diff --git a/src/lighteval/metrics/dynamic_metrics.py b/src/lighteval/metrics/dynamic_metrics.py index e83c9ea6..7dd33d02 100644 --- a/src/lighteval/metrics/dynamic_metrics.py +++ b/src/lighteval/metrics/dynamic_metrics.py @@ -249,7 +249,7 @@ def sample_level_fn(golds: list[str], predictions: list[str], formatted_doc: Doc raise ValueError(f"No gold targets found for at least one gold. Gold: {golds}, Pred: {predictions}") if all(len(p) == 0 for p in extracted_predictions): - logger.warning(f"No predictions found for all predictions. Gold: {golds}, Pred: {predictions}") + logger.warning(f"We did not manage to extract a prediction in the correct format. Gold: {golds}, Pred: {predictions}") # We have to use timeout because the sypmy to str conversion can be very slow try: