diff --git a/.nojekyll b/.nojekyll index 9add1d60c..ea9863a21 100644 --- a/.nojekyll +++ b/.nojekyll @@ -1 +1 @@ -159de4cc \ No newline at end of file +ffb47d96 \ No newline at end of file diff --git a/docs/custom_integrations.html b/docs/custom_integrations.html index d7747aa45..22c3fd8bf 100644 --- a/docs/custom_integrations.html +++ b/docs/custom_integrations.html @@ -581,6 +581,8 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
  • gemma2
  • gemma3
  • gemma3_text
    • +
  • gemma3n
    • +
  • gemma3n_text
  • glm
  • glm4
  • llama
    • diff --git a/docs/multimodal.html b/docs/multimodal.html index 2ac254bc3..7a17aabc9 100644 --- a/docs/multimodal.html +++ b/docs/multimodal.html @@ -482,10 +482,17 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
  • Llava-1.5
  • Mistral-Small-3.1
  • Gemma-3
    • +
  • Gemma-3n
  • Qwen2-VL
  • Qwen2.5-VL
    • -
  • Dataset Format
    • +
  • Dataset Format +
    • +
  • FAQ
    • @@ -520,6 +527,7 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
  • Llava-1.5
  • Mistral-Small-3.1
  • Gemma-3
    • +
  • Gemma-3n
  • Qwen2-VL
  • Qwen2.5-VL
    • @@ -616,17 +624,49 @@ Tip chat_template: gemma3 +
    +

    Gemma-3n

    +
    +
    +
    + +
    +
    +Warning +
    +
    +
    +

    The model’s initial loss and grad norm will be very high. We suspect this to be due to the Conv in the vision layers.

    +
    +
    +
    +
    +
    + +
    +
    +Tip +
    +
    +
    +

    Please make sure to install timm via pip3 install timm==1.0.17

    +
    +
    +
    base_model: google/gemma-3n-E2B-it
+
+chat_template: gemma3n
    +

    Qwen2-VL

    -
    base_model: Qwen/Qwen2-VL-7B-Instruct
-
-chat_template: qwen2_vl
    +
    base_model: Qwen/Qwen2-VL-7B-Instruct
+
+chat_template: qwen2_vl

    Qwen2.5-VL

    -
    base_model: Qwen/Qwen2.5-VL-7B-Instruct
-
-chat_template: qwen2_vl  # same as qwen2-vl
    +
    base_model: Qwen/Qwen2.5-VL-7B-Instruct
+
+chat_template: qwen2_vl  # same as qwen2-vl
    @@ -635,8 +675,10 @@ Tip +
    +

    Image

    @@ -654,6 +696,22 @@ Note
    +

    For image loading, you can use the following keys within content alongside "type": "image":

    +
      +
    • "path": "/path/to/image.jpg"
      • +
    • "url": "https://example.com/image.jpg"
      • +
    • "base64": "..."
      • +
    • "image": PIL.Image
      • +
    +
    +
    +

    Audio

    +

    For audio loading, you can use the following keys within content alongside "type": "audio":

    +
    @@ -664,41 +722,46 @@ Tip
    -

    For image loading, you can use the following keys within content alongside "type": "image":

    -
      -
    • "path": "/path/to/image.jpg"
      • -
    • "url": "https://example.com/image.jpg"
      • -
    • "base64": "..."
      • -
    • "image": PIL.Image
      • -
    +

    You may need to install librosa via pip3 install librosa==0.11.0.

    +
    +
    +

    Example

    Here is an example of a multi-modal dataset:

    -
    [
-  {
-    "messages": [
-        {
-            "role": "system",
-            "content": [
-              {"type": "text", "text": "You are a helpful assistant."}
-              ]
-        },
-        {
-            "role": "user",
-            "content": [
-                {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg"},
-                {"type": "text", "text": "Describe this image in detail."}
-            ]
-        },
-        {
-            "role": "assistant",
-            "content": [
-              {"type": "text", "text": "The image is a bee."}
-            ]
-        }
-    ]
-  }
-]
    +
    [
+  {
+    "messages": [
+        {
+            "role": "system",
+            "content": [
+              {"type": "text", "text": "You are a helpful assistant."}
+              ]
+        },
+        {
+            "role": "user",
+            "content": [
+                {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg"},
+                {"type": "text", "text": "Describe this image in detail."}
+            ]
+        },
+        {
+            "role": "assistant",
+            "content": [
+              {"type": "text", "text": "The image is a bee."}
+            ]
+        }
+    ]
+  }
+]
    +
    +
    +
    +

    FAQ

    +
      +
    1. PIL.UnidentifiedImageError: cannot identify image file ...
      2. +
    +

    PIL could not retrieve the file at url using requests. Please check for typo. One alternative reason is that the request is blocked by the server.

    diff --git a/search.json b/search.json index dc56f02d8..1b302d417 100644 --- a/search.json +++ b/search.json @@ -1153,7 +1153,7 @@ "href": "docs/custom_integrations.html#cut-cross-entropy", "title": "Custom Integrations", "section": "Cut Cross Entropy", - "text": "Cut Cross Entropy\nCut Cross Entropy (CCE) reduces VRAM usage through optimization on the cross-entropy operation during loss calculation.\nSee https://github.com/apple/ml-cross-entropy\n\nRequirements\n\nPyTorch 2.4.0 or higher\n\n\n\nInstallation\nRun the following command to install cut_cross_entropy[transformers] if you don’t have it already.\n\nIf you are in dev environment\n\npython scripts/cutcrossentropy_install.py | sh\n\nIf you are installing from pip\n\npip3 uninstall -y cut-cross-entropy && pip3 install \"cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@631d646\"\n\n\nUsage\nplugins:\n - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin\n\n\nSupported Models\n\ncohere\ncohere2\ngemma\ngemma2\ngemma3\ngemma3_text\nglm\nglm4\nllama\nllama4\nllama4_text\nmistral\nmistral3\nmllama\nphi\nphi3\nphi4_multimodal\nqwen2\nqwen2_vl\nqwen2_moe\nqwen2_5_vl\nqwen3\nqwen3_moe\n\n\n\nCitation\n@article{wijmans2024cut,\n author = {Erik Wijmans and\n Brody Huval and\n Alexander Hertzberg and\n Vladlen Koltun and\n Philipp Kr\\\"ahenb\\\"uhl},\n title = {Cut Your Losses in Large-Vocabulary Language Models},\n journal = {arXiv},\n year = {2024},\n url = {https://arxiv.org/abs/2411.09009},\n}\nPlease see reference here", + "text": "Cut Cross Entropy\nCut Cross Entropy (CCE) reduces VRAM usage through optimization on the cross-entropy operation during loss calculation.\nSee https://github.com/apple/ml-cross-entropy\n\nRequirements\n\nPyTorch 2.4.0 or higher\n\n\n\nInstallation\nRun the following command to install cut_cross_entropy[transformers] if you don’t have it already.\n\nIf you are in dev environment\n\npython scripts/cutcrossentropy_install.py | sh\n\nIf you are installing from pip\n\npip3 uninstall -y cut-cross-entropy && pip3 install \"cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@631d646\"\n\n\nUsage\nplugins:\n - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin\n\n\nSupported Models\n\ncohere\ncohere2\ngemma\ngemma2\ngemma3\ngemma3_text\ngemma3n\ngemma3n_text\nglm\nglm4\nllama\nllama4\nllama4_text\nmistral\nmistral3\nmllama\nphi\nphi3\nphi4_multimodal\nqwen2\nqwen2_vl\nqwen2_moe\nqwen2_5_vl\nqwen3\nqwen3_moe\n\n\n\nCitation\n@article{wijmans2024cut,\n author = {Erik Wijmans and\n Brody Huval and\n Alexander Hertzberg and\n Vladlen Koltun and\n Philipp Kr\\\"ahenb\\\"uhl},\n title = {Cut Your Losses in Large-Vocabulary Language Models},\n journal = {arXiv},\n year = {2024},\n url = {https://arxiv.org/abs/2411.09009},\n}\nPlease see reference here", "crumbs": [ "Advanced Features", "Custom Integrations" @@ -2655,7 +2655,7 @@ "href": "docs/multimodal.html", "title": "MultiModal / Vision Language Models (BETA)", "section": "", - "text": "Mllama\nLlama4\nPixtral\nLlava-1.5\nMistral-Small-3.1\nGemma-3\nQwen2-VL\nQwen2.5-VL", + "text": "Mllama\nLlama4\nPixtral\nLlava-1.5\nMistral-Small-3.1\nGemma-3\nGemma-3n\nQwen2-VL\nQwen2.5-VL", "crumbs": [ "How To Guides", "MultiModal / Vision Language Models (BETA)" @@ -2666,7 +2666,7 @@ "href": "docs/multimodal.html#supported-models", "title": "MultiModal / Vision Language Models (BETA)", "section": "", - "text": "Mllama\nLlama4\nPixtral\nLlava-1.5\nMistral-Small-3.1\nGemma-3\nQwen2-VL\nQwen2.5-VL", + "text": "Mllama\nLlama4\nPixtral\nLlava-1.5\nMistral-Small-3.1\nGemma-3\nGemma-3n\nQwen2-VL\nQwen2.5-VL", "crumbs": [ "How To Guides", "MultiModal / Vision Language Models (BETA)" @@ -2677,7 +2677,7 @@ "href": "docs/multimodal.html#usage", "title": "MultiModal / Vision Language Models (BETA)", "section": "Usage", - "text": "Usage\nMultimodal support is limited and doesn’t have full feature parity.\nHere are the hyperparams you’ll need to use to finetune a multimodal model.\nprocessor_type: AutoProcessor\n\nskip_prepare_dataset: true\nremove_unused_columns: false # leave columns in place as they are needed to handle image embeddings during training\nsample_packing: false # not yet supported with multimodal\n\nchat_template: # see in next section\n\n# example dataset\ndatasets:\n - path: HuggingFaceH4/llava-instruct-mix-vsft\n type: chat_template\n split: train[:1%]\n field_messages: messages\n\n# (optional) if doing lora, only finetune the Language model,\n# leave the vision model and vision tower frozen\n# load_in_8bit: true\nadapter: lora\nlora_target_modules: 'model.language_model.layers.[\\d]+.(mlp|cross_attn|self_attn).(up|down|gate|q|k|v|o)_proj'\n\n# (optional) if you want to resize images to a set size\nimage_size: 512\nimage_resize_algorithm: bilinear\nPlease see examples folder for full configs.\n\n\n\n\n\n\nWarning\n\n\n\nSome of our chat_templates have been extended to support broader dataset types. This should not break any existing configs.\n\n\n\nMllama\nbase_model: meta-llama/Llama-3.2-11B-Vision-Instruct\n\nchat_template: llama3_2_vision\n\n\nLlama4\nbase_model: meta-llama/Llama-4-Scout-17B-16E-Instruct\n\nchat_template: llama4\n\n\nPixtral\nbase_model: mistralai/Pixtral-12B-2409\n\nchat_template: pixtral\n\n\nLlava-1.5\nbase_model: llava-hf/llava-1.5-7b-hf\n\nchat_template: llava\n\n\nMistral-Small-3.1\nbase_model: mistralai/Mistral-Small-3.1-24B-Instruct-2503\n\nchat_template: mistral_v7_tekken\n\n\nGemma-3\n\n\n\n\n\n\nTip\n\n\n\nThe Gemma3-1B model is a text-only model, so please train as regular text model.\n\n\nFor multi-modal 4B/12B/27B models, use the following config:\nbase_model: google/gemma-3-4b-it\n\nchat_template: gemma3\n\n\nQwen2-VL\nbase_model: Qwen/Qwen2-VL-7B-Instruct\n\nchat_template: qwen2_vl\n\n\nQwen2.5-VL\nbase_model: Qwen/Qwen2.5-VL-7B-Instruct\n\nchat_template: qwen2_vl # same as qwen2-vl", + "text": "Usage\nMultimodal support is limited and doesn’t have full feature parity.\nHere are the hyperparams you’ll need to use to finetune a multimodal model.\nprocessor_type: AutoProcessor\n\nskip_prepare_dataset: true\nremove_unused_columns: false # leave columns in place as they are needed to handle image embeddings during training\nsample_packing: false # not yet supported with multimodal\n\nchat_template: # see in next section\n\n# example dataset\ndatasets:\n - path: HuggingFaceH4/llava-instruct-mix-vsft\n type: chat_template\n split: train[:1%]\n field_messages: messages\n\n# (optional) if doing lora, only finetune the Language model,\n# leave the vision model and vision tower frozen\n# load_in_8bit: true\nadapter: lora\nlora_target_modules: 'model.language_model.layers.[\\d]+.(mlp|cross_attn|self_attn).(up|down|gate|q|k|v|o)_proj'\n\n# (optional) if you want to resize images to a set size\nimage_size: 512\nimage_resize_algorithm: bilinear\nPlease see examples folder for full configs.\n\n\n\n\n\n\nWarning\n\n\n\nSome of our chat_templates have been extended to support broader dataset types. This should not break any existing configs.\n\n\n\nMllama\nbase_model: meta-llama/Llama-3.2-11B-Vision-Instruct\n\nchat_template: llama3_2_vision\n\n\nLlama4\nbase_model: meta-llama/Llama-4-Scout-17B-16E-Instruct\n\nchat_template: llama4\n\n\nPixtral\nbase_model: mistralai/Pixtral-12B-2409\n\nchat_template: pixtral\n\n\nLlava-1.5\nbase_model: llava-hf/llava-1.5-7b-hf\n\nchat_template: llava\n\n\nMistral-Small-3.1\nbase_model: mistralai/Mistral-Small-3.1-24B-Instruct-2503\n\nchat_template: mistral_v7_tekken\n\n\nGemma-3\n\n\n\n\n\n\nTip\n\n\n\nThe Gemma3-1B model is a text-only model, so please train as regular text model.\n\n\nFor multi-modal 4B/12B/27B models, use the following config:\nbase_model: google/gemma-3-4b-it\n\nchat_template: gemma3\n\n\nGemma-3n\n\n\n\n\n\n\nWarning\n\n\n\nThe model’s initial loss and grad norm will be very high. We suspect this to be due to the Conv in the vision layers.\n\n\n\n\n\n\n\n\nTip\n\n\n\nPlease make sure to install timm via pip3 install timm==1.0.17\n\n\nbase_model: google/gemma-3n-E2B-it\n\nchat_template: gemma3n\n\n\nQwen2-VL\nbase_model: Qwen/Qwen2-VL-7B-Instruct\n\nchat_template: qwen2_vl\n\n\nQwen2.5-VL\nbase_model: Qwen/Qwen2.5-VL-7B-Instruct\n\nchat_template: qwen2_vl # same as qwen2-vl", "crumbs": [ "How To Guides", "MultiModal / Vision Language Models (BETA)" @@ -2688,7 +2688,18 @@ "href": "docs/multimodal.html#dataset-format", "title": "MultiModal / Vision Language Models (BETA)", "section": "Dataset Format", - "text": "Dataset Format\nFor multi-modal datasets, we adopt an extended chat_template format similar to OpenAI’s Message format.\n\nA message is a list of role and content.\nrole can be system, user, assistant, etc.\ncontent is a list of type and (text or image or path or url or base64).\n\n\n\n\n\n\n\nNote\n\n\n\nFor backwards compatibility:\n\nIf the dataset has a images or image column of list[Image], it will be appended to the first content list as {\"type\": \"image\", \"image\": ...}. However, if the content already has a {\"type\": \"image\"} but no image key, it will be set the image key.\nIf content is a string, it will be converted to a list with type as text.\n\n\n\n\n\n\n\n\n\nTip\n\n\n\nFor image loading, you can use the following keys within content alongside \"type\": \"image\":\n\n\"path\": \"/path/to/image.jpg\"\n\"url\": \"https://example.com/image.jpg\"\n\"base64\": \"...