From b2034c645ee8eb71276b781c0ab7fe2f6f0e136b Mon Sep 17 00:00:00 2001 From: Quarto GHA Workflow Runner Date: Wed, 17 Sep 2025 09:44:04 +0000 Subject: [PATCH] Built site for gh-pages --- .nojekyll | 2 +- docs/qat.html | 10 +- docs/quantize.html | 11 +- search.json | 4 +- sitemap.xml | 396 ++++++++++++++++++++++----------------------- 5 files changed, 214 insertions(+), 209 deletions(-) diff --git a/.nojekyll b/.nojekyll index e1b123edd..a55cc4c5f 100644 --- a/.nojekyll +++ b/.nojekyll @@ -1 +1 @@ -1d43a3d7 \ No newline at end of file +e807e5c1 \ No newline at end of file diff --git a/docs/qat.html b/docs/qat.html index 5239ecf0f..ea44201e5 100644 --- a/docs/qat.html +++ b/docs/qat.html @@ -544,10 +544,16 @@ and the QAT documentation in the Configuring QAT in Axolotl

To enable QAT in axolotl, add the following to your configuration file:

qat:
-  activation_dtype: # Optional[str] = "int8". Fake quantization layout to use for activation quantization. Valid options are "int4" and "int8"
-  weight_dtype: # Optional[str] = "int8". Fake quantization layout to use for weight quantization. Valid options are "int4" and "int8"
+  activation_dtype: # Optional[str] = "int8". Fake quantization layout to use for activation quantization. Valid options are "int4", "int8", "float8"
+  weight_dtype: # Optional[str] = "int8". Fake quantization layout to use for weight quantization. Valid options are "int4", "fp8", and "nvfp4".
   group_size: # Optional[int] = 32. The number of elements in each group for per-group fake quantization
   fake_quant_after_n_steps: # Optional[int] = None. The number of steps to apply fake quantization after
+

We support the following quantization schemas: +- Int4WeightOnly (requires the fbgemm-gpu extra when installing Axolotl) +- Int8DynamicActivationInt4Weight +- Float8DynamicActivationFloat8Weight +- Float8DynamicActivationInt4Weight +- NVFP4

Once you have finished training, you must quantize your model by using the same quantization configuration which you used to train the model with. You can use the quantize command to do this.

diff --git a/docs/quantize.html b/docs/quantize.html index ae94e78f1..c6d6bb218 100644 --- a/docs/quantize.html +++ b/docs/quantize.html @@ -548,8 +548,8 @@ Note

Quantization is configured using the quantization key in your configuration file.

base_model: # The path to the model to quantize.
 quantization:
-  weight_dtype: # Optional[str] = "int8". Fake quantization layout to use for weight quantization. Valid options are uintX for X in [1, 2, 3, 4, 5, 6, 7], or int4, or int8
-  activation_dtype: # Optional[str] = "int8". Fake quantization layout to use for activation quantization. Valid options are "int4" and "int8"
+  activation_dtype: # Optional[str] = "int8". Fake quantization layout to use for activation quantization. Valid options are "int4", "int8", "float8"
+  weight_dtype: # Optional[str] = "int8". Fake quantization layout to use for weight quantization. Valid options are "int4", "fp8", and "nvfp4".
   group_size: # Optional[int] = 32. The number of elements in each group for per-group fake quantization
   quantize_embedding: # Optional[bool] = False. Whether to quantize the embedding layer.
 
@@ -560,11 +560,10 @@ you used to train the model:

 
# qat.yml
 qat:
   activation_dtype: int8
-  weight_dtype: int8
+  weight_dtype: int4
   group_size: 256
-  quantize_embedding: true
-
-output_dir: # The path to the output directory used during training where the final checkpoint has been saved.

+
+output_dir: # The path to the output directory used during training where the final checkpoint has been saved.
axolotl quantize qat.yml

This ensures that an identical quantization configuration is used to quantize the model as was used to train it.

