53,10 @@ local_rank:
 logging_steps: 1
 xformers_attention:
 flash_attention: true
-s2_attention:
 
 warmup_steps: 10
-evals_per_epoch: 4
-saves_per_epoch: 1
+eval_steps: 20
+save_steps:
 debug:
 deepspeed:
 weight_decay: 0.0

examples/code-llama/34b/qlora.yml

@@ -1,9 +1,8 @@
 base_model: codellama/CodeLlama-34b-hf
-# optionally might have model_type or tokenizer_type
+base_model_config: codellama/CodeLlama-34b-hf
 model_type: LlamaForCausalLM
 tokenizer_type: CodeLlamaTokenizer
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
+is_llama_derived_model: true
 
 load_in_8bit: false
 load_in_4bit: true
@@ -13,8 +12,8 @@ datasets:
   - path: mhenrichsen/alpaca_2k_test
     type: alpaca
 dataset_prepared_path:
-val_set_size: 0.05
-output_dir: ./outputs/qlora-out
+val_set_size: 0.01
+output_dir: ./qlora-out
 
 adapter: qlora
 lora_model_dir:
@@ -33,20 +32,20 @@ lora_fan_in_fan_out:
 wandb_project:
 wandb_entity:
 wandb_watch:
-wandb_name:
+wandb_run_id:
 wandb_log_model:
 
 gradient_accumulation_steps: 4
 micro_batch_size: 2
-num_epochs: 4
+num_epochs: 3
 optimizer: paged_adamw_32bit
 lr_scheduler: cosine
 learning_rate: 0.0002
 
 train_on_inputs: false
 group_by_length: false
-bf16: auto
-fp16:
+bf16: true
+fp16: false
 tf32: false
 
 gradient_checkpointing: true
@@ -58,8 +57,8 @@ xformers_attention:
 flash_attention: true
 
 warmup_steps: 10
-evals_per_epoch: 4
-saves_per_epoch: 1
+eval_steps: 20
+save_steps:
 debug:
 deepspeed:
 weight_decay: 0.0

examples/code-llama/7b/lora.yml

@@ -1,9 +1,8 @@
 base_model: codellama/CodeLlama-7b-hf
-# optionally might have model_type or tokenizer_type
+base_model_config: codellama/CodeLlama-7b-hf
 model_type: LlamaForCausalLM
 tokenizer_type: CodeLlamaTokenizer
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
+is_llama_derived_model: true
 
 load_in_8bit: true
 load_in_4bit: false
@@ -13,8 +12,8 @@ datasets:
   - path: mhenrichsen/alpaca_2k_test
     type: alpaca
 dataset_prepared_path:
-val_set_size: 0.05
-output_dir: ./outputs/lora-out
+val_set_size: 0.01
+output_dir: ./lora-out
 
 sequence_len: 4096
 sample_packing: true
@@ -31,20 +30,20 @@ lora_fan_in_fan_out:
 wandb_project:
 wandb_entity:
 wandb_watch:
-wandb_name:
+wandb_run_id:
 wandb_log_model:
 
 gradient_accumulation_steps: 4
 micro_batch_size: 2
-num_epochs: 4
+num_epochs: 3
 optimizer: adamw_bnb_8bit
 lr_scheduler: cosine
 learning_rate: 0.0002
 
 train_on_inputs: false
 group_by_length: false
-bf16: auto
-fp16:
+bf16: true
+fp16: false
 tf32: false
 
 gradient_checkpointing: true
@@ -54,11 +53,10 @@ local_rank:
 logging_steps: 1
 xformers_attention:
 flash_attention: true
-s2_attention:
 
 warmup_steps: 10
-evals_per_epoch: 4
-saves_per_epoch: 1
+eval_steps: 20
+save_steps:
 debug:
 deepspeed:
 weight_decay: 0.0

examples/code-llama/7b/qlora.yml

@@ -1,9 +1,8 @@
 base_model: codellama/CodeLlama-7b-hf
-# optionally might have model_type or tokenizer_type
+base_model_config: codellama/CodeLlama-7b-hf
 model_type: LlamaForCausalLM
 tokenizer_type: CodeLlamaTokenizer
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
+is_llama_derived_model: true
 
