[E2E Accuracy] timm jx_nest_base amp_fp16 inference accuracy failed randomly

[mengfei25]

🐛 Describe the bug

Details in https://github.com/intel/torch-xpu-ops/actions/runs/11361002852

dev	name	batch_size	accuracy
xpu	jx_nest_base	8	fail_accuracy
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	pass
xpu	jx_nest_base	8	fail_accuracy

Versions

env:
pytorch: bdb42e7c944eb8c3bbfa0327e49e5db797a0bd92
torch-xpu-ops: 1d217ae
keep_torch_xpu_ops: false
python: 3.10
TRITON_COMMIT_ID: 91b14bf5593cf58a8541f3e6b9125600a867d4ef
TORCH_COMMIT_ID: bdb42e7c944eb8c3bbfa0327e49e5db797a0bd92
TRANSFORMERS_VERSION: 243e186efbf7fb93328dd6b34927a4e8c8f24395
DRIVER_VERSION: 803.61
KERNEL_VERSION: 5.15.0-73-generic #80-Ubuntu SMP Mon May 15 15:18:26 UTC 2023
BUNDLE_VERSION: 0.5.3
OS_PRETTY_NAME: Ubuntu 22.04.2 LTS
GCC_VERSION: 11

[retonym]

could not reproduce the random issue locally. This model passed in last weekly test.
Will check the condition in next weekly test.

[retonym]

amp_fp16 inference is not meta dashboard targeted datatype, move to milestone: PT2.7

[mengfei25]

fail accuracy in https://github.com/intel/torch-xpu-ops/actions/runs/13076390637/job/36490679565

[DDEle]

Fail accuracy in 4a9333a and pytorch/pytorch@93316cf https://github.com/intel/torch-xpu-ops/actions/runs/13428785399/job/37516867627