tt-metal: ttnn.repeat throws runtime memory error

Describe the bug When trying to expand a tensor of shape [1, 1, 32, 16] to [1, 1, 32, 16*2048] using ttnn.repeat, the kernel fails with the following error FATAL | 32KB runtime args targeting kernel reader_concat_interleaved_start_id on (x=0,y=0) are too large. Cannot be written as they will run into memory region reserved for result. Max allowable size is 1 KB.

However, the same output tensor can be successfully created using ttnn.repeat_interleave api

To Reproduce Run below test code in your tt-metal environment

import ttnn
import torch
from tests.ttnn.utils_for_testing import assert_with_pcc


device = ttnn.open_device(device_id=0)
torch_input_tensor = torch.randn((1, 1, 32, 16), dtype=torch.bfloat16)
repeat_shape = (1, 1, 1, 2048)

torch_result_repeat_interleave = torch_input_tensor.repeat_interleave(2048, dim=3)
torch_result_repeat = torch_input_tensor.repeat(repeat_shape)

repeat_tensor = torch.randn(repeat_shape, dtype=torch.bfloat16)
repeat_tensor = ttnn.from_torch(repeat_tensor, layout=ttnn.TILE_LAYOUT, device=device)

input_tensor = ttnn.from_torch(torch_input_tensor, layout=ttnn.TILE_LAYOUT, device=device)

repeat_interleaved_output = ttnn.repeat_interleave(input_tensor, 2048, dim=3)
repeat_interleaved_output = ttnn.to_torch(repeat_interleaved_output)
print(repeat_interleaved_output.shape)
assert_with_pcc(torch_result_repeat_interleave, repeat_interleaved_output, 0.9999)
print("Repeat Interleave Passed")

repeat_output = ttnn.repeat(input_tensor, repeat_tensor.shape)
repeat_output = ttnn.to_torch(repeat_output)
print(repeat_output.shape)
assert_with_pcc(torch_result_repeat_interleave, repeat_output, 0.9999)

ttnn.close_device(device)

Expected behavior If ttnn.repeat_interleave passes then it is also expected to work with ttnn.repeat

Screenshots