 load_in_8bit: false
 load_in_4bit: true
@@ -13,8 +12,8 @@ datasets:
   - path: mhenrichsen/alpaca_2k_test
     type: alpaca
 dataset_prepared_path:
-val_set_size: 0.05
-output_dir: ./outputs/qlora-out
+val_set_size: 0.01
+output_dir: ./qlora-out
 
 adapter: qlora
 lora_model_dir:
@@ -33,20 +32,20 @@ lora_fan_in_fan_out:
 wandb_project:
 wandb_entity:
 wandb_watch:
-wandb_name:
+wandb_run_id:
 wandb_log_model:
 
 gradient_accumulation_steps: 4
 micro_batch_size: 2
-num_epochs: 4
+num_epochs: 3
 optimizer: paged_adamw_32bit
 lr_scheduler: cosine
 learning_rate: 0.0002
 
 train_on_inputs: false
 group_by_length: false
-bf16: auto
-fp16:
+bf16: true
+fp16: false
 tf32: false
 
 gradient_checkpointing: true
@@ -58,8 +57,8 @@ xformers_attention:
 flash_attention: true
 
 warmup_steps: 10
-evals_per_epoch: 4
-saves_per_epoch: 1
+eval_steps: 20
+save_steps:
 debug:
 deepspeed:
 weight_decay: 0.0

examples/colab-notebooks/colab-axolotl-example.ipynb

@@ -1,357 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Setting up"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "import torch\n",
-    "# Check so there is a gpu available, a T4(free tier) is enough to run this notebook\n",
-    "assert (torch.cuda.is_available()==True)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "!pip install --no-build-isolation axolotl[deepspeed]"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Hugging Face login (optional)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "from huggingface_hub import notebook_login\n",
-    "notebook_login()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Example configuration"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "import yaml\n",
-    "\n",
-    "yaml_string = \"\"\"\n",
-    "base_model: NousResearch/Meta-Llama-3.1-8B\n",
-    "\n",
-    "load_in_8bit: false\n",
-    "load_in_4bit: true\n",
-    "strict: false\n",
-    "\n",
-    "datasets:\n",
-    "  - path: tatsu-lab/alpaca\n",
-    "    type: alpaca\n",
-    "dataset_prepared_path: last_run_prepared\n",
-    "val_set_size: 0.05\n",
-    "output_dir: ./outputs/lora-out\n",
-    "\n",
-    "sequence_len: 2048\n",
-    "sample_packing: true\n",
-    "eval_sample_packing: true\n",
-    "pad_to_sequence_len: true\n",
-    "\n",
-    "adapter: qlora\n",
-    "lora_model_dir:\n",
-    "lora_r: 32\n",
-    "lora_alpha: 16\n",
-    "lora_dropout: 0.05\n",
-    "lora_target_linear: true\n",
-    "lora_fan_in_fan_out:\n",
-    "lora_modules_to_save:\n",
-    "  - embed_tokens\n",
-    "  - lm_head\n",
-    "\n",
-    "wandb_project:\n",
-    "wandb_entity:\n",
-    "wandb_watch:\n",
-    "wandb_name:\n",
-    "wandb_log_model:\n",
-    "\n",
-    "gradient_accumulation_steps: 2\n",
-    "micro_batch_size: 1\n",
-    "num_epochs: 1\n",
-    "optimizer: paged_adamw_8bit\n",
-    "lr_scheduler: cosine\n",
-    "learning_rate: 2e-5\n",
-    "\n",
-    "train_on_inputs: false\n",
-    "group_by_length: false\n",
-    "bf16: auto\n",
-    "fp16:\n",
-    "tf32: false\n",
-    "\n",
-    "gradient_checkpointing: true\n",
-    "early_stopping_patience:\n",
-    "resume_from_checkpoint:\n",
-    "logging_steps: 1\n",
-    "xformers_attention:\n",
-    "flash_attention: false\n",
-    "sdp_attention: true\n",
-    "\n",
-    "warmup_steps: 1\n",
-    "max_steps: 25\n",
-    "evals_per_epoch: 1\n",
-    "eval_table_size:\n",
-    "saves_per_epoch: 1\n",
-    "debug:\n",
-    "deepspeed:\n",
-    "weight_decay: 0.0\n",
-    "fsdp:\n",
-    "fsdp_config:\n",
-    "special_tokens:\n",
-    "  pad_token: <|end_of_text|>\n",
-    "\"\"\"\n",
-    "\n",
-    "\n",
-    "# Convert the YAML string to a Python dictionary\n",
-    "yaml_dict = yaml.safe_load(yaml_string)\n",
-    "\n",
-    "# Specify your file path\n",
-    "file_path = 'test_axolotl.yaml'\n",
-    "\n",
-    "# Write the YAML file\n",
-    "with open(file_path, 'w') as file:\n",
-    "    yaml.dump(yaml_dict, file)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Above we have a configuration file with base LLM model and datasets specified, among many other things. Axolotl can automatically detect whether the specified datasets are on HuggingFace repo or local machine.\n",
-    "\n",
-    "The Axolotl configuration options encompass model and dataset selection, data pre-processing, and training. Let's go through them line by line:\n",
-    "\n",
-    "*   \"base model\": String value, specifies the underlying pre-trained LLM that will be used for finetuning\n",
-    "\n",
-    "Next we have options for model weights quantization. Quantization allows for reduction in occupied memory on GPUs.\n",
-    "\n",
-    "*   \"load_in_8bit\": Boolean value, whether to quantize the model weights into 8-bit integer.\n",
-    "\n",
-    "*   \"load_in_4bit\": Boolean value, whether to quantize the model weights into 4-bit integer.\n",
-    "\n",
-    "*   \"strict\": Boolean value. If false, it allows for overriding established configuration options in the yaml file when executing in command-line interface.\n",
-    "\n",
-    "*   \"datasets\": a list of dicts that contain path and type of data sets as well as other optional configurations where datasets are concerned. Supports multiple datasets.\n",
-    "\n",
-    "*   \"val_set_size\": Either a float value less than one or an integer less than the total size of dataset. Sets the size of validation set from the whole dataset. If float, sets the proportion of the dataset assigned for validation. If integer, sets the direct size of validation set.\n",
-    "\n",
-    "*   \"output_dir\": String value. Path of trained model.\n",
-    "\n",
-    "For data preprocessing:\n",
-    "\n",
-    "*   \"sequence_len\": Integer. Specifies the maximum sequence length of the input. Typically 2048 or less.\n",
-    "\n",
-    "*   \"pad_to_sequence_len\": Boolean. Padding input to maximum sequence length.\n",
-    "\n",
-    "*   \"sample_packing\": Boolean. Specifies whether to use multi-packing with block diagonal attention.\n",
-    "\n",
-    "*   \"special_tokens\": Python dict, optional. Allows users to specify the additional special tokens to be ignored by the tokenizer.\n",
-    "\n",
-    "For LoRA configuration and its hyperparamters:\n",
-    "\n",
-    "*   \"adapter\": String. Either \"lora\" or \"qlora\", depending on user's choice.\n",
-    "\n",
-    "*   \"lora_model_dir\": String, Optional. Path to directory that contains LoRA model, if there is already a trained LoRA model the user would like to use.\n",
-    "\n",
-    "*   \"lora_r\": Integer. Refers to the rank of LoRA decomposition matrices. Higher value will reduce LoRA efficiency. Recommended to be set to 8.\n",
-    "\n",
-    "*   \"lora_alpha\": Integer. Scale the weight matrices by $\\frac{\\text{lora_alpha}}{\\text{lora_r}}$Recommended to be fixed at 16.\n",
-    "\n",
-    "*   \"lora_dropout\": Float that is 1 or less. The dropout probability of a lora layer.\n",
-    "\n",
-    "*   \"lora_target_linear\": Boolean. If true, lora will target all linear modules in the transformers architecture.\n",
-    "\n",
-    "*   \"lora_modules_to_save\": If you added new tokens to the tokenizer, you may need to save some LoRA modules because they need to know the new tokens.\n",
-    "\n",
-    "See [LoRA](https://arxiv.org/abs/2106.09685) for detailed explanation of LoRA implementation.\n",
-    "\n",
-    "For the training configurations:\n",
-    "\n",
-    "*   \"gradient_accumulation_steps\": Integer. The number of steps over which to accumulate gradient for batch training. E.g. if 2, backprop is performed every two steps.\n",
-    "\n",
-    "*   \"micro_batch_size\": Integer. Batch size per gpu / gradient_accumulation_steps\n",
-    "\n",
-    "*   \"num_epochs\": Integer. Number of epochs. One epoch is when training has looped over every batch in the whole data set once.\n",
-    "\n",
-    "*   \"optimizer\": The optimizer to use for the training.\n",
-    "\n",
-    "*   \"learning_rate\": The learning rate.\n",
-    "\n",
