astral-demo-5 / README.md
adowu's picture
Upload folder using huggingface_hub
c924f88 verified
|
raw
history blame
1.29 kB
---
base_model:
- adowu/autocodit4
- alpindale/Mistral-7B-v0.2-hf
- beowolx/MistralHermes-CodePro-7B-v1
library_name: transformers
tags:
- mergekit
- merge
---
# merge
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 [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [adowu/autocodit4](https://huggingface.co/adowu/autocodit4) as a base.
### Models Merged
The following models were included in the merge:
* [alpindale/Mistral-7B-v0.2-hf](https://huggingface.co/alpindale/Mistral-7B-v0.2-hf)
* [beowolx/MistralHermes-CodePro-7B-v1](https://huggingface.co/beowolx/MistralHermes-CodePro-7B-v1)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
models:
- model: adowu/autocodit4
parameters:
density: 1.0
weight: 0.8
- model: alpindale/Mistral-7B-v0.2-hf
parameters:
density: 0.6
weight: 1.0
- model: beowolx/MistralHermes-CodePro-7B-v1
parameters:
density: 0.6
weight: 1.0
merge_method: dare_ties
base_model: adowu/autocodit4
parameters:
normalize: true
int8_mask: true
dtype: bfloat16
vocab_size: 128000
```