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
59
Columbidae/GLM-Air-Q4KM-Yarn
Updated
•
142
Columbidae/Nemo-Base-Pruned-LP-Double
8B
•
Updated
•
12
Columbidae/Nemo-Base-Pruned-LP
8B
•
Updated
•
14
Columbidae/Nemo-Base-Pruned-Proj
7B
•
Updated
•
34
Columbidae/Nemo-Base-Pruned-S0
8B
•
Updated
•
19
Columbidae/Nemo-Base-Pruned-S0-Smaller
8B
•
Updated
•
15
Columbidae/Nemo-Base-Pruned-S2-Proj
6B
•
Updated
•
8
Columbidae/Nemo-Base-Pruned-S2
8B
•
Updated
•
10
Columbidae/Nemo-Base-Pruned-S1
Text Generation
•
11B
•
Updated
•
5
Columbidae/Ministral-8B-Instruct-2410
8B
•
Updated
•
3
datasets
0
None public yet