Removing unnecessary `device=device` in modeling_llama.py (#24696)

* Update modeling_llama.py

Removing unnecessary `device=device`

* fix in all occurrences of _make_causal_mask

src/transformers/models/owlvit/modeling_owlvit.py

@@ -792,7 +792,7 @@ def _make_causal_mask(
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
     mask_cond = torch.arange(mask.size(-1), device=device)
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)

src/transformers/models/pegasus/modeling_pegasus.py

@@ -80,7 +80,7 @@ def _make_causal_mask(
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
     mask_cond = torch.arange(mask.size(-1), device=device)
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)

src/transformers/models/pegasus_x/modeling_pegasus_x.py

@@ -98,7 +98,7 @@ def _make_causal_mask(
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
     mask_cond = torch.arange(mask.size(-1), device=device)
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)

src/transformers/models/plbart/modeling_plbart.py

@@ -85,7 +85,7 @@ def _make_causal_mask(
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
     mask_cond = torch.arange(mask.size(-1), device=device)
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)

src/transformers/models/speech_to_text/modeling_speech_to_text.py

@@ -70,7 +70,7 @@ def _make_causal_mask(
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
     mask_cond = torch.arange(mask.size(-1), device=device)
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)

src/transformers/models/speech_to_text_2/modeling_speech_to_text_2.py

@@ -50,7 +50,7 @@ def _make_causal_mask(
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
     mask_cond = torch.arange(mask.size(-1), device=device)
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)

src/transformers/models/speecht5/modeling_speecht5.py

@@ -97,7 +97,7 @@ def _make_causal_mask(
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
     mask_cond = torch.arange(mask.size(-1), device=device)
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)

src/transformers/models/time_series_transformer/modeling_time_series_transformer.py

@@ -233,7 +233,7 @@ def _make_causal_mask(
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
     mask_cond = torch.arange(mask.size(-1), device=device)
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)

src/transformers/models/trocr/modeling_trocr.py

@@ -50,7 +50,7 @@ def _make_causal_mask(
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
     mask_cond = torch.arange(mask.size(-1), device=device)
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)

src/transformers/models/whisper/modeling_whisper.py

@@ -80,7 +80,7 @@ def _make_causal_mask(
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
     mask_cond = torch.arange(mask.size(-1), device=device)
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)

src/transformers/models/x_clip/modeling_x_clip.py

@@ -746,7 +746,7 @@ def _make_causal_mask(
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
     mask_cond = torch.arange(mask.size(-1), device=device)
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)

src/transformers/models/xglm/modeling_xglm.py

@@ -135,7 +135,7 @@ def _make_causal_mask(
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
     mask_cond = torch.arange(mask.size(-1), device=device)
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)

templates/adding_a_new_model/cookiecutter-template-{{cookiecutter.modelname}}/modeling_{{cookiecutter.lowercase_modelname}}.py

@@ -1620,7 +1620,7 @@ def _make_causal_mask(input_ids_shape: torch.Size, dtype: torch.dtype, past_key_
     Make causal mask used for bi-directional self-attention.
     """
     bsz, tgt_len = input_ids_shape
-    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min))
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min)
     mask_cond = torch.arange(mask.size(-1))
     mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
     mask = mask.to(dtype)