Andrew Wu 90090fa9e8 DPO support loss types (#3566) ...

* Support loss_type/loss_weights DPO

* Validate dpo loss type/weights only set for dpo

* Tests: Update ipo tests to use new path

* Docs: Update docs for new ipo path

* PR fixes - typo/validation

* PR nit - warning

* chore: fix warnings arg

---------

Co-authored-by: NanoCode012 <nano@axolotl.ai>

2026-04-23 00:25:28 -04:00

..

agents

DPO support loss types (#3566)

2026-04-23 00:25:28 -04:00

dataset-formats

handle trainable/masked spans in content and reasoning content (#3592)

2026-04-10 14:11:10 -04:00

images

Ray Train Axolotl Integration (#2251)

2025-01-29 00:10:19 -05:00

scripts

build examples readmes with quarto (#3046)

2025-12-25 19:17:25 +07:00

.gitignore

build examples readmes with quarto (#3046)

2025-12-25 19:17:25 +07:00

amd_hpc.qmd

transformers v5 upgrade (#3272)

2026-01-27 17:08:24 -05:00

attention.qmd

feat: add FA4 (#3481)

2026-03-16 00:13:18 -04:00

batch_vs_grad.qmd

Feat: update doc (#1475) [skip ci]

2024-04-04 13:43:40 +09:00

checkpoint_saving.qmd

docs for checkpiont saving (#3335) [skip ci]

2025-12-30 12:40:32 -05:00

choosing_method.qmd

feat(docs): comprehensive improvement (#3564)

2026-04-02 08:01:26 -04:00

cli.qmd

fix: clarify how to use lm_eval plugin (#3404) [skip ci]

2026-02-15 07:52:30 -05:00

custom_integrations.qmd

densemixer plugin integration (#2868)

2025-07-07 17:05:19 -04:00

dataset_loading.qmd

Config doc autogen (#2718)

2025-06-18 15:36:53 -04:00

dataset_preprocessing.qmd

Autodoc generation with quartodoc (#2419)

2025-03-21 12:26:47 -04:00

debugging.qmd

feat: move to uv first (#3545)

2026-04-21 10:16:03 -04:00

docker.qmd

feat: move to uv first (#3545)

2026-04-21 10:16:03 -04:00

ebft.qmd

feat(docs): comprehensive improvement (#3564)

2026-04-02 08:01:26 -04:00

expert_quantization.qmd

fix(validation): add validation for lora target linear with quantize experts (#3461)

2026-03-06 09:19:05 -05:00

faq.qmd

fix: force train split for json,csv,txt for test_datasets and misc doc changes (#3226)

2025-10-22 15:23:20 -07:00

fsdp_qlora.qmd

Add FSDP v2 swap memory support + QLoRA compatibility fixes (#3167)

2025-09-26 10:23:59 +01:00

getting-started.qmd

feat(docs): comprehensive improvement (#3564)

2026-04-02 08:01:26 -04:00

gradient_checkpointing.qmd

support offloading layers to CPU (#3512) [skip ci]

2026-03-21 22:47:02 -04:00

grpo.qmd

feat(docs): comprehensive improvement (#3564)

2026-04-02 08:01:26 -04:00

inference.qmd

Feat: minor docs improvements for RLHF and faq on embeddings (#2401) [skip ci]

2025-03-17 08:39:04 -04:00

input_output.qmd

Feat(doc): Reorganize documentation, fix broken syntax, update notes (#2348)

2025-02-25 16:09:37 +07:00

installation.qmd

feat: move to uv first (#3545)

2026-04-21 10:16:03 -04:00

lora_optims.qmd

fix: improve lora kernels failure message and handle trust_remote_code (#3378) [skip ci]

2026-02-10 17:58:40 +07:00

lr_groups.qmd

fix: force train split for json,csv,txt for test_datasets and misc doc changes (#3226)

2025-10-22 15:23:20 -07:00

mac.qmd

Feat(doc): Reorganize documentation, fix broken syntax, update notes (#2348)

2025-02-25 16:09:37 +07:00

mixed_precision.qmd

basic torchao fp8 mixed precision training (#2926)

2025-07-22 16:27:47 -04:00

multi-gpu.qmd

Feat: add Olmo3 (BC with Olmo and Olmo2) (#3275)

2025-11-24 10:21:31 +07:00

multi-node.qmd

CLI: add --launcher option, support launcher args, cleanup, refactor (#2924)

2025-07-30 15:46:56 -04:00

multimodal.qmd

Gemma4 fixes and profiler (#3591)

2026-04-10 16:46:17 -04:00

multipack.qmd

Bootstrap Hosted Axolotl Docs w/Quarto (#1429)

2024-03-21 22:28:36 -07:00

nccl.qmd

Feat(doc): Reorganize documentation, fix broken syntax, update notes (#2348)

2025-02-25 16:09:37 +07:00

nd_parallelism.qmd

feat: update nd parallelism readme (#3039)

2025-08-08 12:45:36 +01:00

optimizations.qmd

support offloading layers to CPU (#3512) [skip ci]

2026-03-21 22:47:02 -04:00

optimizers.qmd

feat: add complete optimizer docs (#3017) [skip ci]

2025-08-06 08:01:51 -04:00

qat.qmd

feat(doc): add optimizations table of content to our improvements (#3175) [skip ci]

2025-09-24 16:13:49 -04:00

quantize.qmd

qat doc updates (#3162) [skip-ci]

2025-09-17 10:38:15 +01:00

ray-integration.qmd

Feat(doc): Reorganize documentation, fix broken syntax, update notes (#2348)

2025-02-25 16:09:37 +07:00

reward_modelling.qmd

Center rewards coefficient (#3124)

2025-09-03 16:22:37 -04:00

rlhf.qmd

DPO support loss types (#3566)

2026-04-23 00:25:28 -04:00

sequence_parallelism.qmd

Distributed/ND-Parallel (#2977)

2025-07-31 15:25:02 -04:00

streaming.qmd

Streaming SFT support (#3101)

2025-09-02 12:08:44 -04:00

telemetry.qmd

feat: Add opt-out Telemetry (#3237)

2025-11-18 11:35:25 +07:00

torchao.qmd

Feat(doc): Reorganize documentation, fix broken syntax, update notes (#2348)

2025-02-25 16:09:37 +07:00

training_stability.qmd

feat(docs): comprehensive improvement (#3564)

2026-04-02 08:01:26 -04:00

vllm_serving.qmd

feat(docs): comprehensive improvement (#3564)

2026-04-02 08:01:26 -04:00