From 5f1a4306b0e56ef8d0f8e99bf2c6c2f51cf1ebf4 Mon Sep 17 00:00:00 2001 From: Wing Lian Date: Tue, 22 Jul 2025 20:40:44 -0400 Subject: [PATCH] don't check dataset labels during preprocess for GRPO (#2952) [skip ci] * don't check dataset labels during preprocess for GRPO * use enum check per PR feedback --- src/axolotl/common/datasets.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/axolotl/common/datasets.py b/src/axolotl/common/datasets.py index a9b4c0f0f..761317dfb 100644 --- a/src/axolotl/common/datasets.py +++ b/src/axolotl/common/datasets.py @@ -122,7 +122,7 @@ def load_preference_datasets( math.ceil(len(train_dataset) * cfg.num_epochs / cfg.batch_size) ) - if (cli_args and cli_args.debug) or cfg.debug: + if ((cli_args and cli_args.debug) or cfg.debug) and cfg.rl != RLType.ORPO: LOG.info("check_dataset_labels...") num_examples = cli_args.debug_num_examples if cli_args else 1