refactor trainer to prevent circular dependencies later

fix loader default
2024-12-16 14:16:36 -05:00
parent 1ed4de73b6
commit 17ba9dcfdb
7 changed files with 1232 additions and 1161 deletions
--- a/docs/rlhf.qmd
+++ b/docs/rlhf.qmd
@@ -29,7 +29,7 @@ datasets:
    type: chatml.intel
  - path: argilla/ultrafeedback-binarized-preferences
    split: train
-    type: chatml.argilla
+    type: chatml
 ```

 #### IPO