diff --git a/llmfoundry/data/finetuning/dataloader.py b/llmfoundry/data/finetuning/dataloader.py index 661729ff8a..fce694f160 100644 --- a/llmfoundry/data/finetuning/dataloader.py +++ b/llmfoundry/data/finetuning/dataloader.py @@ -336,6 +336,7 @@ def build_finetuning_dataloader( replication_factor if replication_factor > 1 else None, rank=dist.get_global_rank() // replication_factor if replication_factor > 1 else None, + seed=dataset_cfg.get('shuffle_seed', 0), ) assert streaming_dataset is not None # for pyright