From 421eea620c0db6349bd7db0b05d06106ba44c23b Mon Sep 17 00:00:00 2001 From: Quarto GHA Workflow Runner Date: Tue, 16 Sep 2025 18:58:53 +0000 Subject: [PATCH] Built site for gh-pages --- .nojekyll | 2 +- FAQS.html | 27 +- docs/amd_hpc.html | 55 +-- docs/api/cli.args.html | 49 ++- docs/api/cli.art.html | 29 +- docs/api/cli.checks.html | 31 +- docs/api/cli.cloud.base.html | 29 +- docs/api/cli.cloud.modal_.html | 31 +- docs/api/cli.config.html | 35 +- docs/api/cli.delinearize_llama4.html | 29 +- docs/api/cli.evaluate.html | 31 +- docs/api/cli.inference.html | 35 +- docs/api/cli.main.html | 45 +- docs/api/cli.merge_lora.html | 31 +- docs/api/cli.merge_sharded_fsdp_weights.html | 35 +- docs/api/cli.preprocess.html | 31 +- docs/api/cli.quantize.html | 29 +- docs/api/cli.train.html | 31 +- docs/api/cli.utils.args.html | 33 +- docs/api/cli.utils.fetch.html | 29 +- docs/api/cli.utils.html | 27 +- docs/api/cli.utils.load.html | 29 +- docs/api/cli.utils.sweeps.html | 29 +- docs/api/cli.utils.train.html | 35 +- docs/api/cli.vllm_serve.html | 33 +- docs/api/common.architectures.html | 27 +- docs/api/common.const.html | 27 +- docs/api/common.datasets.html | 37 +- docs/api/convert.html | 41 +- docs/api/core.builders.base.html | 31 +- docs/api/core.builders.causal.html | 31 +- docs/api/core.builders.rl.html | 29 +- docs/api/core.chat.format.chatml.html | 27 +- docs/api/core.chat.format.llama3x.html | 27 +- docs/api/core.chat.format.shared.html | 27 +- docs/api/core.chat.messages.html | 51 ++- docs/api/core.datasets.chat.html | 31 +- ...core.datasets.transforms.chat_builder.html | 31 +- docs/api/core.trainers.base.html | 39 +- docs/api/core.trainers.dpo.trainer.html | 31 +- docs/api/core.trainers.grpo.sampler.html | 33 +- docs/api/core.trainers.grpo.trainer.html | 37 +- docs/api/core.trainers.mamba.html | 31 +- docs/api/core.trainers.mixins.optimizer.html | 31 +- ...core.trainers.mixins.rng_state_loader.html | 29 +- docs/api/core.trainers.mixins.scheduler.html | 33 +- docs/api/core.trainers.trl.html | 37 +- docs/api/core.trainers.utils.html | 27 +- docs/api/core.training_args.html | 39 +- docs/api/datasets.html | 31 +- docs/api/evaluate.html | 31 +- docs/api/index.html | 27 +- docs/api/integrations.base.html | 119 +++--- .../integrations.cut_cross_entropy.args.html | 29 +- docs/api/integrations.grokfast.optimizer.html | 27 +- docs/api/integrations.kd.trainer.html | 33 +- docs/api/integrations.liger.args.html | 29 +- docs/api/integrations.lm_eval.args.html | 29 +- docs/api/integrations.spectrum.args.html | 29 +- docs/api/kernels.geglu.html | 31 +- docs/api/kernels.lora.html | 61 +-- docs/api/kernels.quantize.html | 29 +- docs/api/kernels.swiglu.html | 31 +- docs/api/kernels.utils.html | 27 +- docs/api/loaders.adapter.html | 31 +- docs/api/loaders.constants.html | 27 +- docs/api/loaders.model.html | 33 +- docs/api/loaders.patch_manager.html | 35 +- docs/api/loaders.processor.html | 27 +- docs/api/loaders.tokenizer.html | 33 +- docs/api/logging_config.html | 35 +- docs/api/models.mamba.modeling_mamba.html | 27 +- .../monkeypatch.btlm_attn_hijack_flash.html | 27 +- ...onkeypatch.data.batch_dataset_fetcher.html | 35 +- ...ch.gradient_checkpointing.offload_cpu.html | 31 +- ...h.gradient_checkpointing.offload_disk.html | 67 +-- .../monkeypatch.llama_attn_hijack_flash.html | 31 +- ...onkeypatch.llama_attn_hijack_xformers.html | 27 +- docs/api/monkeypatch.llama_expand_mask.html | 27 +- .../monkeypatch.llama_patch_multipack.html | 27 +- docs/api/monkeypatch.lora_kernels.html | 41 +- ...monkeypatch.mistral_attn_hijack_flash.html | 27 +- docs/api/monkeypatch.mixtral.html | 27 +- docs/api/monkeypatch.multipack.html | 27 +- docs/api/monkeypatch.relora.html | 29 +- ...onkeypatch.stablelm_attn_hijack_flash.html | 31 +- docs/api/monkeypatch.trainer_fsdp_optim.html | 29 +- .../monkeypatch.transformers_fa_utils.html | 31 +- docs/api/monkeypatch.unsloth_.html | 27 +- docs/api/monkeypatch.utils.html | 33 +- docs/api/prompt_strategies.alpaca_chat.html | 43 +- .../prompt_strategies.alpaca_instruct.html | 27 +- .../prompt_strategies.alpaca_w_system.html | 43 +- docs/api/prompt_strategies.base.html | 27 +- ...rompt_strategies.bradley_terry.llama3.html | 29 +- docs/api/prompt_strategies.chat_template.html | 61 +-- docs/api/prompt_strategies.completion.html | 33 +- .../prompt_strategies.dpo.chat_template.html | 27 +- docs/api/prompt_strategies.dpo.chatml.html | 35 +- docs/api/prompt_strategies.dpo.llama3.html | 35 +- .../prompt_strategies.dpo.passthrough.html | 27 +- .../prompt_strategies.dpo.user_defined.html | 27 +- docs/api/prompt_strategies.dpo.zephyr.html | 27 +- docs/api/prompt_strategies.input_output.html | 33 +- docs/api/prompt_strategies.kto.chatml.html | 33 +- docs/api/prompt_strategies.kto.llama3.html | 33 +- .../prompt_strategies.kto.user_defined.html | 27 +- docs/api/prompt_strategies.llama2_chat.html | 41 +- docs/api/prompt_strategies.messages.chat.html | 31 +- docs/api/prompt_strategies.metharme.html | 33 +- docs/api/prompt_strategies.orcamini.html | 31 +- .../prompt_strategies.orpo.chat_template.html | 53 ++- docs/api/prompt_strategies.pygmalion.html | 33 +- ...prompt_strategies.stepwise_supervised.html | 31 +- docs/api/prompt_strategies.user_defined.html | 35 +- docs/api/prompt_tokenizers.html | 81 ++-- docs/api/train.html | 51 ++- docs/api/utils.bench.html | 29 +- docs/api/utils.callbacks.comet_.html | 29 +- docs/api/utils.callbacks.lisa.html | 27 +- docs/api/utils.callbacks.mlflow_.html | 29 +- docs/api/utils.callbacks.perplexity.html | 31 +- docs/api/utils.callbacks.profiler.html | 31 +- docs/api/utils.callbacks.qat.html | 31 +- docs/api/utils.chat_templates.html | 27 +- docs/api/utils.collators.batching.html | 43 +- docs/api/utils.collators.core.html | 27 +- docs/api/utils.collators.mamba.html | 29 +- docs/api/utils.collators.mm_chat.html | 31 +- .../utils.ctx_managers.sequence_parallel.html | 45 +- docs/api/utils.data.sft.html | 29 +- docs/api/utils.data.streaming.html | 27 +- docs/api/utils.dict.html | 31 +- docs/api/utils.distributed.html | 45 +- docs/api/utils.freeze.html | 35 +- docs/api/utils.lora.html | 29 +- docs/api/utils.model_shard_quant.html | 31 +- docs/api/utils.optimizers.adopt.html | 31 +- docs/api/utils.quantization.html | 41 +- docs/api/utils.samplers.multipack.html | 55 +-- docs/api/utils.schedulers.html | 51 ++- docs/api/utils.schemas.config.html | 31 +- docs/api/utils.schemas.datasets.html | 45 +- docs/api/utils.schemas.enums.html | 35 +- docs/api/utils.schemas.integrations.html | 39 +- docs/api/utils.schemas.model.html | 33 +- docs/api/utils.schemas.multimodal.html | 33 +- docs/api/utils.schemas.peft.html | 35 +- docs/api/utils.schemas.training.html | 33 +- docs/api/utils.schemas.trl.html | 29 +- docs/api/utils.schemas.utils.html | 29 +- docs/api/utils.tokenization.html | 35 +- docs/api/utils.trainer.html | 39 +- docs/batch_vs_grad.html | 27 +- docs/cli.html | 105 ++--- docs/config-reference.html | 31 +- docs/custom_integrations.html | 105 ++--- docs/dataset-formats/conversation.html | 87 ++-- docs/dataset-formats/index.html | 103 ++--- docs/dataset-formats/inst_tune.html | 73 ++-- docs/dataset-formats/pretraining.html | 37 +- docs/dataset-formats/stepwise_supervised.html | 31 +- docs/dataset-formats/template_free.html | 57 +-- docs/dataset-formats/tokenized.html | 35 +- docs/dataset_loading.html | 67 +-- docs/dataset_preprocessing.html | 27 +- docs/debugging.html | 57 +-- docs/docker.html | 33 +- docs/faq.html | 35 +- docs/fsdp_qlora.html | 27 +- docs/getting-started.html | 57 +-- docs/gradient_checkpointing.html | 33 +- docs/inference.html | 59 +-- docs/input_output.html | 27 +- docs/installation.html | 69 +-- docs/lora_optims.html | 39 +- docs/lr_groups.html | 31 +- docs/mac.html | 27 +- docs/mixed_precision.html | 45 +- docs/multi-gpu.html | 51 ++- docs/multi-node.html | 43 +- docs/multimodal.html | 77 ++-- docs/multipack.html | 27 +- docs/nccl.html | 37 +- docs/nd_parallelism.html | 31 +- docs/optimizers.html | 43 +- docs/qat.html | 31 +- docs/quantize.html | 37 +- docs/ray-integration.html | 39 +- docs/reward_modelling.html | 39 +- docs/rlhf.html | 193 +++++---- docs/sequence_parallelism.html | 35 +- docs/streaming.html | 41 +- docs/torchao.html | 31 +- docs/unsloth.html | 37 +- .../colab-axolotl-example.html | 75 ++-- index.html | 41 +- search.json | 4 +- ...p-e9895ec3143e9833a687747e8d39d226.min.css | 12 - ...p-f9d679a32da2b248d4ca48a0e58e089e.min.css | 12 + site_libs/bootstrap/bootstrap-icons.css | 34 +- site_libs/bootstrap/bootstrap-icons.woff | Bin 176200 -> 180288 bytes site_libs/quarto-html/axe/axe-check.js | 145 +++++++ ...dark-b651517ce65839d647a86e2780455cfb.css} | 2 +- site_libs/quarto-html/quarto.js | 2 + site_libs/quarto-search/autocomplete.umd.js | 4 +- sitemap.xml | 396 +++++++++--------- src/axolotl/integrations/LICENSE.html | 27 +- .../cut_cross_entropy/ACKNOWLEDGEMENTS.html | 27 +- 209 files changed, 4822 insertions(+), 3261 deletions(-) delete mode 100644 site_libs/bootstrap/bootstrap-e9895ec3143e9833a687747e8d39d226.min.css create mode 100644 site_libs/bootstrap/bootstrap-f9d679a32da2b248d4ca48a0e58e089e.min.css create mode 100644 site_libs/quarto-html/axe/axe-check.js rename site_libs/quarto-html/{quarto-syntax-highlighting-dark-befe23ebd2f54d8af2c8a89d1a1611f1.css => quarto-syntax-highlighting-dark-b651517ce65839d647a86e2780455cfb.css} (98%) diff --git a/.nojekyll b/.nojekyll index b0532e970..e1b123edd 100644 --- a/.nojekyll +++ b/.nojekyll @@ -1 +1 @@ -6fb18589 \ No newline at end of file +1d43a3d7 \ No newline at end of file diff --git a/FAQS.html b/FAQS.html index 3aa04ba1c..29023c65a 100644 --- a/FAQS.html +++ b/FAQS.html @@ -2,7 +2,7 @@ - + @@ -32,14 +32,15 @@ ul.task-list li input[type="checkbox"] { + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + diff --git a/docs/multi-gpu.html b/docs/multi-gpu.html index 9c110ec36..360512d02 100644 --- a/docs/multi-gpu.html +++ b/docs/multi-gpu.html @@ -2,7 +2,7 @@ - + @@ -67,14 +67,15 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin + - + - + diff --git a/docs/multi-node.html b/docs/multi-node.html index a4fd61d25..059e4f76e 100644 --- a/docs/multi-node.html +++ b/docs/multi-node.html @@ -2,7 +2,7 @@ - + @@ -68,14 +68,15 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + - + - + + @@ -67,14 +67,15 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin + - + - + - + - + @@ -128,7 +129,8 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
@@ -154,6 +156,10 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
-
import os
+
import os
 
 # Optionally, upload your own JSONL to your Google Drive
 GOOGLE_DRIVE_PATH = ""  # ex: "MyDrive/Colab\ Notebooks/train.jsonl"
@@ -585,7 +591,7 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
     # make sure file exists
     if not os.path.isfile(tmp_path):
         raise ValueError(f"File {tmp_path} does not exist")
-    dataset_id = tmp_path

+    dataset_id = tmp_path
@@ -593,7 +599,7 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});

Configure for Supervised Fine-Tuning (SFT)

-
from axolotl.cli.config import load_cfg
+
from axolotl.cli.config import load_cfg
 from axolotl.utils.dict import DictDefault
 
 # Axolotl provides full control and transparency over model and training configuration
@@ -657,7 +663,7 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
 )
 
 # validates the configuration
-cfg = load_cfg(config)

+cfg = load_cfg(config)
[2025-05-08 13:40:27,488] [INFO] [root.register:348] [PID:174] Attempting to load plugin: axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
@@ -685,21 +691,21 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});
-
from axolotl.utils import patch_optimized_env
+
from axolotl.utils import patch_optimized_env
 
 # speedup downloads from HF 🤗 and set "PYTORCH_CUDA_ALLOC_CONF" env to save memory
-patch_optimized_env()

+patch_optimized_env()

Datasets

Axolotl has a robust suite of loaders and transforms to parse most open datasets of any format into the appropriate chat template for your model. Axolotl will mask input tokens from the user’s prompt so that the train loss is only calculated against the model’s response. For more information, see our documentation on dataset preparation.

-
from axolotl.common.datasets import load_datasets
+
from axolotl.common.datasets import load_datasets
 
 # Load, parse and tokenize the datasets to be formatted with qwen3 chat template
 # Drop long samples from the dataset that overflow the max sequence length
-dataset_meta = load_datasets(cfg=cfg)

+dataset_meta = load_datasets(cfg=cfg)
+ - + - + + - + - + + - + - +