Fix orpo/dpo trainer (#1286)

Browse source

* change the colab notebook for dpo zephyr and orpo

* use original tokenizer

* Update README.md

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

...

This commit is contained in:

cell-dame 2024-11-27 04:02:06 +05:30 • committed by GitHub

parent 8a6da33bd0

commit 98a78ddfe4

No known key found for this signature in database

GPG key ID: B5690EEEBB952194

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download patch file Download diff file Expand all files Collapse all files

Diff content is not available