LAION-AI / LAION-AI/Open-Assistant

Allow cancellation of prediction while running prompt

Open
#2,815 13 comments 2 reactions 1 assignee View on GitHub

@0xfacade is already working on this.

Since May 22, 2023.

inference
Dominant language
Python
Stars
37.4k
Forks
3.3k
No merged PRs in 30d

Description

Open Assistant is great, but sometimes it will predict a long answer where I can spot a misinterpretation right away. Whether this is because my prompt was faulty and I realise this too late, or the model hallucinates. Either way, having to wait for the entire prediction to load significantly reduces the UX (due to waiting time). It would be useful to be able to abort the prediction.

Model OA_SFT_Llama_30B_6 
Top K 50 
Top P 0.95 
Temperature 0.9 
Repetition penalty 1.2 
Max new tokens 1024

Contributor guide

Open the contributing guide

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.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.