diff --git a/search.json b/search.json index 1977249ff..4f9292145 100644 --- a/search.json +++ b/search.json @@ -208,7 +208,7 @@ "href": "docs/quantize.html#configuring-quantization-in-axolotl", "title": "Quantization with torchao", "section": "Configuring Quantization in Axolotl", - "text": "Configuring Quantization in Axolotl\nQuantization is configured using the quantization key in your configuration file.\nbase_model: # The path to the model to quantize.\nquantization:\n weight_dtype: # Optional[str] = \"int8\". Fake quantization layout to use for weight quantization. Valid options are uintX for X in [1, 2, 3, 4, 5, 6, 7], or int4, or int8\n activation_dtype: # Optional[str] = \"int8\". Fake quantization layout to use for activation quantization. Valid options are \"int4\" and \"int8\"\n group_size: # Optional[int] = 32. The number of elements in each group for per-group fake quantization\n quantize_embedding: # Optional[bool] = False. Whether to quantize the embedding layer.\n\noutput_dir: # The path to the output directory.\nOnce quantization is complete, your quantized model will be saved in the {output_dir}/quantized directory.\nYou may also use the quantize command to quantize a model which has been trained with QAT - you can do this by using the existing QAT configuration file which\nyou used to train the model:\n# qat.yml\nqat:\n activation_dtype: int8\n weight_dtype: int8\n group_size: 256\n quantize_embedding: true\n\noutput_dir: # The path to the output directory used during training where the final checkpoint has been saved.\naxolotl quantize qat.yml\nThis ensures that an identical quantization configuration is used to quantize the model as was used to train it.\n\n\n\n\n\n\nNote\n\n\n\nIf you have configured pushing to hub with hub_model_id, your model hub name will have the quantization schema appended to it,\ne.g. axolotl-ai-cloud/qat-nvfp4-llama3B will become axolotl-ai-cloud/qat-nvfp4-llama3B-nvfp4w", + "text": "Configuring Quantization in Axolotl\nQuantization is configured using the quantization key in your configuration file.\nbase_model: # The path to the model to quantize.\nquantization:\n activation_dtype: # Optional[str] = \"int8\". Fake quantization layout to use for activation quantization. Valid options are \"int4\", \"int8\", \"float8\"\n weight_dtype: # Optional[str] = \"int8\". Fake quantization layout to use for weight quantization. Valid options are \"int4\", \"fp8\", and \"nvfp4\".\n group_size: # Optional[int] = 32. The number of elements in each group for per-group fake quantization\n quantize_embedding: # Optional[bool] = False. Whether to quantize the embedding layer.\n\noutput_dir: # The path to the output directory.\nOnce quantization is complete, your quantized model will be saved in the {output_dir}/quantized directory.\nYou may also use the quantize command to quantize a model which has been trained with QAT - you can do this by using the existing QAT configuration file which\nyou used to train the model:\n# qat.yml\nqat:\n activation_dtype: int8\n weight_dtype: int4\n group_size: 256\n\noutput_dir: # The path to the output directory used during training where the final checkpoint has been saved.\naxolotl quantize qat.yml\nThis ensures that an identical quantization configuration is used to quantize the model as was used to train it.\n\n\n\n\n\n\nNote\n\n\n\nIf you have configured pushing to hub with hub_model_id, your model hub name will have the quantization schema appended to it,\ne.g. axolotl-ai-cloud/qat-nvfp4-llama3B will become axolotl-ai-cloud/qat-nvfp4-llama3B-nvfp4w", "crumbs": [ "How To Guides", "Quantization with torchao" @@ -4038,7 +4038,7 @@ "href": "docs/qat.html#configuring-qat-in-axolotl", "title": "Quantization Aware Training (QAT)", "section": "Configuring QAT in Axolotl", - "text": "Configuring QAT in Axolotl\nTo enable QAT in axolotl, add the following to your configuration file:\nqat:\n activation_dtype: # Optional[str] = \"int8\". Fake quantization layout to use for activation quantization. Valid options are \"int4\" and \"int8\"\n weight_dtype: # Optional[str] = \"int8\". Fake quantization layout to use for weight quantization. Valid options are \"int4\" and \"int8\"\n group_size: # Optional[int] = 32. The number of elements in each group for per-group fake quantization\n fake_quant_after_n_steps: # Optional[int] = None. The number of steps to apply fake quantization after\nOnce you have finished training, you must quantize your model by using the same quantization configuration which you used to train the model with. You can use the quantize command to do this.", + "text": "Configuring QAT in Axolotl\nTo enable QAT in axolotl, add the following to your configuration file:\nqat:\n activation_dtype: # Optional[str] = \"int8\". Fake quantization layout to use for activation quantization. Valid options are \"int4\", \"int8\", \"float8\"\n weight_dtype: # Optional[str] = \"int8\". Fake quantization layout to use for weight quantization. Valid options are \"int4\", \"fp8\", and \"nvfp4\".\n group_size: # Optional[int] = 32. The number of elements in each group for per-group fake quantization\n fake_quant_after_n_steps: # Optional[int] = None. The number of steps to apply fake quantization after\nWe support the following quantization schemas:\n- Int4WeightOnly (requires the fbgemm-gpu extra when installing Axolotl)\n- Int8DynamicActivationInt4Weight\n- Float8DynamicActivationFloat8Weight\n- Float8DynamicActivationInt4Weight\n- NVFP4\nOnce you have finished training, you must quantize your model by using the same quantization configuration which you used to train the model with. You can use the quantize command to do this.", "crumbs": [ "How To Guides", "Quantization Aware Training (QAT)" diff --git a/sitemap.xml b/sitemap.xml index 7f3486590..13801b961 100644 --- a/sitemap.xml +++ b/sitemap.xml @@ -2,794 +2,794 @@ https://docs.axolotl.ai/examples/colab-notebooks/colab-axolotl-example.html - 2025-09-16T18:53:10.076Z + 2025-09-17T09:38:24.776Z https://docs.axolotl.ai/src/axolotl/integrations/LICENSE.html - 2025-09-16T18:53:10.093Z + 2025-09-17T09:38:24.792Z https://docs.axolotl.ai/FAQS.html - 2025-09-16T18:53:10.066Z + 2025-09-17T09:38:24.766Z https://docs.axolotl.ai/docs/unsloth.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.772Z https://docs.axolotl.ai/docs/reward_modelling.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/docker.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/batch_vs_grad.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.767Z https://docs.axolotl.ai/docs/streaming.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.772Z https://docs.axolotl.ai/docs/nccl.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/quantize.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/nd_parallelism.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/custom_integrations.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.767Z https://docs.axolotl.ai/docs/debugging.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/gradient_checkpointing.html - 2025-09-16T18:53:10.069Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/multimodal.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/dataset_loading.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/api/integrations.spectrum.args.html - 2025-09-16T18:56:42.167Z + 2025-09-17T09:41:55.684Z https://docs.axolotl.ai/docs/api/loaders.adapter.html - 2025-09-16T18:56:41.303Z + 2025-09-17T09:41:54.825Z https://docs.axolotl.ai/docs/api/core.builders.base.html - 2025-09-16T18:56:40.934Z + 2025-09-17T09:41:54.462Z https://docs.axolotl.ai/docs/api/cli.merge_sharded_fsdp_weights.html - 2025-09-16T18:56:41.138Z + 2025-09-17T09:41:54.662Z https://docs.axolotl.ai/docs/api/cli.cloud.modal_.html - 2025-09-16T18:56:41.168Z + 2025-09-17T09:41:54.692Z https://docs.axolotl.ai/docs/api/monkeypatch.stablelm_attn_hijack_flash.html - 2025-09-16T18:56:41.723Z + 2025-09-17T09:41:55.244Z https://docs.axolotl.ai/docs/api/integrations.kd.trainer.html - 2025-09-16T18:56:42.157Z + 2025-09-17T09:41:55.674Z https://docs.axolotl.ai/docs/api/cli.delinearize_llama4.html - 2025-09-16T18:56:41.103Z + 2025-09-17T09:41:54.627Z https://docs.axolotl.ai/docs/api/cli.evaluate.html - 2025-09-16T18:56:41.049Z + 2025-09-17T09:41:54.575Z https://docs.axolotl.ai/docs/api/utils.schemas.model.html - 2025-09-16T18:56:41.936Z + 2025-09-17T09:41:55.454Z https://docs.axolotl.ai/docs/api/cli.art.html - 2025-09-16T18:56:41.073Z + 2025-09-17T09:41:54.598Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.llama3.html - 2025-09-16T18:56:41.487Z + 2025-09-17T09:41:55.009Z https://docs.axolotl.ai/docs/api/core.chat.format.llama3x.html - 2025-09-16T18:56:40.983Z + 2025-09-17T09:41:54.510Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.chatml.html - 2025-09-16T18:56:41.518Z + 2025-09-17T09:41:55.040Z https://docs.axolotl.ai/docs/api/cli.args.html - 2025-09-16T18:56:41.070Z + 2025-09-17T09:41:54.594Z https://docs.axolotl.ai/docs/api/monkeypatch.gradient_checkpointing.offload_cpu.html - 2025-09-16T18:56:41.749Z + 2025-09-17T09:41:55.269Z https://docs.axolotl.ai/docs/api/evaluate.html - 2025-09-16T18:56:40.857Z + 2025-09-17T09:41:54.384Z https://docs.axolotl.ai/docs/api/prompt_strategies.input_output.html - 2025-09-16T18:56:41.449Z + 2025-09-17T09:41:54.971Z https://docs.axolotl.ai/docs/api/cli.vllm_serve.html - 2025-09-16T18:56:41.158Z + 2025-09-17T09:41:54.682Z https://docs.axolotl.ai/docs/api/cli.quantize.html - 2025-09-16T18:56:41.151Z + 2025-09-17T09:41:54.675Z https://docs.axolotl.ai/docs/api/cli.utils.sweeps.html - 2025-09-16T18:56:41.198Z + 2025-09-17T09:41:54.722Z https://docs.axolotl.ai/docs/api/core.builders.rl.html - 2025-09-16T18:56:40.944Z + 2025-09-17T09:41:54.472Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.chatml.html - 2025-09-16T18:56:41.497Z + 2025-09-17T09:41:55.019Z https://docs.axolotl.ai/docs/api/prompt_strategies.metharme.html - 2025-09-16T18:56:41.460Z + 2025-09-17T09:41:54.982Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.passthrough.html - 2025-09-16T18:56:41.502Z + 2025-09-17T09:41:55.023Z https://docs.axolotl.ai/docs/api/core.datasets.chat.html - 2025-09-16T18:56:40.989Z + 2025-09-17T09:41:54.517Z https://docs.axolotl.ai/docs/api/utils.distributed.html - 2025-09-16T18:56:41.871Z + 2025-09-17T09:41:55.390Z https://docs.axolotl.ai/docs/api/integrations.base.html - 2025-09-16T18:56:42.144Z + 2025-09-17T09:41:55.662Z https://docs.axolotl.ai/docs/api/index.html - 2025-09-16T18:56:40.787Z + 2025-09-17T09:41:54.315Z https://docs.axolotl.ai/docs/api/monkeypatch.gradient_checkpointing.offload_disk.html - 2025-09-16T18:56:41.775Z + 2025-09-17T09:41:55.295Z https://docs.axolotl.ai/docs/api/utils.quantization.html - 2025-09-16T18:56:41.913Z + 2025-09-17T09:41:55.431Z https://docs.axolotl.ai/docs/api/utils.dict.html - 2025-09-16T18:56:41.877Z + 2025-09-17T09:41:55.396Z https://docs.axolotl.ai/docs/api/core.trainers.dpo.trainer.html - 2025-09-16T18:56:41.253Z + 2025-09-17T09:41:54.776Z https://docs.axolotl.ai/docs/api/utils.schemas.config.html - 2025-09-16T18:56:41.929Z + 2025-09-17T09:41:55.447Z https://docs.axolotl.ai/docs/api/prompt_strategies.orcamini.html - 2025-09-16T18:56:41.464Z + 2025-09-17T09:41:54.986Z https://docs.axolotl.ai/docs/api/utils.collators.mamba.html - 2025-09-16T18:56:42.213Z + 2025-09-17T09:41:55.728Z https://docs.axolotl.ai/docs/api/datasets.html - 2025-09-16T18:56:40.863Z + 2025-09-17T09:41:54.391Z https://docs.axolotl.ai/docs/api/convert.html - 2025-09-16T18:56:40.876Z + 2025-09-17T09:41:54.404Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.llama3.html - 2025-09-16T18:56:41.510Z + 2025-09-17T09:41:55.032Z https://docs.axolotl.ai/docs/api/utils.tokenization.html - 2025-09-16T18:56:41.782Z + 2025-09-17T09:41:55.302Z https://docs.axolotl.ai/docs/api/core.trainers.base.html - 2025-09-16T18:56:41.225Z + 2025-09-17T09:41:54.748Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_expand_mask.html - 2025-09-16T18:56:41.677Z + 2025-09-17T09:41:55.198Z https://docs.axolotl.ai/docs/api/loaders.model.html - 2025-09-16T18:56:41.288Z + 2025-09-17T09:41:54.810Z https://docs.axolotl.ai/docs/api/logging_config.html - 2025-09-16T18:56:40.928Z + 2025-09-17T09:41:54.456Z https://docs.axolotl.ai/docs/api/core.chat.format.shared.html - 2025-09-16T18:56:40.984Z + 2025-09-17T09:41:54.512Z https://docs.axolotl.ai/docs/api/models.mamba.modeling_mamba.html - 2025-09-16T18:56:42.186Z + 2025-09-17T09:41:55.704Z https://docs.axolotl.ai/docs/api/utils.schemas.enums.html - 2025-09-16T18:56:42.000Z + 2025-09-17T09:41:55.518Z https://docs.axolotl.ai/docs/api/utils.callbacks.profiler.html - 2025-09-16T18:56:42.269Z + 2025-09-17T09:41:55.783Z https://docs.axolotl.ai/docs/api/utils.bench.html - 2025-09-16T18:56:41.798Z + 2025-09-17T09:41:55.317Z https://docs.axolotl.ai/docs/api/utils.samplers.multipack.html - 2025-09-16T18:56:42.259Z + 2025-09-17T09:41:55.773Z https://docs.axolotl.ai/docs/api/core.trainers.mamba.html - 2025-09-16T18:56:41.246Z + 2025-09-17T09:41:54.769Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_chat.html - 2025-09-16T18:56:41.403Z + 2025-09-17T09:41:54.925Z https://docs.axolotl.ai/docs/api/monkeypatch.btlm_attn_hijack_flash.html - 2025-09-16T18:56:41.716Z + 2025-09-17T09:41:55.236Z https://docs.axolotl.ai/docs/api/core.chat.messages.html - 2025-09-16T18:56:40.980Z + 2025-09-17T09:41:54.507Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_patch_multipack.html - 2025-09-16T18:56:41.717Z + 2025-09-17T09:41:55.238Z https://docs.axolotl.ai/docs/api/monkeypatch.mixtral.html - 2025-09-16T18:56:41.745Z + 2025-09-17T09:41:55.265Z https://docs.axolotl.ai/docs/api/core.trainers.mixins.scheduler.html - 2025-09-16T18:56:41.330Z + 2025-09-17T09:41:54.852Z https://docs.axolotl.ai/docs/api/loaders.patch_manager.html - 2025-09-16T18:56:41.313Z + 2025-09-17T09:41:54.835Z https://docs.axolotl.ai/docs/api/utils.optimizers.adopt.html - 2025-09-16T18:56:41.884Z + 2025-09-17T09:41:55.403Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_attn_hijack_xformers.html - 2025-09-16T18:56:41.669Z + 2025-09-17T09:41:55.190Z https://docs.axolotl.ai/docs/api/utils.callbacks.perplexity.html - 2025-09-16T18:56:42.266Z + 2025-09-17T09:41:55.780Z https://docs.axolotl.ai/docs/api/loaders.tokenizer.html - 2025-09-16T18:56:41.296Z + 2025-09-17T09:41:54.819Z https://docs.axolotl.ai/docs/api/prompt_strategies.pygmalion.html - 2025-09-16T18:56:41.471Z + 2025-09-17T09:41:54.993Z https://docs.axolotl.ai/docs/api/utils.lora.html - 2025-09-16T18:56:41.789Z + 2025-09-17T09:41:55.308Z https://docs.axolotl.ai/docs/api/prompt_strategies.user_defined.html - 2025-09-16T18:56:41.424Z + 2025-09-17T09:41:54.946Z https://docs.axolotl.ai/docs/api/core.trainers.mixins.rng_state_loader.html - 2025-09-16T18:56:41.323Z + 2025-09-17T09:41:54.846Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.user_defined.html - 2025-09-16T18:56:41.500Z + 2025-09-17T09:41:55.022Z https://docs.axolotl.ai/docs/api/kernels.swiglu.html - 2025-09-16T18:56:41.653Z + 2025-09-17T09:41:55.174Z https://docs.axolotl.ai/docs/api/utils.schemas.multimodal.html - 2025-09-16T18:56:41.978Z + 2025-09-17T09:41:55.497Z https://docs.axolotl.ai/docs/api/cli.main.html - 2025-09-16T18:56:41.032Z + 2025-09-17T09:41:54.559Z https://docs.axolotl.ai/docs/api/utils.ctx_managers.sequence_parallel.html - 2025-09-16T18:56:41.354Z + 2025-09-17T09:41:54.876Z https://docs.axolotl.ai/docs/api/monkeypatch.multipack.html - 2025-09-16T18:56:41.672Z + 2025-09-17T09:41:55.193Z https://docs.axolotl.ai/docs/api/core.chat.format.chatml.html - 2025-09-16T18:56:40.981Z + 2025-09-17T09:41:54.509Z https://docs.axolotl.ai/docs/api/cli.cloud.base.html - 2025-09-16T18:56:41.162Z + 2025-09-17T09:41:54.686Z https://docs.axolotl.ai/docs/api/core.trainers.trl.html - 2025-09-16T18:56:41.240Z + 2025-09-17T09:41:54.763Z https://docs.axolotl.ai/docs/mixed_precision.html - 2025-09-16T18:53:10.071Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/installation.html - 2025-09-16T18:53:10.071Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/mac.html - 2025-09-16T18:53:10.071Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/sequence_parallelism.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.772Z https://docs.axolotl.ai/docs/faq.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/dataset-formats/tokenized.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/dataset-formats/stepwise_supervised.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/dataset-formats/pretraining.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/dataset-formats/template_free.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/dataset-formats/index.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.767Z https://docs.axolotl.ai/docs/dataset-formats/conversation.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.767Z https://docs.axolotl.ai/docs/dataset-formats/inst_tune.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/lr_groups.html - 2025-09-16T18:53:10.071Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/inference.html - 2025-09-16T18:53:10.071Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/lora_optims.html - 2025-09-16T18:53:10.071Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/multipack.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/amd_hpc.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.767Z https://docs.axolotl.ai/docs/api/monkeypatch.trainer_fsdp_optim.html - 2025-09-16T18:56:41.727Z + 2025-09-17T09:41:55.247Z https://docs.axolotl.ai/docs/api/cli.inference.html - 2025-09-16T18:56:41.118Z + 2025-09-17T09:41:54.642Z https://docs.axolotl.ai/docs/api/utils.collators.mm_chat.html - 2025-09-16T18:56:42.218Z + 2025-09-17T09:41:55.732Z https://docs.axolotl.ai/docs/api/utils.model_shard_quant.html - 2025-09-16T18:56:41.794Z + 2025-09-17T09:41:55.314Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_attn_hijack_flash.html - 2025-09-16T18:56:41.668Z + 2025-09-17T09:41:55.189Z https://docs.axolotl.ai/docs/api/integrations.grokfast.optimizer.html - 2025-09-16T18:56:42.149Z + 2025-09-17T09:41:55.666Z https://docs.axolotl.ai/docs/api/utils.schemas.training.html - 2025-09-16T18:56:41.943Z + 2025-09-17T09:41:55.461Z https://docs.axolotl.ai/docs/api/utils.schemas.integrations.html - 2025-09-16T18:56:41.991Z + 2025-09-17T09:41:55.509Z https://docs.axolotl.ai/docs/api/utils.schemas.utils.html - 2025-09-16T18:56:42.005Z + 2025-09-17T09:41:55.523Z https://docs.axolotl.ai/docs/api/cli.checks.html - 2025-09-16T18:56:41.080Z + 2025-09-17T09:41:54.604Z https://docs.axolotl.ai/docs/api/utils.callbacks.qat.html - 2025-09-16T18:56:42.285Z + 2025-09-17T09:41:55.799Z https://docs.axolotl.ai/docs/api/monkeypatch.mistral_attn_hijack_flash.html - 2025-09-16T18:56:41.671Z + 2025-09-17T09:41:55.192Z https://docs.axolotl.ai/docs/api/utils.data.sft.html - 2025-09-16T18:56:41.892Z + 2025-09-17T09:41:55.411Z https://docs.axolotl.ai/docs/api/utils.chat_templates.html - 2025-09-16T18:56:41.784Z + 2025-09-17T09:41:55.303Z https://docs.axolotl.ai/docs/api/monkeypatch.utils.html - 2025-09-16T18:56:41.714Z + 2025-09-17T09:41:55.235Z https://docs.axolotl.ai/docs/api/cli.utils.fetch.html - 2025-09-16T18:56:41.186Z + 2025-09-17T09:41:54.710Z https://docs.axolotl.ai/docs/api/core.builders.causal.html - 2025-09-16T18:56:40.939Z + 2025-09-17T09:41:54.467Z https://docs.axolotl.ai/docs/api/utils.schemas.trl.html - 2025-09-16T18:56:41.973Z + 2025-09-17T09:41:55.491Z https://docs.axolotl.ai/docs/api/utils.data.streaming.html - 2025-09-16T18:56:41.886Z + 2025-09-17T09:41:55.405Z https://docs.axolotl.ai/docs/api/monkeypatch.unsloth_.html - 2025-09-16T18:56:41.735Z + 2025-09-17T09:41:55.255Z https://docs.axolotl.ai/docs/api/prompt_strategies.base.html - 2025-09-16T18:56:41.355Z + 2025-09-17T09:41:54.878Z https://docs.axolotl.ai/docs/api/cli.merge_lora.html - 2025-09-16T18:56:41.126Z + 2025-09-17T09:41:54.650Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_instruct.html - 2025-09-16T18:56:41.404Z + 2025-09-17T09:41:54.926Z https://docs.axolotl.ai/docs/api/loaders.constants.html - 2025-09-16T18:56:41.314Z + 2025-09-17T09:41:54.837Z https://docs.axolotl.ai/docs/api/monkeypatch.transformers_fa_utils.html - 2025-09-16T18:56:41.733Z + 2025-09-17T09:41:55.253Z https://docs.axolotl.ai/docs/api/monkeypatch.relora.html - 2025-09-16T18:56:41.676Z + 2025-09-17T09:41:55.197Z https://docs.axolotl.ai/docs/api/utils.schedulers.html - 2025-09-16T18:56:41.851Z + 2025-09-17T09:41:55.370Z https://docs.axolotl.ai/docs/api/kernels.lora.html - 2025-09-16T18:56:41.632Z + 2025-09-17T09:41:55.153Z https://docs.axolotl.ai/docs/api/utils.collators.batching.html - 2025-09-16T18:56:42.209Z + 2025-09-17T09:41:55.724Z https://docs.axolotl.ai/docs/api/prompt_strategies.completion.html - 2025-09-16T18:56:41.443Z + 2025-09-17T09:41:54.965Z https://docs.axolotl.ai/docs/api/core.trainers.grpo.trainer.html - 2025-09-16T18:56:41.264Z + 2025-09-17T09:41:54.787Z https://docs.axolotl.ai/docs/api/common.architectures.html - 2025-09-16T18:56:42.168Z + 2025-09-17T09:41:55.686Z https://docs.axolotl.ai/docs/api/cli.utils.html - 2025-09-16T18:56:41.169Z + 2025-09-17T09:41:54.693Z https://docs.axolotl.ai/docs/api/prompt_strategies.orpo.chat_template.html - 2025-09-16T18:56:41.540Z + 2025-09-17T09:41:55.061Z https://docs.axolotl.ai/docs/api/core.training_args.html - 2025-09-16T18:56:40.957Z + 2025-09-17T09:41:54.484Z https://docs.axolotl.ai/docs/api/utils.schemas.datasets.html - 2025-09-16T18:56:41.961Z + 2025-09-17T09:41:55.479Z https://docs.axolotl.ai/docs/api/prompt_strategies.chat_template.html - 2025-09-16T18:56:41.389Z + 2025-09-17T09:41:54.911Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_w_system.html - 2025-09-16T18:56:41.416Z + 2025-09-17T09:41:54.938Z https://docs.axolotl.ai/docs/api/integrations.cut_cross_entropy.args.html - 2025-09-16T18:56:42.148Z + 2025-09-17T09:41:55.665Z https://docs.axolotl.ai/docs/api/common.const.html - 2025-09-16T18:56:42.170Z + 2025-09-17T09:41:55.687Z https://docs.axolotl.ai/docs/api/monkeypatch.lora_kernels.html - 2025-09-16T18:56:41.706Z + 2025-09-17T09:41:55.227Z https://docs.axolotl.ai/docs/api/kernels.quantize.html - 2025-09-16T18:56:41.661Z + 2025-09-17T09:41:55.182Z https://docs.axolotl.ai/docs/api/cli.config.html - 2025-09-16T18:56:41.098Z + 2025-09-17T09:41:54.622Z https://docs.axolotl.ai/docs/api/prompt_strategies.llama2_chat.html - 2025-09-16T18:56:41.437Z + 2025-09-17T09:41:54.959Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.zephyr.html - 2025-09-16T18:56:41.499Z + 2025-09-17T09:41:55.020Z https://docs.axolotl.ai/docs/api/kernels.geglu.html - 2025-09-16T18:56:41.643Z + 2025-09-17T09:41:55.164Z https://docs.axolotl.ai/docs/api/utils.freeze.html - 2025-09-16T18:56:41.806Z + 2025-09-17T09:41:55.325Z https://docs.axolotl.ai/docs/api/utils.callbacks.mlflow_.html - 2025-09-16T18:56:42.275Z + 2025-09-17T09:41:55.788Z https://docs.axolotl.ai/docs/api/core.trainers.grpo.sampler.html - 2025-09-16T18:56:41.276Z + 2025-09-17T09:41:54.799Z https://docs.axolotl.ai/docs/api/cli.utils.train.html - 2025-09-16T18:56:41.210Z + 2025-09-17T09:41:54.734Z https://docs.axolotl.ai/docs/api/cli.utils.load.html - 2025-09-16T18:56:41.192Z + 2025-09-17T09:41:54.716Z https://docs.axolotl.ai/docs/api/utils.collators.core.html - 2025-09-16T18:56:42.188Z + 2025-09-17T09:41:55.705Z https://docs.axolotl.ai/docs/api/integrations.liger.args.html - 2025-09-16T18:56:42.160Z + 2025-09-17T09:41:55.677Z