axolotl

Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, multimodal support

npx skills add https://github.com/firecrawl/ai-research-skills --skill axolotl

Axolotl Skill

Comprehensive assistance with axolotl development, generated from official documentation.

When to Use This Skill

This skill should be triggered when:

  • Working with axolotl
  • Asking about axolotl features or APIs
  • Implementing axolotl solutions
  • Debugging axolotl code
  • Learning axolotl best practices

Quick Reference

Common Patterns

Pattern 1: To validate that acceptable data transfer speeds exist for your training job, running NCCL Tests can help pinpoint bottlenecks, for example:

./build/all_reduce_perf -b 8 -e 128M -f 2 -g 3

Pattern 2: Configure your model to use FSDP in the Axolotl yaml. For example:

fsdp_version: 2
fsdp_config:
  offload_params: true
  state_dict_type: FULL_STATE_DICT
  auto_wrap_policy: TRANSFORMER_BASED_WRAP
  transformer_layer_cls_to_wrap: LlamaDecoderLayer
  reshard_after_forward: true

Pattern 3: The context_parallel_size should be a divisor of the total number of GPUs. For example:

context_parallel_size

Pattern 4: For example: - With 8 GPUs and no sequence parallelism: 8 different batches processed per step - With 8 GPUs and context_parallel_size=4: Only 2 different batches processed per step (each split across 4 GPUs) - If your per-GPU micro_batch_size is 2, the global batch size decreases from 16 to 4

context_parallel_size=4

Pattern 5: Setting save_compressed: true in your configuration enables saving models in a compressed format, which: - Reduces disk space usage by approximately 40% - Maintains compatibility with vLLM for accelerated inference - Maintains compatibility with llmcompressor for further optimization (example: quantization)

save_compressed: true

Pattern 6: Note It is not necessary to place your integration in the integrations folder. It can be in any location, so long as it’s installed in a package in your python env. See this repo for an example: https://github.com/axolotl-ai-cloud/diff-transformer

integrations

Pattern 7: Handle both single-example and batched data. - single example: sample[‘input_ids’] is a list[int] - batched data: sample[‘input_ids’] is a list[list[int]]

utils.trainer.drop_long_seq(sample, sequence_len=2048, min_sequence_len=2)

Example Code Patterns

Example 1 (python):

cli.cloud.modal_.ModalCloud(config, app=None)

Example 2 (python):

cli.cloud.modal_.run_cmd(cmd, run_folder, volumes=None)

Example 3 (python):

core.trainers.base.AxolotlTrainer(
    *_args,
    bench_data_collator=None,
    eval_data_collator=None,
    dataset_tags=None,
    **kwargs,
)

Example 4 (python):

core.trainers.base.AxolotlTrainer.log(logs, start_time=None)

Example 5 (python):

prompt_strategies.input_output.RawInputOutputPrompter()

Reference Files

This skill includes comprehensive documentation in references/:

  • api.md - Api documentation
  • dataset-formats.md - Dataset-Formats documentation
  • other.md - Other documentation

Use view to read specific reference files when detailed information is needed.

Working with This Skill

For Beginners

Start with the getting_started or tutorials reference files for foundational concepts.

For Specific Features

Use the appropriate category reference file (api, guides, etc.) for detailed information.

For Code Examples

The quick reference section above contains common patterns extracted from the official docs.

Resources

references/

Organized documentation extracted from official sources. These files contain:

  • Detailed explanations
  • Code examples with language annotations
  • Links to original documentation
  • Table of contents for quick navigation

scripts/

Add helper scripts here for common automation tasks.

assets/

Add templates, boilerplate, or example projects here.

Notes

  • This skill was automatically generated from official documentation
  • Reference files preserve the structure and examples from source docs
  • Code examples include language detection for better syntax highlighting
  • Quick reference patterns are extracted from common usage examples in the docs

Updating

To refresh this skill with updated documentation:

  1. Re-run the scraper with the same configuration
  2. The skill will be rebuilt with the latest information

Больше skills от firecrawl

oracle
firecrawl
Лучшие практики использования oracle CLI (объединение подсказок и файлов, движки, сессии и шаблоны прикрепления файлов).
official
pinecone
firecrawl
Управляемая векторная база данных для производственных AI-приложений. Полностью управляемая, с автоматическим масштабированием, гибридным поиском (плотный + разреженный), фильтрацией по метаданным и пространствами имён.
official
sentence-transformers
firecrawl
Фреймворк для современных эмбеддингов предложений, текстов и изображений. Предоставляет более 5000 предварительно обученных моделей для семантического сходства, кластеризации и поиска.…
official
wp-playground
firecrawl
Используется для рабочих процессов WordPress Playground: быстрые одноразовые экземпляры WP в браузере или локально через @wp-playground/cli (server, run-blueprint, build-snapshot),…
official
wp-plugin-development
firecrawl
Используйте при разработке плагинов WordPress: архитектура и хуки, активация/деактивация/удаление, административный интерфейс и Settings API, хранение данных, cron/задачи, безопасность…
official
wp-project-triage
firecrawl
Используйте, когда требуется детерминированная проверка репозитория WordPress (плагин/тема/блочная тема/WP Core/Gutenberg/полный сайт), включая инструменты/тесты/версии…
official
wp-rest-api
firecrawl
Используйте при создании, расширении или отладке конечных точек/маршрутов REST API WordPress: register_rest_route, классы контроллеров WP_REST_Controller, схема/аргументы…
official
wp-wpcli-and-ops
firecrawl
Использовать при работе с WP-CLI (wp) для операций с WordPress: безопасный поиск-замена, экспорт/импорт БД, управление плагинами/темами/пользователями/контентом, cron, очистка кэша,…
official