46 lines
1.3 KiB
Markdown
46 lines
1.3 KiB
Markdown
---
|
|
base_model:
|
|
- POLARIS-Project/Polaris-7B-Preview
|
|
- deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
|
|
- nvidia/AceReason-Nemotron-7B
|
|
library_name: transformers
|
|
tags:
|
|
- mergekit
|
|
- merge
|
|
|
|
---
|
|
# merged-breadcrumbs
|
|
|
|
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 [Model Breadcrumbs with TIES](https://arxiv.org/abs/2312.06795) merge method using [deepseek-ai/DeepSeek-R1-Distill-Qwen-7B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) as a base.
|
|
|
|
### Models Merged
|
|
|
|
The following models were included in the merge:
|
|
* [POLARIS-Project/Polaris-7B-Preview](https://huggingface.co/POLARIS-Project/Polaris-7B-Preview)
|
|
* ./padded-maestro
|
|
* [nvidia/AceReason-Nemotron-7B](https://huggingface.co/nvidia/AceReason-Nemotron-7B)
|
|
|
|
### Configuration
|
|
|
|
The following YAML configuration was used to produce this model:
|
|
|
|
```yaml
|
|
merge_method: breadcrumbs_ties
|
|
base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
|
|
parameters:
|
|
density: 0.15 # Retain top 15% (Trims bottom 85% SFT noise)
|
|
gamma: 0.99 # Trim the top 1% (Destroys extreme RL outliers)
|
|
weight: 1.0
|
|
models:
|
|
- model: nvidia/AceReason-Nemotron-7B
|
|
- model: ./padded-maestro
|
|
- model: POLARIS-Project/Polaris-7B-Preview
|
|
dtype: bfloat16
|
|
|
|
```
|