axolotl

axolotl — una habilidad instalable para agentes de IA, publicada por firecrawl/ai-research-skills.

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

Más skills de firecrawl

firecrawl-research-index
firecrawl
Encuentra los artículos que responden a una consulta de investigación con Firecrawl Research, utilizando búsqueda semántica, expansión semántica y estructural, y verificación en el cuerpo del texto. Usa siempre esta habilidad para cualquier tarea de búsqueda de literatura o recuperación de artículos, ya sea para consultas de un solo artículo o conjuntos completos de múltiples artículos.
data-analysisresearchweb-scraping
oracle
firecrawl
Mejores prácticas para usar la CLI de oracle (prompt + agrupación de archivos, motores, sesiones y patrones de adjuntar archivos).
pinecone
firecrawl
Base de datos vectorial gestionada para aplicaciones de IA en producción. Completamente gestionada, con escalado automático, búsqueda híbrida (densa + dispersa), filtrado por metadatos y espacios de nombres.…
wpds
firecrawl
Úsalo al construir interfaces de usuario que aprovechen el Sistema de Diseño de WordPress (WPDS) y sus componentes, tokens, patrones, etc.
audiocraft-audio-generation
firecrawl
Biblioteca de PyTorch para generación de audio que incluye texto a música (MusicGen) y texto a sonido (AudioGen). Úsala cuando necesites generar música a partir de texto…
skypilot-multi-cloud-orchestration
firecrawl
Orquestración multinube para cargas de trabajo de ML con optimización automática de costos. Úsalo cuando necesites ejecutar entrenamiento o trabajos por lotes en múltiples nubes, aprovechar…
firecrawl-seo-audit
firecrawl
Auditar el SEO de un sitio web con Firecrawl. Úsalo cuando el usuario solicite una auditoría SEO, revisión de metadatos y encabezados, análisis del mapa del sitio o estructura del sitio, oportunidades de palabras clave, comparación de SERP de competidores o recomendaciones priorizadas de optimización de búsqueda.
data-analysisresearchweb-scraping
gh-issues
firecrawl
Obtener issues de GitHub, generar subagentes para implementar correcciones y abrir PRs, luego monitorear y abordar comentarios de revisión de PRs. Uso: /gh-issues [owner/repo] [--label…