diff --git a/verifiers/rl/trainer/config.py b/verifiers/rl/trainer/config.py index b709696c7..c0c928015 100644 --- a/verifiers/rl/trainer/config.py +++ b/verifiers/rl/trainer/config.py @@ -298,6 +298,9 @@ def __post_init__(self): lora_alpha=self.lora_alpha, target_modules=self.lora_target_modules, task_type="CAUSAL_LM", + use_rslora=self.lora_use_rslora, + lora_dropout=self.lora_dropout, + modules_to_save=self.lora_modules_to_save, ) self.per_device_train_batch_size = self.micro_batch_size