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:

Copy code
./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:

Copy code
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:

Copy code
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

Copy code
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)

Copy code
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

Copy code
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]]

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

Example Code Patterns

Example 1 (python):

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

Example 2 (python):

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

Example 3 (python):

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

Example 4 (python):

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

Example 5 (python):

pythonCopy code
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

Agent operating procedure

1. Check the environment. Check GPU type, memory and driver/CUDA versions (nvidia-smi), framework versions, and dataset location and size.
2. Pin down the inputs. Confirm formats, identifiers and parameters from the data or the user. Ask rather than guess any value that changes the result.
3. Run a small version first. Do a smoke run: tiny model or subset, few steps, and confirm loss decreases and checkpoints save.
4. Execute the full task using the instructions and references above.
5. Validate the result. Track metrics on held-out data, compare against a baseline, and record seeds, configs and hardware.
6. Report. State what was run (versions, commands, parameters), what was checked, and what is still uncertain.

Integrity rules

• Never fabricate results, parameters, identifiers, citations or statistics. If something cannot be run or verified, say so plainly.
• Never claim training results without logs; estimate compute cost before launching large jobs.
• Treat version-specific details here as possibly outdated: confirm them against the official documentation for the installed version.
• Ask before actions that cost money, consume shared GPUs or cloud quota, touch personal or patient data, or cannot be undone.

Related skills

• llama-factory: Guides fine-tuning of large language models with LLaMA-Factory, covering the WebUI no-code interface, training across 100+ supported models…
• unsloth: Provides guidance on fine-tuning large language models with Unsloth, a library for faster, lower-memory training using LoRA and QLoRA, base…
• peft-fine-tuning: Fine-tunes LLMs with Hugging Face PEFT, using LoRA, QLoRA, IA3, AdaLoRA, prefix tuning, and prompt tuning so that under 1% of parameters ar…