diff --git a/examples/llama-2/qlora-fsdp.yml b/examples/llama-2/qlora-fsdp.yml
index 30916ed45a..274bd7312d 100644
--- a/examples/llama-2/qlora-fsdp.yml
+++ b/examples/llama-2/qlora-fsdp.yml
@@ -65,6 +65,7 @@ deepspeed:
 weight_decay: 0.0
 fsdp:
   - full_shard
+  - auto_wrap
 fsdp_config:
   fsdp_limit_all_gathers: true
   fsdp_sync_module_states: true
@@ -73,4 +74,5 @@ fsdp_config:
   fsdp_cpu_ram_efficient_loading: true
   fsdp_transformer_layer_cls_to_wrap: LlamaDecoderLayer
   fsdp_state_dict_type: SHARDED_STATE_DICT
+  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
 special_tokens:
diff --git a/src/axolotl/utils/trainer.py b/src/axolotl/utils/trainer.py
index 2de2c54cce..6625080755 100644
--- a/src/axolotl/utils/trainer.py
+++ b/src/axolotl/utils/trainer.py
@@ -198,7 +198,7 @@ def calculate_total_num_steps(cfg, train_dataset, update=True):
             .apply(lambda x: len(x))  # pylint: disable=unnecessary-lambda
             .values
         )
-        LOG.debug(f"total_num_tokens: {total_num_tokens}", main_process_only=True)
+        LOG.debug(f"total_num_tokens: {total_num_tokens:_}", main_process_only=True)
         if update:
             cfg.total_num_tokens = total_num_tokens
 
@@ -212,7 +212,7 @@ def calculate_total_num_steps(cfg, train_dataset, update=True):
             .sum()
         )
         LOG.debug(
-            f"`total_supervised_tokens: {total_supervised_tokens}`",
+            f"`total_supervised_tokens: {total_supervised_tokens:_}`",
             main_process_only=True,
         )
         if update:
@@ -239,7 +239,7 @@ def calculate_total_num_steps(cfg, train_dataset, update=True):
                 * cfg.num_epochs
             )
             LOG.debug(
-                f"total_num_tokens: {cfg.total_num_tokens}, total_num_steps: {total_num_steps}",
+                f"total_num_tokens: {cfg.total_num_tokens:_}, total_num_steps: {total_num_steps:_}",
                 main_process_only=True,
             )
         else: