Minor documentation rephrasing

huggingface · Sep 13, 2023 · b5a3f5f · b5a3f5f
1 parent 6eaab24
commit b5a3f5f
Showing 1 changed file with 4 additions and 2 deletions.
diff --git a/src/accelerate/big_modeling.py b/src/accelerate/big_modeling.py
@@ -336,7 +336,8 @@ def dispatch_model(
             called directly during the forward, for instance if a `dense` linear layer is registered, but at forward,
             `dense.weight` and `dense.bias` are used in some operations instead of calling `dense` directly.
         force_hooks (`bool`, *optional*, defaults to `False`):
-            Whether or not to force hooks to be attached to the model even if it's on a single device.
+            Whether or not to force device hooks to be attached to the model even if all layers are dispatched to a
+            single device.
     """
     # Error early if the device map is incomplete.
     check_device_map(model, device_map)
@@ -487,7 +488,8 @@ def load_checkpoint_and_dispatch(
             called directly during the forward, for instance if a `dense` linear layer is registered, but at forward,
             `dense.weight` and `dense.bias` are used in some operations instead of calling `dense` directly.
         force_hooks (`bool`, *optional*, defaults to `False`):
-            Whether or not to force hooks to be attached to the model even if it's on a single device.
+            Whether or not to force device hooks to be attached to the model even if all layers are dispatched to a
+            single device.
 
     Example: