Padded Qwen Models
Collection
Qwen2.5 models with the tokenizers padded with extra unused special tokens to match the model embedding size. Workaround for various bugs.
•
5 items
•
Updated
This is a copy of Qwen/QwQ-32B made for experimental purposes.
The tokenizer of this model has been padded out to contain a vocab size of 152064 to match what's in the config.
- Downloads last month
- 126
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support