ORPO (#1419)

* orpo trainer

* rl handling for orpo

* support for remove_unused_columns

* orpo fixes

* fix loader for orpo

* chore: lint

* fix default for remove_unused_columns

* roll ORPO into the main AxolotlTrainer so it can be compatible with some of the other techniques like relora

* better handling of system message for orpo

* revert system prompt changes for chat templtes

* no need for else condition

* split dataset parsing into it's own component

src/axolotl/train.py

@@ -85,7 +85,7 @@ def train(
     model.generation_config.do_sample = True
 
     model_ref = None
-    if cfg.rl:
+    if cfg.rl and cfg.rl != "orpo":
         if cfg.adapter and not cfg.rl_adapter_ref_model:
             # use built-in trl autounwrap
             LOG.debug("Passing model_ref: None to RL trainer")