xpu: enable gpt2 and decision_transformer tests for xpu pytorch backend

Note that running xpu tests requires TRANSFORMERS_TEST_DEVICE_SPEC=spec.py
passed to the test runner:

  import torch
  DEVICE_NAME = 'xpu'
  MANUAL_SEED_FN = torch.xpu.manual_seed
  EMPTY_CACHE_FN = torch.xpu.empty_cache
  DEVICE_COUNT_FN = torch.xpu.device_count

Signed-off-by: Dmitry Rogozhkin <dmitry.v.rogozhkin@intel.com>

src/transformers/models/decision_transformer/modeling_decision_transformer.py

@@ -22,7 +22,6 @@
 import torch
 import torch.utils.checkpoint
 from torch import nn
-from torch.cuda.amp import autocast
 
 from ...activations import ACT2FN
 from ...modeling_outputs import BaseModelOutputWithPastAndCrossAttentions
@@ -219,7 +218,7 @@ def _upcast_and_reordered_attn(self, query, key, value, attention_mask=None, hea
             scale_factor /= float(self.layer_idx + 1)
 
         # Upcast (turn off autocast) and reorder (Scale K by 1 / root(dk))
-        with autocast(enabled=False):
+        with torch.amp.autocast(query.device.type, enabled=False):
             q, k = query.reshape(-1, q_seq_len, dk), key.transpose(-1, -2).reshape(-1, dk, k_seq_len)
             attn_weights = torch.baddbmm(attn_weights, q.float(), k.float(), beta=0, alpha=scale_factor)
             attn_weights = attn_weights.reshape(bsz, num_heads, q_seq_len, k_seq_len)

src/transformers/models/gpt2/modeling_gpt2.py

@@ -25,7 +25,6 @@
 import torch.nn.functional as F
 import torch.utils.checkpoint
 from torch import nn
-from torch.cuda.amp import autocast
 from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
 
 from ...activations import ACT2FN
@@ -249,7 +248,7 @@ def _upcast_and_reordered_attn(self, query, key, value, attention_mask=None, hea
             scale_factor /= float(self.layer_idx + 1)
 
         # Upcast (turn off autocast) and reorder (Scale K by 1 / root(dk))
-        with autocast(enabled=False):
+        with torch.amp.autocast(query.device.type, enabled=False):
             q, k = query.reshape(-1, q_seq_len, dk), key.transpose(-1, -2).reshape(-1, dk, k_seq_len)
             attn_weights = torch.baddbmm(attn_weights, q.float(), k.float(), beta=0, alpha=scale_factor)
             attn_weights = attn_weights.reshape(bsz, num_heads, q_seq_len, k_seq_len)