diff --git a/src/datasets/arrow_dataset.py b/src/datasets/arrow_dataset.py
index 24d8aadfb60..5c34c349a3e 100644
--- a/src/datasets/arrow_dataset.py
+++ b/src/datasets/arrow_dataset.py
@@ -1320,7 +1320,7 @@ def set_format(
         self._format_kwargs = format_kwargs
         self._format_columns = columns
         self._output_all_columns = output_all_columns
-        logger.info(
+        logger.debug(
             "Set __getitem__(key) output type to %s for %s columns "
             " (when key is int or slice) and %s output other (un-formatted) columns.",
             "python objects" if type is None else type,
diff --git a/src/datasets/arrow_writer.py b/src/datasets/arrow_writer.py
index 713d2bbf7b3..a3f4fc7cfc7 100644
--- a/src/datasets/arrow_writer.py
+++ b/src/datasets/arrow_writer.py
@@ -423,7 +423,7 @@ def finalize(self, close_stream=True):
         self.pa_writer.close()
         if close_stream:
             self.stream.close()
-        logger.info(
+        logger.debug(
             "Done writing %s %s in %s bytes %s.",
             self._num_examples,
             self.unit,
diff --git a/src/datasets/builder.py b/src/datasets/builder.py
index 0e3a63acc48..2f7a762ad1d 100644
--- a/src/datasets/builder.py
+++ b/src/datasets/builder.py
@@ -729,7 +729,7 @@ def as_dataset(
                 % (self.name, self._cache_dir_root)
             )
 
-        logger.info(
+        logger.debug(
             "Constructing Dataset for split %s, from %s", split or ", ".join(self.info.splits), self._cache_dir
         )
 
diff --git a/src/datasets/load.py b/src/datasets/load.py
index dc658d042d0..0302380ce6f 100644
--- a/src/datasets/load.py
+++ b/src/datasets/load.py
@@ -171,7 +171,7 @@ def get_imports(file_path: str):
     with open(file_path, mode="r", encoding="utf-8") as f:
         lines.extend(f.readlines())
 
-    logger.info("Checking %s for additional imports.", file_path)
+    logger.debug("Checking %s for additional imports.", file_path)
     imports: List[Tuple[str, str, str, Optional[str]]] = []
     is_in_docstring = False
     for line in lines:
diff --git a/src/datasets/utils/filelock.py b/src/datasets/utils/filelock.py
index b67ffbf90ec..3e5bacafe27 100644
--- a/src/datasets/utils/filelock.py
+++ b/src/datasets/utils/filelock.py
@@ -272,7 +272,7 @@ def acquire(self, timeout=None, poll_intervall=0.05):
                         self._acquire()
 
                 if self.is_locked:
-                    logger().info("Lock %s acquired on %s", lock_id, lock_filename)
+                    logger().debug("Lock %s acquired on %s", lock_id, lock_filename)
                     break
                 elif timeout >= 0 and time.time() - start_time > timeout:
                     logger().debug("Timeout on acquiring lock %s on %s", lock_id, lock_filename)
@@ -315,7 +315,7 @@ def release(self, force=False):
                     logger().debug("Attempting to release lock %s on %s", lock_id, lock_filename)
                     self._release()
                     self._lock_counter = 0
-                    logger().info("Lock %s released on %s", lock_id, lock_filename)
+                    logger().debug("Lock %s released on %s", lock_id, lock_filename)
 
         return None