-    "*   \"lr_scheduler\": The learning rate scheduler to use for adjusting learning rate during training.\n",
-    "\n",
-    "*   \"train_on_inputs\": Boolean. Whether to ignore or include the user's prompt from the training labels.\n",
-    "\n",
-    "*   \"group_by_length\": Boolean. Whether to group similarly sized data to minimize padding.\n",
-    "\n",
-    "*   \"bf16\": Either \"auto\", \"true\", or \"false\". Whether to use CUDA bf16 floating point format. If set to \"auto\", will automatically apply bf16 should the gpu supports it.\n",
-    "\n",
-    "*   \"fp16\": Optional. Specifies whether to use CUDA fp16. Automatically set to true if \"bf16\" is set to true. Otherwise false.\n",
-    "\n",
-    "*   \"tf32\": Boolean. Whether to use CUDA tf32. Will override bf16.\n",
-    "\n",
-    "*   \"gradient_checkpointing\": Boolean. Whether to use gradient checkpointing https://huggingface.co/docs/transformers/v4.18.0/en/performance#gradient-checkpointing\n",
-    "\n",
-    "*   \"gradient_checkpointing_kwargs\": Python Dict. Fed into the trainer.\n",
-    "\n",
-    "*   \"logging_steps\": Integer. Log training information over every specified number of steps.\n",
-    "\n",
-    "*   \"flash_attention\": Boolean. Whether to use the [flash attention](https://github.com/Dao-AILab/flash-attention) mechanism.\n",
-    "\n",
-    "*   \"sdp_attention\": Boolean. Whether to use the Scaled Dot Product attention mechanism (the attention mechanism in the [original implementation](https://arxiv.org/abs/1706.03762) of transformers.)\n",
-    "\n",
-    "*   \"warmup_steps\": Integer. The number of pre-training steps where a very low learning rate is used.\n",
-    "\n",
-    "*   \"evals_per_epoch\": Integer. Number of evaluations to be performed within one training epoch.\n",
-    "\n",
-    "*   \"saves_per_epoch\": Integer. Number of times the model is saved in one training epoch.\n",
-    "\n",
-    "*   \"weight_decay\": Positive Float. Sets the \"strength\" of weight decay (i.e. setting the coefficient of L2 regularization)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "The above is but a snippet aiming to get users familiarized with the types of streamlined configuration options axolotl provides. For a full list of configuration options, see [here](https://axolotl-ai-cloud.github.io/axolotl/docs/config.html)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Train the model"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "!accelerate launch -m axolotl.cli.train /content/test_axolotl.yaml"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Predict with trained model"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "!accelerate launch -m axolotl.cli.inference /content/test_axolotl.yaml \\\n",
-    "    --lora_model_dir=\"./outputs/lora-out\" --gradio"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Deeper Dive"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "It is also helpful to gain some familiarity over some of the core inner workings of axolotl"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Configuration Normalization"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Axolotl uses a custom Dict class, called ```DictDefault```\n",
-    "to store configurations specified in the yaml configuration file (into a Python variable named ```cfg```). The definition for this custom Dict can be found in the [utils/dict.py](https://github.com/axolotl-ai-cloud/axolotl/blob/main/src/axolotl/utils/dict.py)\n",
-    "\n",
-    "```DictDefault``` is amended such that calling a missing key from it will result in a ```None``` return type. This is important because if some configuration options aren't specified by the user, the ```None``` type allows Axolotl to perform boolean operations to determine the default settings for missing configurations. For more examples on how this is done, check out [utils/config/__init__.py](https://github.com/axolotl-ai-cloud/axolotl/blob/main/src/axolotl/utils/config/__init__.py)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Loading Models, Tokenizers, and Trainer"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "If we inspect [cli.train.py](https://github.com/axolotl-ai-cloud/axolotl/blob/main/src/axolotl/cli/train.py), we will find that most of the heavy lifting were done by the function ```train()``` which is itself imported from [src/axolotl/train.py](https://github.com/axolotl-ai-cloud/axolotl/blob/main/src/axolotl/train.py).\n",
-    "\n",
-    "```train()``` takes care of loading the appropriate tokenizer and pre-trained model through ```load_model()``` and ```load_tokenizer()``` from [src/axolotl/utils/models.py](https://github.com/axolotl-ai-cloud/axolotl/blob/main/src/axolotl/utils/models.py) respectively.\n",
-    "\n",
-    "```load_tokenizer()``` loads in the appropriate tokenizer given the desired model, as well as chat templates.\n",
-    "\n",
-    "```ModelLoader``` class follows after tokenizer has been selected. It will automatically discern the base model type, load in the desired model, as well as applying model-appropriate attention mechanism modifications (e.g. flash attention). Depending on which base model the user chooses in the configuration, ```ModelLoader``` will utilize the corresponding \"attention hijacking\" script. For example, if the user specified the base model to be ```NousResearch/Meta-Llama-3.1-8B```, which is of llama type, and set ```flash_attn``` to ```True```, ```ModelLoader``` will load in [llama_attn_hijack_flash.py](https://github.com/axolotl-ai-cloud/axolotl/blob/main/src/axolotl/monkeypatch/llama_attn_hijack_flash.py). For a list of supported attention hijacking, please refer to the directory [/src/axolotl/monkeypatch/](https://github.com/axolotl-ai-cloud/axolotl/tree/main/src/axolotl/monkeypatch)\n",
-    "\n",
-    "Another important operation encompassed in ```train()``` is setting up the training that takes into account of user-specified traning configurations (e.g. num_epochs, optimizer) through the use of ```setup_trainer()``` from [/src/axolotl/utils/trainer.py](https://github.com/axolotl-ai-cloud/axolotl/blob/main/src/axolotl/utils/trainer.py), which in turn relies on modules from [/src/axolotl/core/trainer_builder.py](https://github.com/axolotl-ai-cloud/axolotl/blob/main/src/axolotl/core/trainer_builder.py).\n",
-    "```trainer_builder.py``` provides a list of trainer object options bespoke for the task type (Causal or Reinforcement learning ('dpo', 'ipo', 'kto') )"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Monkey patch\n",
-    "\n",
-    "The [Monkey patch directory](https://github.com/axolotl-ai-cloud/axolotl/tree/main/src/axolotl/monkeypatch) is where model architecture/optimization patching scripts are stored (these are modifications that are not implemented in the official releases, hence the name monkey patch). It includes attention jacking, ReLoRA, and unsloth optimization."
-   ]
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "Python 3",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "name": "python",
-   "version": "3.9.6"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 2
-}

