diff --git a/src/lighteval/logging/evaluation_tracker.py b/src/lighteval/logging/evaluation_tracker.py index 2694bae81..63fe99e8a 100644 --- a/src/lighteval/logging/evaluation_tracker.py +++ b/src/lighteval/logging/evaluation_tracker.py @@ -214,7 +214,7 @@ def save(self) -> None: details_datasets: dict[str, Dataset] = {} for task_name, task_details in self.details_logger.details.items(): # Create a dataset from the dictionary - we force cast to str to avoid formatting problems for nested objects - dataset = Dataset.from_list([asdict(detail) for detail in task_details]) + dataset = Dataset.from_list([{k: str(v) for k, v in asdict(detail).items()} for detail in task_details]) # We don't keep 'id' around if it's there column_names = dataset.column_names