Commit 1a4e4cad authored by yangql's avatar yangql
Browse files

修复qwen3-moe-awq的bug问题

parent d36deb1a
......@@ -332,6 +332,9 @@ class Qwen3MoeModel(nn.Module):
self.padding_idx = config.pad_token_id
self.vocab_size = config.vocab_size
self.config = config
if self.config.quantization_config["bits"] == 4:
os.environ['LLAMA_NN'] = '0'
os.environ['LM_NN'] = '0'
self.embed_tokens = VocabParallelEmbedding(
config.vocab_size,
config.hidden_size,
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment