EngineX-Ascend/enginex-ascend-910-llama.cpp
10
0
Fork 0
Code Issues Pull Requests Actions 4 Projects Releases Wiki Activity

llama-chat : fix wrong template in GLM4-0414 (#13140)

Browse Source 
* fix wrong template in GLM4-0414

* fix spaces

* no bos token since it is already in the template

* moved the chatgml4 check to higher priority

* restored template for old GLM models

* moved the GLM4 template check in the correct place with correct check
This commit is contained in:
matteo
2025-04-27 21:57:32 +02:00
committed by GitHub
parent e291450b76
commit ced44be342
2 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
convert_hf_to_gguf.py
View File

@@ -5154,7 +5154,7 @@ class Glm4Model(TextModel):
special_vocab._set_special_token("eos", tokenizer.get_added_vocab()["<|endoftext|>"])
special_vocab._set_special_token("eot", tokenizer.get_added_vocab()["<|user|>"])
special_vocab._set_special_token("unk", tokenizer.get_added_vocab()["<|endoftext|>"])
special_vocab._set_special_token("bos", tokenizer.get_added_vocab()["[gMASK]"])
special_vocab._set_special_token("bos", tokenizer.get_added_vocab()["<|endoftext|>"])
special_vocab.add_to_gguf(self.gguf_writer)
def set_gguf_parameters(self):