Skip to content

Latest commit

 

History

History
99 lines (69 loc) · 4.81 KB

index.md

File metadata and controls

99 lines (69 loc) · 4.81 KB
title description logo
About taxonomy trees and QnA YAMLs
The overview of 🐶 InstructLab's Taxonomy.
images/ilab_dog.png

About taxonomy trees and QnA YAMLs

InstructLab 🐶 uses a novel synthetic data-based alignment tuning method for Large Language Models (LLMs.) The "lab" in InstructLab 🐶 stands for Large-Scale Alignment for ChatBots [1].

The LAB method is driven by taxonomies, which are largely created manually and with care.

This repository contains a taxonomy tree that allows you to create models tuned with your data (enhanced via synthetic data generation) using the LAB 🐶 method.

[1] Shivchander Sudalairaj*, Abhishek Bhandwaldar*, Aldo Pareja*, Kai Xu, David D. Cox, Akash Srivastava*. "LAB: Large-Scale Alignment for ChatBots", arXiv preprint arXiv: 2403.01081, 2024. (* denotes equal contributions)

Intro to skills and knowledge

Skill and knowledge are the types of data that you can add to the taxonomy tree. You can then use these types to train a model on data it might not already know.

Knowledge

Knowledge for an AI model consists of data and facts. When creating knowledge sets for a model, you are providing it with additional data and information so the model can answer questions more accurately. Where skills are the information that trains an AI model on how to do something, knowledge is based on the model’s ability to answer questions that involve facts, data, or references. For example, you can create a data set that includes a product’s documentation and the model can learn the information provided in that documentation.

Skills

A skill is a capability domain that intends to train the AI model on submitted information. When you make a skill, you are teaching the model how to do a task. Skills on RHEL AI are split into categories:

  • Composition skill: Compositional skills allow AI models to perform specific tasks or functions. There are two types of compositional skills: ** Freeform compositional skills: These are performative skills that do not require additional context or information to function. ** Grounded compositional skills: These are performative skills that require additional context. For example, you can teach the model to read a table, where the additional context is an example of the table layout. Foundation skills: Foundational skills are skills that involve math, reasoning, and coding.

InstructLab QnA YAML files

You can teach LLMs new information by creating a qna.yaml file that contains information of your knowledge or details of your skill.

For more information on creating skills and knowledge YAML files, see:

Choosing domains for the taxonomy

In general, we use the Dewey Decimal Classification (DDC) System to determine our domains (and subdomains) in the taxonomy. This DDC SUMMARIES document is a great resource for determining where a topic might be classified.

If you are unsure where to put your knowledge or compositional skill, create a folder in the miscellaneous_unknown folder under the knowledge or compositional_skills folders.

Taxonomy tree Layout

The taxonomy tree is organized in a cascading directory structure. At the end of each branch, there is a YAML file (qna.yaml) that contains the examples for that domain. Maintainers can decide to change the names of the existing branches or to add new branches.

!!! important Folder names do not have spaces. Use underscores between words.

Taxonomy diagram

!!! note These diagrams shows a subset of the taxonomy. It is not a complete representation.

 flowchart TD;
   na[not accepting contributions\n at this time]:::na
   taxonomy --> foundational_skill & compositional_skills & knowledge

   foundational_skill:::na --> reasoning:::na
   reasoning:::na --> common_sense_reasoning:::na
   reasoning:::na --> mathematical_reasoning:::na
   reasoning:::na --> theory_of_mind:::na

   compositional_skills --> engineering
   compositional_skills --> grounded
   compositional_skills --> lingustics

   grounded --> grounded/arts
   grounded --> grounded/geography
   grounded --> grounded/history
   grounded --> grounded/science

   knowledge --> knowledge/arts

   knowledge --> knowledge/miscellaneous_unknown
   knowledge --> knowledge/science
   knowledge --> knowledge/technology
   knowledge/science --> animals --> birds --> black_capped_chickadee --> black_capped_chikadee-a & black_capped_chikadee-q
   knowledge/science --> astronomy --> constellations --> phoenix --> phoenix-a & phoenix-q

   black_capped_chikadee-a{attribution.txt}
   black_capped_chikadee-q{qna.yaml}
   phoenix-a{attribution.txt}
   phoenix-q{qna.yaml}
   classDef na fill:#EEE
Loading