diff --git a/src/axolotl/utils/config/models/input/v0_4_1/__init__.py b/src/axolotl/utils/config/models/input/v0_4_1/__init__.py index ad332da2da..4e7b11950c 100644 --- a/src/axolotl/utils/config/models/input/v0_4_1/__init__.py +++ b/src/axolotl/utils/config/models/input/v0_4_1/__init__.py @@ -664,8 +664,8 @@ def check_sample_packing_wo_flash(cls, data): and not data.get("flash_attention") and not data.get("sdp_attention") ): - raise ValueError( - "sample_packing requires flash_attention or sdp_attention to be set to true" + LOG.warning( + "sample_packing without flash_attention or sdp_attention does not handle cross-attention." ) return data