|
|
--- |
|
|
base_model: |
|
|
- deepseek-ai/DeepSeek-R1-Distill-Qwen-32B |
|
|
- Qwen/QwQ-32B |
|
|
library_name: transformers |
|
|
tags: |
|
|
- mergekit |
|
|
- merge |
|
|
--- |
|
|
# QwQ-R1-Distill-Merge-32B |
|
|
|
|
|
Testing locally it behaved very well for math problems. It usually starts a problem without the <think> tag, but ends by closing it when using chatml template. |
|
|
|
|
|
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit). |
|
|
|
|
|
## Merge Details |
|
|
### Merge Method |
|
|
|
|
|
This model was merged using the SLERP merge method. |
|
|
|
|
|
### Models Merged |
|
|
|
|
|
The following models were included in the merge: |
|
|
* /models/Qwen/QwQ-32B |
|
|
* /models/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B |
|
|
|
|
|
### Configuration |
|
|
|
|
|
The following YAML configuration was used to produce this model: |
|
|
|
|
|
```yaml |
|
|
base_model: /models/Qwen/QwQ-32B |
|
|
dtype: bfloat16 |
|
|
merge_method: slerp |
|
|
parameters: |
|
|
t: |
|
|
- filter: self_attn |
|
|
value: [0.0, 0.5, 0.3, 0.7, 1.0] |
|
|
- filter: mlp |
|
|
value: [1.0, 0.5, 0.7, 0.3, 0.0] |
|
|
- value: 0.5 |
|
|
slices: |
|
|
- sources: |
|
|
- layer_range: [0, 64] |
|
|
model: /models/Qwen/QwQ-32B |
|
|
- layer_range: [0, 64] |
|
|
model: /models/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B |
|
|
``` |
