Skip to content

Commit

Permalink
my first skill
Browse files Browse the repository at this point in the history
Signed-off-by: Yogendra Srivastava <ysrivastava@in.ibm.com>
  • Loading branch information
yogendrasri committed Jan 17, 2025
1 parent dfa3afa commit 18555b0
Show file tree
Hide file tree
Showing 2 changed files with 34 additions and 0 deletions.
3 changes: 3 additions & 0 deletions compositional_skills/technology/attribution.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
Title of work: InstructLab
License of the work: https://github.com/yogendrasri/instructlab.git
Creator names: Yogendra
31 changes: 31 additions & 0 deletions compositional_skills/technology/qna.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,31 @@
created_by: yogendrasri
version: 3
task_description: my first InstructLab skill implementation
seed_examples:
- question: What is InstructLab
answer: >
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.
- question: What is InstructLab Core?
answer: >
instructlab is the Core package for the InstructLab project that contains
the ilab Command-Line Interface (CLI) tool and allows you to perform the
following actions:
Download a pre-trained Large Language Model (LLM).
Chat with the LLM.
- question: what is InstructLab User Interface (UI)?
answer: >-
The InstructLab User Interface (UI) is an additional component in the
InstructLab project to make Knowledge/Skills contributions more accessible
to the general user.
- question: How does InstructLab work?
answer: >-
InstructLab is driven by taxonomies and works by empowering users to add
new skills and knowledge to a pre-trained LLM.
- question: How can I contribute?
answer: >-
You can begin your contribution journey by reading over the Contributing
guide and joining the Community Discord Server or the Community Slack
Channel.

0 comments on commit 18555b0

Please sign in to comment.