Error when running python ./utils/convert-helper-bitnet.py models/Llama3-8B-1.58-100B-tokens

[ady133t]

Unable to quantize Llama3-8B-1.58-100B-tokens, this is the log.

Traceback (most recent call last):

File "D:\Git\BitNet\utils\convert-ms-to-gguf-bitnet.py", line 1855, in
main()

File "D:\Git\BitNet\utils\convert-ms-to-gguf-bitnet.py", line 1841, in main
model = convert_model_names(model, params, args.skip_unknown)

File "D:\Git\BitNet\utils\convert-ms-to-gguf-bitnet.py", line 1591, in convert_model_names
raise ValueError(f"Unexpected tensor name: {name}. Use --skip-unknown to ignore it (e.g. LLaVA)")

ValueError: Unexpected tensor name: model.layers.0.mlp.down_proj.weight_scale. Use --skip-unknown to ignore it (e.g. LLaVA)

Error executing command: D:\Program_Files\MiniConda\envs\bitnet-cpp\python.exe D:\Git\BitNet\utils\convert-ms-to-gguf-bitnet.py D:\Git\BitNet\models\Llama3-8B-1.58-100B-tokens --vocab-type bpe --outtype f32 --concurrency 1 --outfile D:\Git\BitNet\models\Llama3-8B-1.58-100B-tokens\ggml-model-f32-bitnet.gguf

Return code: 1

An error occurred: Command '['D:\Program_Files\MiniConda\envs\bitnet-cpp\python.exe', 'D:\Git\BitNet\utils\convert-ms-to-gguf-bitnet.py', 'D:\Git\BitNet\models\Llama3-8B-1.58-100B-tokens', '--vocab-type', 'bpe', '--outtype', 'f32', '--concurrency', '1', '--outfile', 'D:\Git\BitNet\models\Llama3-8B-1.58-100B-tokens\ggml-model-f32-bitnet.gguf']' returned non-zero exit status 1.

Cleaning up intermediate files...

Removing preprocessed file: D:\Git\BitNet\models\Llama3-8B-1.58-100B-tokens\model.safetensors

Restoring original 'D:\Git\BitNet\models\Llama3-8B-1.58-100B-tokens\model.safetensors' from 'D:\Git\BitNet\models\Llama3-8B-1.58-100B-tokens\model.safetensors.backup'