lm-sys / lm-sys/FastChat

Modified loss for Multi-turn conversations

Open
#1,861 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
39.5k
Forks
4.8k
PR merge metrics
No merged PRs in 30d

Description

Hi, upon reading the blog post of Vicuna, I see it stated that:
"Our training recipe builds on top of Stanford’s alpaca with the following improvements.
- Multi-turn conversations: We adjust the training loss to account for multi-turn conversations and compute the fine-tuning loss solely on the chatbot's output."

I wonder how impactful the loss modification is to the performance of the model? How much the difference between a chatbot model trained with and without this loss modification trick?

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No files, tests, or entry points are named in the issue. Start by reviewing the Vicuna training recipe and its multi-turn loss description, then compare model performance with and without that loss modification; done means reporting the measured difference.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.