examples/dbrx/16bit-lora.yaml

@@ -1,84 +0,0 @@
-base_model: LnL-AI/dbrx-base-converted-v2
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
-
-trust_remote_code: true
-
-load_in_8bit: false
-load_in_4bit: false
-strict: false
-
-datasets:
-  - path: tatsu-lab/alpaca
-    type: alpaca
-dataset_prepared_path: last_run_prepared
-val_set_size: 0.0
-output_dir: ./outputs/out
-
-sequence_len: 512
-sample_packing: false
-pad_to_sequence_len: false
-
-wandb_project:
-wandb_entity:
-wandb_watch:
-wandb_name:
-wandb_log_model:
-
-adapter: lora
-lora_model_dir:
-lora_r: 8
-lora_alpha: 16
-lora_dropout: 0.05
-# w1, w2, & v1 will hang the trainer
-lora_target_modules:
-  - q_proj # attn
-  - k_proj # attn
-  - v_proj # attn
-  - out_proj # attn
-  - layer # router
-#  - w1
-#  - w2
-#  - v1
-
-gradient_accumulation_steps: 1
-micro_batch_size: 1
-num_epochs: 1
-optimizer: paged_adamw_8bit
-lr_scheduler: cosine
-learning_rate: 0.0002
-
-train_on_inputs: false
-group_by_length: false
-bf16: auto
-fp16:
-tf32: false
-
-gradient_checkpointing: false  # don't use with fsdp_activation_checkpointing
-gradient_checkpointing_kwargs:
-  use_reentrant: false
-early_stopping_patience:
-resume_from_checkpoint:
-local_rank:
-logging_steps: 1
-xformers_attention:
-flash_attention: true
-
-warmup_steps: 10
-evals_per_epoch:
-saves_per_epoch: 1
-debug:
-weight_decay: 0.0
-fsdp:
-  - full_shard
-  - auto_wrap
-fsdp_config:
-  fsdp_limit_all_gathers: true
-  fsdp_sync_module_states: true
-  fsdp_offload_params: false
-  fsdp_use_orig_params: false
-  fsdp_cpu_ram_efficient_loading: true
-  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
-  fsdp_transformer_layer_cls_to_wrap: DbrxBlock
-  fsdp_state_dict_type: FULL_STATE_DICT
-  fsdp_activation_checkpointing: true

