aws / aws/amazon-sagemaker-feedback

Improve Instruction Fine-Tuning Documentation on AWS Jumpstart

Open
#98 1 comment 0 reactions 0 assignees View on GitHub
documentation
Dominant language
No language data
Stars
10
Forks
3
PR merge metrics
No merged PRs in 30d

Description

### Product Category

JumpStart

### Feedback Category

Clarity and Comprehensibility, Missing Examples and Tutorials

### Reference Link

_No response_

### Details

It would be beneficial to clearly state if pre-processing of samples is necessary (any special tokens that need to be inserted for Mixtral 8x7b) / how the expected samples should look like. Also, the layout is messed up.

Contributor guide

Open the contributing guide

Research direction

Start by locating the AWS JumpStart instruction fine-tuning documentation for Mixtral 8x7b and inspect its current layout and sample guidance. Clarify whether preprocessing and special tokens are required, show the expected sample format, and correct the layout; done means the page answers these questions clearly and renders properly.

Written by the indexing model from the issue text.

Assessment

Tech stack
aws
Domain
documentation, machine-learning
Issue type
Documentation
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.