Error in converting llama

[purejomo]

I'm trying to convert my llama2 7b model following Readme

In STEP1,
I set my command like below,
python3 /home/ghlee/transformers/models/llama/convert_llama_weights_to_hf.py --input_dir /home/ghlee/llama/llama-2-7b --model_size 7B --output_dir /home/ghlee/llama.onnx/onnx_converted
and this kind of error occured

Traceback (most recent call last):
File "/home/ghlee/transformers/models/llama/convert_llama_weights_to_hf.py", line 407, in
main()
File "/home/ghlee/transformers/models/llama/convert_llama_weights_to_hf.py", line 396, in main
write_model(
File "/home/ghlee/transformers/models/llama/convert_llama_weights_to_hf.py", line 178, in write_model
f"model.layers.{layer_i}.self_attn.k_proj.weight": permute(
File "/home/ghlee/transformers/models/llama/convert_llama_weights_to_hf.py", line 154, in permute
return w.view(n_heads, dim1 // n_heads // 2, 2, dim2).transpose(1, 2).reshape(dim1, dim2)
RuntimeError: shape '[32, 2, 2, 4096]' is invalid for input of size 16777216

it seems like shape are not matched
but I never had modified the params.jason and config.jason

Is there any way to fix ti?