examples/dbrx/8bit-lora.yaml

@@ -1,84 +0,0 @@
-base_model: LnL-AI/dbrx-base-converted-v2
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
-
-trust_remote_code: true
-
-load_in_8bit: true
-load_in_4bit: false
-strict: false
-
-datasets:
-  - path: tatsu-lab/alpaca
-    type: alpaca
-dataset_prepared_path: last_run_prepared
-val_set_size: 0.0
-output_dir: ./outputs/out
-
-sequence_len: 512
-sample_packing: false
-pad_to_sequence_len: false
-
-wandb_project:
-wandb_entity:
-wandb_watch:
-wandb_name:
-wandb_log_model:
-
-adapter: lora
-lora_model_dir:
-lora_r: 8
-lora_alpha: 16
-lora_dropout: 0.05
-# w1, w2, & v1 will hang the trainer
-lora_target_modules:
-  - q_proj # attn
-  - k_proj # attn
-  - v_proj # attn
-  - out_proj # attn
-  - layer # router
-#  - w1
-#  - w2
-#  - v1
-
-gradient_accumulation_steps: 1
-micro_batch_size: 1
-num_epochs: 1
-optimizer: paged_adamw_8bit
-lr_scheduler: cosine
-learning_rate: 0.0002
-
-train_on_inputs: false
-group_by_length: false
-bf16: auto
-fp16:
-tf32: false
-
-gradient_checkpointing: false  # don't use with fsdp_activation_checkpointing
-gradient_checkpointing_kwargs:
-  use_reentrant: false
-early_stopping_patience:
-resume_from_checkpoint:
-local_rank:
-logging_steps: 1
-xformers_attention:
-flash_attention: true
-
-warmup_steps: 10
-evals_per_epoch:
-saves_per_epoch: 1
-debug:
-weight_decay: 0.0
-fsdp:
-  - full_shard
-  - auto_wrap
-fsdp_config:
-  fsdp_limit_all_gathers: true
-  fsdp_sync_module_states: true
-  fsdp_offload_params: false
-  fsdp_use_orig_params: false
-  fsdp_cpu_ram_efficient_loading: true
-  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
-  fsdp_transformer_layer_cls_to_wrap: DbrxBlock
-  fsdp_state_dict_type: FULL_STATE_DICT
-  fsdp_activation_checkpointing: true

examples/dbrx/README.md

@@ -1,26 +0,0 @@
-# DBRX MoE
-
-Currently, for LoRA, only the `q_proj`, `k_proj`, `v_proj` `out_proj` and `layer` Linear layers are trainable.
-
-We are using the "converted" base models based on [this issue](https://huggingface.co/databricks/dbrx-instruct/discussions/10)
-where the Experts are fused as an `nn.Parameter` rather than a `nn.Linear` layer. However, the implementation
-is still a bit buggy and attempting to train a LoRA adapter over those `w1`, `w2` and `v1` layers
-results in the trainer hanging.
-
-
-### FSDP
-We've tested using the [`LnL-AI/dbrx-base-converted-v2`](https://huggingface.co/LnL-AI/dbrx-base-converted-v2) model as the base model for FSDP.
-
-The high memory usage seen w/ FSDP is due to FSDP not supporting 8bit optimizers.
-
-- 16-bit LoRA w/ FSDP
-  - ✅ w/o CPU Offload - 8x80GB uses ~80GiB/gpu
-  - ❌ w/ CPU Offload - `paged_adamw_8bit` optimizer errors from being on cpu
-- ✅ 8-bit LoRA w/ FSDP
-- ❌ 4-bit QLoRA w/ FSDP - errors w/: `Error an illegal memory access was encountered at line 90 in file /src/csrc/ops.cu`
-- ✅ bf16 full finetune w/ FSDP, freezing all but first 8 layers (8x80GB uses ~78GiB/gpu)
-
-
-### Deepspeed
-
-WIP

examples/dbrx/fft-ds-zero3.yaml

@@ -1,59 +0,0 @@
-base_model: LnL-AI/dbrx-base-converted-v2
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
-
-trust_remote_code: true
-
-load_in_8bit: false
-load_in_4bit: false
-strict: false
-
-datasets:
-  - path: tatsu-lab/alpaca
-    type: alpaca
-dataset_prepared_path: last_run_prepared
-val_set_size: 0.0
-output_dir: ./outputs/out
-
-sequence_len: 512
-sample_packing: false
-pad_to_sequence_len: false
-
-unfrozen_parameters:
-  - transformer.blocks.[0-7].
-
-wandb_project:
-wandb_entity:
-wandb_watch:
-wandb_name:
-wandb_log_model:
-
-gradient_accumulation_steps: 1
-micro_batch_size: 1
-num_epochs: 1
-optimizer: paged_adamw_8bit
-lr_scheduler: cosine
-learning_rate: 0.0002
-
-train_on_inputs: false
-group_by_length: false
-bf16: auto
-fp16:
-tf32: false
-
-gradient_checkpointing: true
-gradient_checkpointing_kwargs:
-  use_reentrant: false
-early_stopping_patience:
-resume_from_checkpoint:
-local_rank:
-logging_steps: 1
-xformers_attention:
-flash_attention: true
-
-warmup_steps: 10
-evals_per_epoch:
-saves_per_epoch: 1
-debug:
-weight_decay: 0.0
-deepspeed: deepspeed_configs/zero3_bf16.json

examples/deepseek-v2/fft-fsdp-16b.yaml

@@ -1,69 +0,0 @@
-base_model: deepseek-ai/DeepSeek-V2-Lite
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
-trust_remote_code: true
-
-load_in_8bit: false
-load_in_4bit: false
-strict: false
-
-datasets:
-  - path: tatsu-lab/alpaca
-    type: alpaca
-dataset_prepared_path: last_run_prepared
-val_set_size: 0.0
-output_dir: ./outputs/out
-
-sequence_len: 2048
-sample_packing: true
-pad_to_sequence_len: true
-
-wandb_project:
-wandb_entity:
-wandb_watch:
-wandb_name:
-wandb_log_model:
-
-gradient_accumulation_steps: 8
-micro_batch_size: 1
-num_epochs: 1
-optimizer: adamw_torch_fused
-lr_scheduler: cosine
-learning_rate: 2e-5
-
-train_on_inputs: false
-group_by_length: false
-bf16: auto
-fp16:
-tf32: false
-
-gradient_checkpointing: true
-gradient_checkpointing_kwargs:
-  use_reentrant: false
-early_stopping_patience:
-resume_from_checkpoint:
-logging_steps: 1
-xformers_attention:
-flash_attention: true
-
-warmup_steps: 100
-evals_per_epoch: 2
-eval_table_size:
-saves_per_epoch: 1
-debug:
-deepspeed:
-weight_decay: 0.0
-special_tokens:
-fsdp:
-  - full_shard
-  - auto_wrap
-fsdp_config:
-  fsdp_limit_all_gathers: true
-  fsdp_sync_module_states: true
-  fsdp_offload_params: true
-  fsdp_use_orig_params: false
-  fsdp_cpu_ram_efficient_loading: true
-  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
-  fsdp_transformer_layer_cls_to_wrap: DeepseekV2DecoderLayer
-  fsdp_state_dict_type: FULL_STATE_DICT
-  fsdp_sharding_strategy: FULL_SHARD

