apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-04 00:06:38 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Fix Speculative Config bug (#3049)

Browse Source 
* fix speculative bug

* fix rl
This commit is contained in:
YuanRisheng
2025-07-29 10:50:48 +08:00
committed by GitHub
parent f2a528f9ae
commit 1a815b7a2a
6 changed files with 21 additions and 58 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
fastdeploy/spec_decode/mtp.py
View File

@@ -73,7 +73,7 @@ class MTPProposer(Proposer):
self.model_config.architectures[0] = "Ernie4_5_MTPForCausalLM"
self.speculative_config.sharing_model = main_model
self.model_config.num_hidden_layers = 1
self.parallel_config.model_name_or_path = self.speculative_config.model_name_or_path
self.parallel_config.model_name_or_path = self.speculative_config.model
self.model_config.pretrained_config.prefix_name = "ernie.mtp_block"
if self.speculative_config.quantization != "":
self.model_config.quantization = self.speculative_config.quantization