\"\n\"image\": PIL.Image\n\n\n\nHere is an example of a multi-modal dataset:\n[\n {\n \"messages\": [\n {\n \"role\": \"system\",\n \"content\": [\n {\"type\": \"text\", \"text\": \"You are a helpful assistant.\"}\n ]\n },\n {\n \"role\": \"user\",\n \"content\": [\n {\"type\": \"image\", \"url\": \"https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg\"},\n {\"type\": \"text\", \"text\": \"Describe this image in detail.\"}\n ]\n },\n {\n \"role\": \"assistant\",\n \"content\": [\n {\"type\": \"text\", \"text\": \"The image is a bee.\"}\n ]\n }\n ]\n }\n]", + "text": "Dataset Format\nFor multi-modal datasets, we adopt an extended chat_template format similar to OpenAI’s Message format.\n\nA message is a list of role and content.\nrole can be system, user, assistant, etc.\ncontent is a list of type and (text, image, path, url, base64, or audio).\n\n\nImage\n\n\n\n\n\n\nNote\n\n\n\nFor backwards compatibility:\n\nIf the dataset has a images or image column of list[Image], it will be appended to the first content list as {\"type\": \"image\", \"image\": ...}. However, if the content already has a {\"type\": \"image\"} but no image key, it will be set the image key.\nIf content is a string, it will be converted to a list with type as text.\n\n\n\nFor image loading, you can use the following keys within content alongside \"type\": \"image\":\n\n\"path\": \"/path/to/image.jpg\"\n\"url\": \"https://example.com/image.jpg\"\n\"base64\": \"...\"\n\"image\": PIL.Image\n\n\n\nAudio\nFor audio loading, you can use the following keys within content alongside \"type\": \"audio\":\n\n\"path\": \"/path/to/audio.mp3\"\n\"url\": \"https://example.com/audio.mp3\"\n\"audio\": np.ndarray\n\n\n\n\n\n\n\nTip\n\n\n\nYou may need to install librosa via pip3 install librosa==0.11.0.\n\n\n\n\nExample\nHere is an example of a multi-modal dataset:\n[\n {\n \"messages\": [\n {\n \"role\": \"system\",\n \"content\": [\n {\"type\": \"text\", \"text\": \"You are a helpful assistant.\"}\n ]\n },\n {\n \"role\": \"user\",\n \"content\": [\n {\"type\": \"image\", \"url\": \"https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg\"},\n {\"type\": \"text\", \"text\": \"Describe this image in detail.\"}\n ]\n },\n {\n \"role\": \"assistant\",\n \"content\": [\n {\"type\": \"text\", \"text\": \"The image is a bee.\"}\n ]\n }\n ]\n }\n]", + "crumbs": [ + "How To Guides", + "MultiModal / Vision Language Models (BETA)" + ] + }, + { + "objectID": "docs/multimodal.html#faq", + "href": "docs/multimodal.html#faq", + "title": "MultiModal / Vision Language Models (BETA)", + "section": "FAQ", + "text": "FAQ\n\nPIL.UnidentifiedImageError: cannot identify image file ...\n\nPIL could not retrieve the file at url using requests. Please check for typo. One alternative reason is that the request is blocked by the server.", "crumbs": [ "How To Guides", "MultiModal / Vision Language Models (BETA)" diff --git a/sitemap.xml b/sitemap.xml index e00cae60c..3c7636253 100644 --- a/sitemap.xml +++ b/sitemap.xml @@ -2,762 +2,762 @@ https://docs.axolotl.ai/TODO.html - 2025-07-21T15:42:15.517Z + 2025-07-22T09:52:23.346Z https://docs.axolotl.ai/index.html - 2025-07-21T15:42:15.537Z + 2025-07-22T09:52:23.366Z https://docs.axolotl.ai/docs/debugging.html - 2025-07-21T15:42:15.519Z + 2025-07-22T09:52:23.348Z https://docs.axolotl.ai/docs/amd_hpc.html - 2025-07-21T15:42:15.518Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/api/utils.callbacks.mlflow_.html - 2025-07-21T15:45:36.473Z + 2025-07-22T09:55:24.104Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_expand_mask.html - 2025-07-21T15:45:35.908Z + 2025-07-22T09:55:23.540Z https://docs.axolotl.ai/docs/api/loaders.patch_manager.html - 2025-07-21T15:45:35.517Z + 2025-07-22T09:55:23.150Z https://docs.axolotl.ai/docs/api/core.chat.format.llama3x.html - 2025-07-21T15:45:35.213Z + 2025-07-22T09:55:22.845Z https://docs.axolotl.ai/docs/api/cli.train.html - 2025-07-21T15:45:35.266Z + 2025-07-22T09:55:22.899Z https://docs.axolotl.ai/docs/api/utils.callbacks.perplexity.html - 2025-07-21T15:45:36.465Z + 2025-07-22T09:55:24.096Z https://docs.axolotl.ai/docs/api/core.chat.messages.html - 2025-07-21T15:45:35.210Z + 2025-07-22T09:55:22.842Z https://docs.axolotl.ai/docs/api/utils.callbacks.lisa.html - 2025-07-21T15:45:36.470Z + 2025-07-22T09:55:24.101Z https://docs.axolotl.ai/docs/api/cli.merge_sharded_fsdp_weights.html - 2025-07-21T15:45:35.352Z + 2025-07-22T09:55:22.984Z https://docs.axolotl.ai/docs/api/cli.sweeps.html - 2025-07-21T15:45:35.366Z + 2025-07-22T09:55:22.998Z https://docs.axolotl.ai/docs/api/utils.chat_templates.html - 2025-07-21T15:45:36.004Z + 2025-07-22T09:55:23.636Z https://docs.axolotl.ai/docs/api/core.chat.format.shared.html - 2025-07-21T15:45:35.214Z + 2025-07-22T09:55:22.847Z https://docs.axolotl.ai/docs/api/core.trainers.mixins.optimizer.html - 2025-07-21T15:45:35.524Z + 2025-07-22T09:55:23.157Z https://docs.axolotl.ai/docs/api/utils.collators.mamba.html - 2025-07-21T15:45:36.413Z + 2025-07-22T09:55:24.044Z https://docs.axolotl.ai/docs/api/logging_config.html - 2025-07-21T15:45:35.159Z + 2025-07-22T09:55:22.791Z https://docs.axolotl.ai/docs/api/utils.collators.mm_chat.html - 2025-07-21T15:45:36.418Z + 2025-07-22T09:55:24.049Z https://docs.axolotl.ai/docs/api/prompt_strategies.completion.html - 2025-07-21T15:45:35.644Z + 2025-07-22T09:55:23.277Z https://docs.axolotl.ai/docs/api/kernels.utils.html - 2025-07-21T15:45:35.857Z + 2025-07-22T09:55:23.489Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.chat_template.html - 2025-07-21T15:45:35.677Z + 2025-07-22T09:55:23.310Z https://docs.axolotl.ai/docs/api/kernels.swiglu.html - 2025-07-21T15:45:35.848Z + 2025-07-22T09:55:23.480Z https://docs.axolotl.ai/docs/api/common.const.html - 2025-07-21T15:45:36.374Z + 2025-07-22T09:55:24.005Z https://docs.axolotl.ai/docs/api/cli.cloud.base.html - 2025-07-21T15:45:35.408Z + 2025-07-22T09:55:23.040Z https://docs.axolotl.ai/docs/api/utils.callbacks.comet_.html - 2025-07-21T15:45:36.477Z + 2025-07-22T09:55:24.108Z https://docs.axolotl.ai/docs/api/monkeypatch.gradient_checkpointing.offload_cpu.html - 2025-07-21T15:45:35.970Z + 2025-07-22T09:55:23.602Z https://docs.axolotl.ai/docs/api/common.architectures.html - 2025-07-21T15:45:36.372Z + 2025-07-22T09:55:24.003Z https://docs.axolotl.ai/docs/api/prompt_strategies.pygmalion.html - 2025-07-21T15:45:35.671Z + 2025-07-22T09:55:23.304Z https://docs.axolotl.ai/docs/api/utils.schemas.peft.html - 2025-07-21T15:45:36.178Z + 2025-07-22T09:55:23.810Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.user_defined.html - 2025-07-21T15:45:35.701Z + 2025-07-22T09:55:23.333Z https://docs.axolotl.ai/docs/api/utils.schemas.datasets.html - 2025-07-21T15:45:36.170Z + 2025-07-22T09:55:23.801Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_w_system.html - 2025-07-21T15:45:35.618Z + 2025-07-22T09:55:23.250Z https://docs.axolotl.ai/docs/api/prompt_strategies.base.html - 2025-07-21T15:45:35.559Z + 2025-07-22T09:55:23.191Z https://docs.axolotl.ai/docs/api/utils.lora.html - 2025-07-21T15:45:36.009Z + 2025-07-22T09:55:23.641Z https://docs.axolotl.ai/docs/api/prompt_strategies.input_output.html - 2025-07-21T15:45:35.650Z + 2025-07-22T09:55:23.283Z https://docs.axolotl.ai/docs/api/utils.schemas.trl.html - 2025-07-21T15:45:36.182Z + 2025-07-22T09:55:23.813Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.zephyr.html - 2025-07-21T15:45:35.699Z + 2025-07-22T09:55:23.332Z https://docs.axolotl.ai/docs/api/integrations.kd.trainer.html - 2025-07-21T15:45:36.361Z + 2025-07-22T09:55:23.992Z https://docs.axolotl.ai/docs/api/monkeypatch.gradient_checkpointing.offload_disk.html - 2025-07-21T15:45:35.996Z + 2025-07-22T09:55:23.628Z https://docs.axolotl.ai/docs/api/utils.optimizers.adopt.html - 2025-07-21T15:45:36.098Z + 2025-07-22T09:55:23.730Z https://docs.axolotl.ai/docs/api/monkeypatch.data.batch_dataset_fetcher.html - 2025-07-21T15:45:35.965Z + 2025-07-22T09:55:23.597Z https://docs.axolotl.ai/docs/api/cli.cloud.modal_.html - 2025-07-21T15:45:35.414Z + 2025-07-22T09:55:23.046Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_chat.html - 2025-07-21T15:45:35.604Z + 2025-07-22T09:55:23.237Z https://docs.axolotl.ai/docs/api/utils.freeze.html - 2025-07-21T15:45:36.026Z + 2025-07-22T09:55:23.658Z https://docs.axolotl.ai/docs/api/prompt_strategies.bradley_terry.llama3.html - 2025-07-21T15:45:35.744Z + 2025-07-22T09:55:23.376Z https://docs.axolotl.ai/docs/api/integrations.base.html - 2025-07-21T15:45:36.349Z + 2025-07-22T09:55:23.980Z https://docs.axolotl.ai/docs/api/monkeypatch.unsloth_.html - 2025-07-21T15:45:35.964Z + 2025-07-22T09:55:23.596Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.chatml.html - 2025-07-21T15:45:35.718Z + 2025-07-22T09:55:23.351Z https://docs.axolotl.ai/docs/api/cli.main.html - 2025-07-21T15:45:35.258Z + 2025-07-22T09:55:22.890Z https://docs.axolotl.ai/docs/api/common.datasets.html - 2025-07-21T15:45:36.388Z + 2025-07-22T09:55:24.019Z https://docs.axolotl.ai/docs/api/train.html - 2025-07-21T15:45:35.074Z + 2025-07-22T09:55:22.706Z https://docs.axolotl.ai/docs/api/core.trainers.base.html - 2025-07-21T15:45:35.429Z + 2025-07-22T09:55:23.062Z https://docs.axolotl.ai/docs/api/core.trainers.mixins.scheduler.html - 2025-07-21T15:45:35.534Z + 2025-07-22T09:55:23.167Z https://docs.axolotl.ai/docs/api/utils.ctx_managers.sequence_parallel.html - 2025-07-21T15:45:35.557Z + 2025-07-22T09:55:23.190Z https://docs.axolotl.ai/docs/api/utils.schemas.config.html - 2025-07-21T15:45:36.140Z + 2025-07-22T09:55:23.771Z https://docs.axolotl.ai/docs/api/loaders.tokenizer.html - 2025-07-21T15:45:35.502Z + 2025-07-22T09:55:23.134Z https://docs.axolotl.ai/docs/api/integrations.liger.args.html - 2025-07-21T15:45:36.364Z + 2025-07-22T09:55:23.995Z https://docs.axolotl.ai/docs/api/cli.config.html - 2025-07-21T15:45:35.318Z + 2025-07-22T09:55:22.950Z https://docs.axolotl.ai/docs/api/loaders.processor.html - 2025-07-21T15:45:35.504Z + 2025-07-22T09:55:23.136Z https://docs.axolotl.ai/docs/api/monkeypatch.utils.html - 2025-07-21T15:45:35.944Z + 2025-07-22T09:55:23.576Z https://docs.axolotl.ai/docs/api/integrations.cut_cross_entropy.args.html - 2025-07-21T15:45:36.352Z + 2025-07-22T09:55:23.983Z https://docs.axolotl.ai/docs/api/core.trainers.dpo.trainer.html - 2025-07-21T15:45:35.460Z + 2025-07-22T09:55:23.092Z https://docs.axolotl.ai/docs/api/loaders.adapter.html - 2025-07-21T15:45:35.509Z + 2025-07-22T09:55:23.141Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.llama3.html - 2025-07-21T15:45:35.688Z + 2025-07-22T09:55:23.320Z https://docs.axolotl.ai/docs/api/core.datasets.transforms.chat_builder.html - 2025-07-21T15:45:35.227Z + 2025-07-22T09:55:22.859Z https://docs.axolotl.ai/docs/api/monkeypatch.transformers_fa_utils.html - 2025-07-21T15:45:35.962Z + 2025-07-22T09:55:23.594Z https://docs.axolotl.ai/docs/api/datasets.html - 2025-07-21T15:45:35.095Z + 2025-07-22T09:55:22.727Z https://docs.axolotl.ai/docs/api/monkeypatch.btlm_attn_hijack_flash.html - 2025-07-21T15:45:35.945Z + 2025-07-22T09:55:23.578Z https://docs.axolotl.ai/docs/api/cli.inference.html - 2025-07-21T15:45:35.332Z + 2025-07-22T09:55:22.964Z https://docs.axolotl.ai/docs/api/cli.utils.html - 2025-07-21T15:45:35.398Z + 2025-07-22T09:55:23.030Z https://docs.axolotl.ai/docs/api/cli.preprocess.html - 2025-07-21T15:45:35.360Z + 2025-07-22T09:55:22.993Z https://docs.axolotl.ai/docs/api/loaders.model.html - 2025-07-21T15:45:35.494Z + 2025-07-22T09:55:23.126Z https://docs.axolotl.ai/docs/api/core.trainers.grpo.sampler.html - 2025-07-21T15:45:35.483Z + 2025-07-22T09:55:23.115Z https://docs.axolotl.ai/docs/api/core.trainers.grpo.trainer.html - 2025-07-21T15:45:35.471Z + 2025-07-22T09:55:23.103Z https://docs.axolotl.ai/docs/batch_vs_grad.html - 2025-07-21T15:42:15.518Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/custom_integrations.html - 2025-07-21T15:42:15.518Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/quantize.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/unsloth.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/ray-integration.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/dataset-formats/stepwise_supervised.html - 2025-07-21T15:42:15.518Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/dataset-formats/template_free.html - 2025-07-21T15:42:15.518Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/dataset-formats/index.html - 2025-07-21T15:42:15.518Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/dataset-formats/pretraining.html - 2025-07-21T15:42:15.518Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/multi-gpu.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/torchao.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/cli.html - 2025-07-21T15:42:15.518Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/nccl.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/dataset_preprocessing.html - 2025-07-21T15:42:15.519Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/faq.html - 2025-07-21T15:42:15.519Z + 2025-07-22T09:52:23.348Z https://docs.axolotl.ai/docs/qat.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/gradient_checkpointing.html - 2025-07-21T15:42:15.519Z + 2025-07-22T09:52:23.348Z https://docs.axolotl.ai/docs/input_output.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.350Z https://docs.axolotl.ai/src/axolotl/integrations/LICENSE.html - 2025-07-21T15:42:15.541Z + 2025-07-22T09:52:23.370Z https://docs.axolotl.ai/src/axolotl/integrations/cut_cross_entropy/ACKNOWLEDGEMENTS.html - 2025-07-21T15:42:15.542Z + 2025-07-22T09:52:23.370Z https://docs.axolotl.ai/docs/mac.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/lr_groups.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/dataset_loading.html - 2025-07-21T15:42:15.519Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/getting-started.html - 2025-07-21T15:42:15.519Z + 2025-07-22T09:52:23.348Z https://docs.axolotl.ai/docs/lora_optims.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/multi-node.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/fsdp_qlora.html - 2025-07-21T15:42:15.519Z + 2025-07-22T09:52:23.348Z https://docs.axolotl.ai/docs/inference.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.350Z https://docs.axolotl.ai/docs/sequence_parallelism.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/rlhf.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/dataset-formats/tokenized.html - 2025-07-21T15:42:15.519Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/dataset-formats/conversation.html - 2025-07-21T15:42:15.518Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/dataset-formats/inst_tune.html - 2025-07-21T15:42:15.518Z + 2025-07-22T09:52:23.347Z https://docs.axolotl.ai/docs/reward_modelling.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/docker.html - 2025-07-21T15:42:15.519Z + 2025-07-22T09:52:23.348Z https://docs.axolotl.ai/docs/installation.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/multimodal.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/docs/config-reference.html - 2025-07-21T15:45:48.903Z + 2025-07-22T09:55:36.223Z https://docs.axolotl.ai/docs/api/prompt_tokenizers.html - 2025-07-21T15:45:35.150Z + 2025-07-22T09:55:22.782Z https://docs.axolotl.ai/docs/api/utils.schedulers.html - 2025-07-21T15:45:36.067Z + 2025-07-22T09:55:23.699Z https://docs.axolotl.ai/docs/api/utils.samplers.multipack.html - 2025-07-21T15:45:36.459Z + 2025-07-22T09:55:24.089Z https://docs.axolotl.ai/docs/api/prompt_strategies.orcamini.html - 2025-07-21T15:45:35.665Z + 2025-07-22T09:55:23.298Z https://docs.axolotl.ai/docs/api/core.trainers.trl.html - 2025-07-21T15:45:35.444Z + 2025-07-22T09:55:23.076Z https://docs.axolotl.ai/docs/api/prompt_strategies.metharme.html - 2025-07-21T15:45:35.661Z + 2025-07-22T09:55:23.294Z https://docs.axolotl.ai/docs/api/utils.callbacks.profiler.html - 2025-07-21T15:45:36.468Z + 2025-07-22T09:55:24.099Z https://docs.axolotl.ai/docs/api/utils.schemas.enums.html - 2025-07-21T15:45:36.209Z + 2025-07-22T09:55:23.841Z https://docs.axolotl.ai/docs/api/core.trainers.mamba.html - 2025-07-21T15:45:35.449Z + 2025-07-22T09:55:23.081Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_attn_hijack_flash.html - 2025-07-21T15:45:35.883Z + 2025-07-22T09:55:23.514Z https://docs.axolotl.ai/docs/api/monkeypatch.relora.html - 2025-07-21T15:45:35.906Z + 2025-07-22T09:55:23.538Z https://docs.axolotl.ai/docs/api/monkeypatch.stablelm_attn_hijack_flash.html - 2025-07-21T15:45:35.953Z + 2025-07-22T09:55:23.585Z https://docs.axolotl.ai/docs/api/loaders.constants.html - 2025-07-21T15:45:35.519Z + 2025-07-22T09:55:23.151Z https://docs.axolotl.ai/docs/api/utils.callbacks.qat.html - 2025-07-21T15:45:36.484Z + 2025-07-22T09:55:24.115Z https://docs.axolotl.ai/docs/api/utils.schemas.model.html - 2025-07-21T15:45:36.147Z + 2025-07-22T09:55:23.778Z https://docs.axolotl.ai/docs/api/prompt_strategies.stepwise_supervised.html - 2025-07-21T15:45:35.654Z + 2025-07-22T09:55:23.287Z https://docs.axolotl.ai/docs/api/integrations.grokfast.optimizer.html - 2025-07-21T15:45:36.353Z + 2025-07-22T09:55:23.984Z https://docs.axolotl.ai/docs/api/convert.html - 2025-07-21T15:45:35.108Z + 2025-07-22T09:55:22.741Z https://docs.axolotl.ai/docs/api/kernels.quantize.html - 2025-07-21T15:45:35.856Z + 2025-07-22T09:55:23.487Z https://docs.axolotl.ai/docs/api/core.training_args.html - 2025-07-21T15:45:35.187Z + 2025-07-22T09:55:22.819Z https://docs.axolotl.ai/docs/api/prompt_strategies.chat_template.html - 2025-07-21T15:45:35.591Z + 2025-07-22T09:55:23.223Z https://docs.axolotl.ai/docs/api/index.html - 2025-07-21T15:45:35.013Z + 2025-07-22T09:55:22.645Z https://docs.axolotl.ai/docs/api/prompt_strategies.llama2_chat.html - 2025-07-21T15:45:35.638Z + 2025-07-22T09:55:23.271Z https://docs.axolotl.ai/docs/api/utils.trainer.html - 2025-07-21T15:45:36.043Z + 2025-07-22T09:55:23.675Z https://docs.axolotl.ai/docs/api/prompt_strategies.messages.chat.html - 2025-07-21T15:45:35.676Z + 2025-07-22T09:55:23.308Z https://docs.axolotl.ai/docs/api/monkeypatch.lora_kernels.html - 2025-07-21T15:45:35.936Z + 2025-07-22T09:55:23.568Z https://docs.axolotl.ai/docs/api/kernels.lora.html - 2025-07-21T15:45:35.828Z + 2025-07-22T09:55:23.459Z https://docs.axolotl.ai/docs/api/cli.vllm_serve.html - 2025-07-21T15:45:35.405Z + 2025-07-22T09:55:23.037Z https://docs.axolotl.ai/docs/api/utils.schemas.multimodal.html - 2025-07-21T15:45:36.187Z + 2025-07-22T09:55:23.818Z https://docs.axolotl.ai/docs/api/utils.schemas.utils.html - 2025-07-21T15:45:36.215Z + 2025-07-22T09:55:23.847Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_attn_hijack_xformers.html - 2025-07-21T15:45:35.884Z + 2025-07-22T09:55:23.516Z https://docs.axolotl.ai/docs/api/integrations.lm_eval.args.html - 2025-07-21T15:45:36.367Z + 2025-07-22T09:55:23.998Z https://docs.axolotl.ai/docs/api/monkeypatch.mistral_attn_hijack_flash.html - 2025-07-21T15:45:35.898Z + 2025-07-22T09:55:23.530Z https://docs.axolotl.ai/docs/api/utils.collators.core.html - 2025-07-21T15:45:36.391Z + 2025-07-22T09:55:24.022Z https://docs.axolotl.ai/docs/api/core.chat.format.chatml.html - 2025-07-21T15:45:35.211Z + 2025-07-22T09:55:22.844Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.passthrough.html - 2025-07-21T15:45:35.702Z + 2025-07-22T09:55:23.335Z https://docs.axolotl.ai/docs/api/core.datasets.chat.html - 2025-07-21T15:45:35.219Z + 2025-07-22T09:55:22.852Z https://docs.axolotl.ai/docs/api/utils.bench.html - 2025-07-21T15:45:36.018Z + 2025-07-22T09:55:23.650Z https://docs.axolotl.ai/docs/api/utils.schemas.training.html - 2025-07-21T15:45:36.152Z + 2025-07-22T09:55:23.783Z https://docs.axolotl.ai/docs/api/utils.collators.batching.html - 2025-07-21T15:45:36.410Z + 2025-07-22T09:55:24.041Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_patch_multipack.html - 2025-07-21T15:45:35.947Z + 2025-07-22T09:55:23.579Z https://docs.axolotl.ai/docs/api/monkeypatch.multipack.html - 2025-07-21T15:45:35.900Z + 2025-07-22T09:55:23.532Z https://docs.axolotl.ai/docs/api/core.builders.causal.html - 2025-07-21T15:45:35.170Z + 2025-07-22T09:55:22.802Z https://docs.axolotl.ai/docs/api/cli.evaluate.html - 2025-07-21T15:45:35.274Z + 2025-07-22T09:55:22.907Z https://docs.axolotl.ai/docs/api/monkeypatch.trainer_fsdp_optim.html - 2025-07-21T15:45:35.956Z + 2025-07-22T09:55:23.588Z https://docs.axolotl.ai/docs/api/core.trainers.utils.html - 2025-07-21T15:45:35.485Z + 2025-07-22T09:55:23.116Z https://docs.axolotl.ai/docs/api/utils.schemas.integrations.html - 2025-07-21T15:45:36.199Z + 2025-07-22T09:55:23.831Z https://docs.axolotl.ai/docs/api/utils.dict.html - 2025-07-21T15:45:36.090Z + 2025-07-22T09:55:23.722Z https://docs.axolotl.ai/docs/api/core.builders.rl.html - 2025-07-21T15:45:35.175Z + 2025-07-22T09:55:22.807Z https://docs.axolotl.ai/docs/api/prompt_strategies.orpo.chat_template.html - 2025-07-21T15:45:35.740Z + 2025-07-22T09:55:23.373Z https://docs.axolotl.ai/docs/api/core.trainers.relora.html - 2025-07-21T15:45:35.454Z + 2025-07-22T09:55:23.086Z https://docs.axolotl.ai/docs/api/integrations.spectrum.args.html - 2025-07-21T15:45:36.371Z + 2025-07-22T09:55:24.002Z https://docs.axolotl.ai/docs/api/cli.quantize.html - 2025-07-21T15:45:35.419Z + 2025-07-22T09:55:23.051Z https://docs.axolotl.ai/docs/api/cli.checks.html - 2025-07-21T15:45:35.300Z + 2025-07-22T09:55:22.933Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.llama3.html - 2025-07-21T15:45:35.710Z + 2025-07-22T09:55:23.343Z https://docs.axolotl.ai/docs/api/utils.model_shard_quant.html - 2025-07-21T15:45:36.015Z + 2025-07-22T09:55:23.647Z https://docs.axolotl.ai/docs/api/utils.quantization.html - 2025-07-21T15:45:36.127Z + 2025-07-22T09:55:23.758Z https://docs.axolotl.ai/docs/api/core.trainers.mixins.rng_state_loader.html - 2025-07-21T15:45:35.527Z + 2025-07-22T09:55:23.160Z https://docs.axolotl.ai/docs/api/kernels.geglu.html - 2025-07-21T15:45:35.838Z + 2025-07-22T09:55:23.470Z https://docs.axolotl.ai/docs/api/utils.data.pretraining.html - 2025-07-21T15:45:36.099Z + 2025-07-22T09:55:23.731Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.user_defined.html - 2025-07-21T15:45:35.720Z + 2025-07-22T09:55:23.353Z https://docs.axolotl.ai/docs/api/core.builders.base.html - 2025-07-21T15:45:35.165Z + 2025-07-22T09:55:22.798Z https://docs.axolotl.ai/docs/api/cli.merge_lora.html - 2025-07-21T15:45:35.340Z + 2025-07-22T09:55:22.972Z https://docs.axolotl.ai/docs/api/monkeypatch.mixtral.html - 2025-07-21T15:45:35.967Z + 2025-07-22T09:55:23.599Z https://docs.axolotl.ai/docs/api/utils.data.sft.html - 2025-07-21T15:45:36.106Z + 2025-07-22T09:55:23.738Z https://docs.axolotl.ai/docs/api/prompt_strategies.user_defined.html - 2025-07-21T15:45:35.626Z + 2025-07-22T09:55:23.258Z https://docs.axolotl.ai/docs/api/utils.tokenization.html - 2025-07-21T15:45:36.003Z + 2025-07-22T09:55:23.635Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.chatml.html - 2025-07-21T15:45:35.698Z + 2025-07-22T09:55:23.331Z https://docs.axolotl.ai/docs/api/models.mamba.modeling_mamba.html - 2025-07-21T15:45:36.389Z + 2025-07-22T09:55:24.020Z https://docs.axolotl.ai/docs/api/cli.args.html - 2025-07-21T15:45:35.294Z + 2025-07-22T09:55:22.926Z https://docs.axolotl.ai/docs/api/evaluate.html - 2025-07-21T15:45:35.084Z + 2025-07-22T09:55:22.717Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_instruct.html - 2025-07-21T15:45:35.606Z + 2025-07-22T09:55:23.238Z https://docs.axolotl.ai/docs/api/utils.distributed.html - 2025-07-21T15:45:36.087Z + 2025-07-22T09:55:23.719Z https://docs.axolotl.ai/docs/multipack.html - 2025-07-21T15:42:15.522Z + 2025-07-22T09:52:23.351Z https://docs.axolotl.ai/examples/colab-notebooks/colab-axolotl-example.html - 2025-07-21T15:42:15.526Z + 2025-07-22T09:52:23.355Z https://docs.axolotl.ai/FAQS.html - 2025-07-21T15:42:15.517Z + 2025-07-22T09:52:23.345Z