[Bug Report] invalid asin_bw backward result

[hschoi4448]

Describe the bug
A clear and concise description of what the bug is.

The asin_bw function returns an invalid gradient value.

To Reproduce
Steps to reproduce the behavior:

1. Copy and past below code to /tests/tt_eager/python_api_testing/unit_testing/backward_ops/test_backward_asin.py

# SPDX-FileCopyrightText: © 2023 Tenstorrent Inc.

# SPDX-License-Identifier: Apache-2.0

import torch
import pytest
import tt_lib
from tests.tt_eager.python_api_testing.unit_testing.backward_ops.utility_funcs import data_gen_pt_tt, compare_results

def data_gen_pt_tt(input_shapes, device, required_grad=False, val=1):
    pt_tensor = (torch.ones(input_shapes, requires_grad=required_grad) * val).bfloat16()
    tt_tensor = (
        tt_lib.tensor.Tensor(pt_tensor, tt_lib.tensor.DataType.BFLOAT16).to(tt_lib.tensor.Layout.TILE).to(device)
    )
    return pt_tensor, tt_tensor


@pytest.mark.parametrize(
    "input_shapes",
    (
        (torch.Size([1, 1, 32, 32])),
    ),
)
def test_bw_asin(input_shapes, device):
    in_data, input_tensor = data_gen_pt_tt(input_shapes, device, True, val=0)
    grad_data, grad_tensor = data_gen_pt_tt(input_shapes, device, False, val=0.95)
    
    print("input_tensor", input_tensor) # 0
    print("grad_tensor", grad_tensor) # 0.94922
    pyt_y = torch.asin(in_data)

    tt_output_tensor_on_device = tt_lib.tensor.asin_bw(grad_tensor, input_tensor)

    in_data.retain_grad()

    pyt_y.backward(gradient=grad_data)

    golden_tensor = [in_data.grad]
    comp_pass = compare_results(tt_output_tensor_on_device, golden_tensor)
    
    print("tt_output_tensor_on_device", tt_output_tensor_on_device) # 2.96875
    print("golden_tensor", golden_tensor) # 0.9492
    
    assert comp_pass

2. Run cmd pytest ./tests/tt_eager/python_api_testing/unit_testing/backward_ops/test_backward_asin.py

input_tensor ttnn.Tensor([[[[ 0.00000,  0.00000,  ...,  0.00000,  0.00000],
               [ 0.00000,  0.00000,  ...,  0.00000,  0.00000],
               ...,
               [ 0.00000,  0.00000,  ...,  0.00000,  0.00000],
               [ 0.00000,  0.00000,  ...,  0.00000,  0.00000]]]], shape=Shape([1, 1, 32, 32]), dtype=DataType::BFLOAT16, layout=Layout::TILE)
grad_tensor ttnn.Tensor([[[[ 0.94922,  0.94922,  ...,  0.94922,  0.94922],
               [ 0.94922,  0.94922,  ...,  0.94922,  0.94922],
               ...,
               [ 0.94922,  0.94922,  ...,  0.94922,  0.94922],
               [ 0.94922,  0.94922,  ...,  0.94922,  0.94922]]]], shape=Shape([1, 1, 32, 32]), dtype=DataType::BFLOAT16, layout=Layout::TILE)

tt_output_tensor_on_device [ttnn.Tensor([[[[ 2.96875,  2.96875,  ...,  2.96875,  2.96875],
               [ 2.96875,  2.96875,  ...,  2.96875,  2.96875],
               ...,
               [ 2.96875,  2.96875,  ...,  2.96875,  2.96875],
               [ 2.96875,  2.96875,  ...,  2.96875,  2.96875]]]], shape=Shape([1, 1, 32, 32]), dtype=DataType::BFLOAT16, layout=Layout::TILE)]
golden_tensor [tensor([[[[0.9492, 0.9492, 0.9492,  ..., 0.9492, 0.9492, 0.9492],
          [0.9492, 0.9492, 0.9492,  ..., 0.9492, 0.9492, 0.9492],
          [0.9492, 0.9492, 0.9492,  ..., 0.9492, 0.9492, 0.9492],
          ...,
          [0.9492, 0.9492, 0.9492,  ..., 0.9492, 0.9492, 0.9492],
          [0.9492, 0.9492, 0.9492,  ..., 0.9492, 0.9492, 0.9492],
          [0.9492, 0.9492, 0.9492,  ..., 0.9492, 0.9492, 0.9492]]]],
       dtype=torch.bfloat16)]

Expected behavior
A clear and concise description of what you expected to happen.

I want asin_bw to return the correct gradient

Screenshots
If applicable, add screenshots to help explain your problem.

Please complete the following environment information:

• OS: [e.g. Ubuntu 20.04]
• Version of software (eg. commit e14b574)

Additional context
Add any other context about the problem here.

[umadevimcw]

Merged to main