examples/deepseek-v2/qlora-fsdp-2_5.yaml

@@ -1,90 +0,0 @@
-base_model: axolotl-quants/DeepSeek-V2.5-bnb-nf4-bf16
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
-
-trust_remote_code: true
-
-load_in_8bit: false
-load_in_4bit: true
-strict: false
-
-
-plugins:
-  - axolotl.integrations.liger.LigerPlugin
-liger_rms_norm: true
-liger_glu_activation: true
-liger_fused_linear_cross_entropy: true
-
-chat_template: deepseek_v2
-datasets:
-  - path: mlabonne/FineTome-100k
-    type: chat_template
-    split: train[:20%]
-    field_messages: conversations
-    message_property_mappings:
-      role: from
-      content: value
-
-dataset_prepared_path: last_run_prepared
-val_set_size: 0.0
-output_dir: ./outputs/out
-
-sequence_len: 4096
-sample_packing: true
-pad_to_sequence_len: true
-
-wandb_project:
-wandb_entity:
-wandb_watch:
-wandb_name:
-wandb_log_model:
-
-adapter: qlora
-lora_r: 256
-lora_alpha: 256
-lora_target_linear: true
-peft_use_rslora: true
-
-gradient_accumulation_steps: 1
-micro_batch_size: 8
-num_epochs: 1
-optimizer: adamw_torch_fused
-lr_scheduler: cosine
-learning_rate: 2e-5
-
-train_on_inputs: false
-group_by_length: false
-bf16: auto
-fp16:
-tf32: false
-
-gradient_checkpointing: true
-gradient_checkpointing_kwargs:
-  use_reentrant: false
-early_stopping_patience:
-resume_from_checkpoint:
-logging_steps: 1
-xformers_attention:
-flash_attention: true
-
-warmup_steps: 100
-evals_per_epoch: 2
-eval_table_size:
-saves_per_epoch: 1
-debug:
-deepspeed:
-weight_decay: 0.0
-special_tokens:
-fsdp:
-  - full_shard
-  - auto_wrap
-fsdp_config:
-  fsdp_limit_all_gathers: true
-  fsdp_sync_module_states: true
-  fsdp_offload_params: true
-  fsdp_use_orig_params: false
-  fsdp_cpu_ram_efficient_loading: true
-  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
-  fsdp_transformer_layer_cls_to_wrap: DeepseekV2DecoderLayer
-  fsdp_state_dict_type: FULL_STATE_DICT
-  fsdp_sharding_strategy: FULL_SHARD

examples/falcon/config-7b-lora.yml

@@ -1,13 +1,9 @@
 base_model: tiiuae/falcon-7b
-# optionally might have model_type or tokenizer_type
+base_model_config: tiiuae/falcon-7b
+trust_remote_code: true
 model_type: AutoModelForCausalLM
 tokenizer_type: AutoTokenizer
-# Automatically upload checkpoint and final model to HF
-# hub_model_id: username/custom_model_name
-
-# required by falcon custom model code: https://huggingface.co/tiiuae/falcon-7b/tree/main
-trust_remote_code: true
-
+is_falcon_derived_model: true
 load_in_8bit: true
 load_in_4bit: false
 gptq: false
@@ -17,7 +13,7 @@ datasets:
   - path: teknium/GPT4-LLM-Cleaned
     type: alpaca:chat
 dataset_prepared_path:
-val_set_size: 0.05
+val_set_size: 0.01
 adapter: lora
 lora_model_dir:
 sequence_len: 2048
@@ -31,9 +27,9 @@ lora_fan_in_fan_out:
 wandb_project:
 wandb_entity:
 wandb_watch:
-wandb_name:
+wandb_run_id:
 wandb_log_model:
-output_dir: ./outputs/falcon-7b
+output_dir: ./falcon-7b
 batch_size: 2
 micro_batch_size: 1
 num_epochs: 4
@@ -43,8 +39,8 @@ lr_scheduler: cosine
 learning_rate: 0.00003
 train_on_inputs: false
 group_by_length: false
-bf16: auto
-fp16:
+bf16: true
+fp16: false
 tf32: true
 gradient_checkpointing: true
 early_stopping_patience:
@@ -56,8 +52,8 @@ flash_attention:
 gptq_groupsize:
 gptq_model_v1:
 warmup_steps: 40
-evals_per_epoch: 4
-saves_per_epoch: 1
+eval_steps: 5
+save_steps: 43
 debug:
 deepspeed:
 weight_decay: 0.0
@@ -65,5 +61,5 @@ fsdp:
 fsdp_config:
 special_tokens:
   pad_token: "<|endoftext|>"
-  bos_token: "<|endoftext|>"
+  bos_token: ">>ABSTRACT<<"
   eos_token: "<|endoftext|>"