Set benchmark to the latest wandb run

utils.py

@@ -291,7 +291,7 @@ def get_benchmarks() -> Tuple[pd.DataFrame, datetime.datetime]:
         bt.logging.error("No benchmark project set.")
         return None, None
     runs = get_wandb_runs(project=BENCHMARK_WANDB_PROJECT, filters=None)
-    for run in runs:
+    for run in runs[::-1]:
         artifacts = list(run.logged_artifacts())
         if artifacts:
             table = artifacts[-1].get("benchmarks")