Qwen2.5 models with the tokenizers padded with extra unused special tokens to match the model embedding size. Workaround for various bugs.
Columba Labs
community
AI & ML interests
Birb make bots go brrr
Recent Activity
View all activity
models
61
Columbidae/granite-4.0-h-small-half-slerp
32B
•
Updated
•
149
•
1
Columbidae/GLM-Air-Q4KM-Yarn-65k
Updated
•
18
Columbidae/GLM-Air-Q4KM-Yarn
Updated
•
1
Columbidae/Nemo-Base-Pruned-LP-Double
8B
•
Updated
•
2
Columbidae/Nemo-Base-Pruned-LP
8B
•
Updated
•
3
Columbidae/Nemo-Base-Pruned-Proj
7B
•
Updated
•
2
Columbidae/Nemo-Base-Pruned-S0
8B
•
Updated
•
1
Columbidae/Nemo-Base-Pruned-S0-Smaller
8B
•
Updated
•
2
Columbidae/Nemo-Base-Pruned-S2-Proj
6B
•
Updated
•
1
Columbidae/Nemo-Base-Pruned-S2
8B
•
Updated
•
3
datasets
0
None public yet