[bloom] fix alibi device placement (#18087)

src/transformers/models/bloom/modeling_bloom.py

@@ -93,7 +93,7 @@ def attention_mask_func(attention_scores, attention_mask, causal_mask):
     )
 
 
-def build_alibi_tensor(max_seq_len, n_head, dtype=torch.bfloat16):
+def build_alibi_tensor(max_seq_len, n_head, device, dtype=torch.bfloat16):
     """
     Link to paper: https://arxiv.org/abs/2108.12409 Alibi tensor is not causal as the original paper mentions, it
     relies on a translation invariance of softmax for quick implementation: with l being a tensor, and a fixed value
@@ -129,7 +129,7 @@ def build_alibi_tensor(max_seq_len, n_head, dtype=torch.bfloat16):
     arange_tensor = torch.arange(max_seq_len).unsqueeze(0).unsqueeze(0)
     alibi = slopes * arange_tensor.expand(n_head, -1, -1)
 
-    alibi = alibi.to(dtype)
+    alibi = alibi.to(device=device, dtype=dtype)
 
     return alibi
 
@@ -147,7 +147,7 @@ def pre_process_alibi_for_pad(alibi, attention_mask, num_heads):
     # This usually happens when the inference is done with past_key_values
     # In this case we re-create the alibi tensor with the correct sequence length
     if attention_mask.shape[-1] != alibi.shape[-1]:
-        alibi = build_alibi_tensor(attention_mask.shape[-1], num_heads, alibi.dtype).repeat(
+        alibi = build_alibi_tensor(attention_mask.shape[-1], num_heads, alibi.device, alibi.dtype).repeat(
             attention_mask.shape[0], 1, 1
         )
     # Get the indexes of the padding tokens
@@ -156,7 +156,7 @@ def pre_process_alibi_for_pad(alibi, attention_mask, num_heads):
 
     # Clone the embeddings  - we can detach because the embeddings are not learned
     # Get a refence tensor
-    slice_reference_alibi = build_alibi_tensor(alibi.shape[-1], num_heads, alibi.dtype)
+    slice_reference_alibi = build_alibi_tensor(alibi.shape[-1], num_heads, alibi.device, alibi.dtype)
 
     # Loop over the batch where the padding is and replace the alibi tensor by the reference tensor
     # Only where you do not have padding. Replace padding tokens by zeros
@@ -767,7 +767,7 @@ class BloomModel(BloomPreTrainedModel):
         current_sequence_length = hidden_states.shape[1]
         if past_key_values[0] is not None:
             current_sequence_length += past_key_values[0][0].shape[1]
-        alibi = build_alibi_tensor(current_sequence_length, self.n_head, hidden_states.dtype)
+        alibi = build_alibi_tensor(current_sequence_length, self.n_head, hidden_states.device, hidden_states.dtype)
 
         for i, (block, layer_past) in enumerate(zip(self.h, past_key_values)):