https://docs.axolotl.ai/docs/api/cli.preprocess.html - 2025-09-16T18:56:41.146Z + 2025-09-17T09:41:54.671Z https://docs.axolotl.ai/docs/api/kernels.utils.html - 2025-09-16T18:56:41.662Z + 2025-09-17T09:41:55.183Z https://docs.axolotl.ai/docs/api/utils.callbacks.lisa.html - 2025-09-16T18:56:42.271Z + 2025-09-17T09:41:55.785Z https://docs.axolotl.ai/docs/api/monkeypatch.data.batch_dataset_fetcher.html - 2025-09-16T18:56:41.744Z + 2025-09-17T09:41:55.264Z https://docs.axolotl.ai/docs/api/core.trainers.mixins.optimizer.html - 2025-09-16T18:56:41.320Z + 2025-09-17T09:41:54.842Z https://docs.axolotl.ai/docs/api/loaders.processor.html - 2025-09-16T18:56:41.297Z + 2025-09-17T09:41:54.820Z https://docs.axolotl.ai/docs/api/core.trainers.utils.html - 2025-09-16T18:56:41.278Z + 2025-09-17T09:41:54.801Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.chat_template.html - 2025-09-16T18:56:41.476Z + 2025-09-17T09:41:54.998Z https://docs.axolotl.ai/docs/api/prompt_tokenizers.html - 2025-09-16T18:56:40.918Z + 2025-09-17T09:41:54.446Z https://docs.axolotl.ai/docs/api/utils.callbacks.comet_.html - 2025-09-16T18:56:42.278Z + 2025-09-17T09:41:55.792Z https://docs.axolotl.ai/docs/api/core.datasets.transforms.chat_builder.html - 2025-09-16T18:56:40.997Z + 2025-09-17T09:41:54.525Z https://docs.axolotl.ai/docs/api/utils.trainer.html - 2025-09-16T18:56:41.823Z + 2025-09-17T09:41:55.343Z https://docs.axolotl.ai/docs/api/cli.train.html - 2025-09-16T18:56:41.041Z + 2025-09-17T09:41:54.567Z https://docs.axolotl.ai/docs/api/integrations.lm_eval.args.html - 2025-09-16T18:56:42.163Z + 2025-09-17T09:41:55.681Z https://docs.axolotl.ai/docs/api/prompt_strategies.messages.chat.html - 2025-09-16T18:56:41.475Z + 2025-09-17T09:41:54.997Z https://docs.axolotl.ai/docs/api/utils.schemas.peft.html - 2025-09-16T18:56:41.970Z + 2025-09-17T09:41:55.488Z https://docs.axolotl.ai/docs/api/prompt_strategies.bradley_terry.llama3.html - 2025-09-16T18:56:41.544Z + 2025-09-17T09:41:55.065Z https://docs.axolotl.ai/docs/api/cli.utils.args.html - 2025-09-16T18:56:41.181Z + 2025-09-17T09:41:54.705Z https://docs.axolotl.ai/docs/api/prompt_strategies.stepwise_supervised.html - 2025-09-16T18:56:41.453Z + 2025-09-17T09:41:54.975Z https://docs.axolotl.ai/docs/api/common.datasets.html - 2025-09-16T18:56:42.185Z + 2025-09-17T09:41:55.703Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.user_defined.html - 2025-09-16T18:56:41.519Z + 2025-09-17T09:41:55.041Z https://docs.axolotl.ai/docs/api/train.html - 2025-09-16T18:56:40.846Z + 2025-09-17T09:41:54.374Z https://docs.axolotl.ai/docs/multi-node.html - 2025-09-16T18:53:10.071Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/input_output.html - 2025-09-16T18:53:10.071Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/ray-integration.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/getting-started.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/optimizers.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/multi-gpu.html - 2025-09-16T18:53:10.071Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/docs/dataset_preprocessing.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/torchao.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.772Z https://docs.axolotl.ai/docs/config-reference.html - 2025-09-16T18:56:57.625Z + 2025-09-17T09:42:11.046Z https://docs.axolotl.ai/docs/rlhf.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.772Z https://docs.axolotl.ai/docs/cli.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.767Z https://docs.axolotl.ai/docs/fsdp_qlora.html - 2025-09-16T18:53:10.068Z + 2025-09-17T09:38:24.768Z https://docs.axolotl.ai/docs/qat.html - 2025-09-16T18:53:10.072Z + 2025-09-17T09:38:24.771Z https://docs.axolotl.ai/src/axolotl/integrations/cut_cross_entropy/ACKNOWLEDGEMENTS.html - 2025-09-16T18:53:10.093Z + 2025-09-17T09:38:24.792Z https://docs.axolotl.ai/index.html - 2025-09-16T18:53:10.088Z + 2025-09-17T09:38:24.788Z