mirror of
https://github.com/deepseek-ai/DeepSeek-Coder.git
synced 2026-11-05 00:57:19 +00:00
[PR #699] Fintune swap tokenizer #291
Labels
No labels
documentation
ENV
pull-request
research question
WIP
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
deepseek-ai/DeepSeek-Coder#291
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
📋 Pull Request Information
Original PR: https://github.com/deepseek-ai/DeepSeek-Coder/pull/699
Author: @taras-sereda
Created: 5/18/2026
Status: 🔄 Open
Base:
main← Head:taras/fintune-swap-tokenizer📝 Commits (2)
da94445update requirements for both inference and finetuning9394394finetune deepseek coder with tokenizer from other model📊 Changes
3 files changed (+135 additions, -15 deletions)
View changed files
📝
finetune/finetune_deepseekcoder.py(+124 -8)📝
finetune/requirements.txt(+2 -2)📝
requirements.txt(+9 -5)📄 Description
This PR allows finetuning DeepSeek coder with a tokenizer take from another model.
Useful for tuning a speculator with the target model's tokenizer.
🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.