423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-08-23 06:12:50 +08:00
Code Issues Packages Projects Releases Wiki Activity

[model] add QwQ 32b (#7179)

Browse Source 
Former-commit-id: 64a6fb9b5056166265abc5acbddffb64cd8b5256
This commit is contained in:
hoshi-hiyouga 2025-03-06 11:58:36 +08:00 committed by GitHub
parent c67d2b9327
commit 002f58ef8e
2 changed files with 6 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/llamafactory/extras/constants.py
View File

@ -2174,6 +2174,10 @@ register_model_group(
DownloadSource.DEFAULT: "Qwen/QwQ-32B-Preview",
DownloadSource.MODELSCOPE: "Qwen/QwQ-32B-Preview",
},
"QwQ-32B-Instruct": {
DownloadSource.DEFAULT: "Qwen/QwQ-32B",
DownloadSource.MODELSCOPE: "Qwen/QwQ-32B",
},
},
template="qwen",
)

5
src/llamafactory/train/trainer_utils.py
View File

@ -600,15 +600,14 @@ def get_swanlab_callback(finetuning_args: "FinetuningArguments") -> "TrainerCall
return
super().setup(args, state, model, **kwargs)
try:
if hasattr(self, "_swanlab"):
swanlab_public_config = self._swanlab.get_run().public.json()
else: # swanlab <= 0.4.9
swanlab_public_config = self._experiment.get_run().public.json()
except Exception as e:
except Exception:
swanlab_public_config = {}
with open(os.path.join(args.output_dir, SWANLAB_CONFIG), "w") as f:
f.write(json.dumps(swanlab_public_config, indent=2))