MrRobotoAI's picture
Upload folder using huggingface_hub
4e95e5b verified
---
base_model:
- WhiteRabbitNeo/Llama-3-WhiteRabbitNeo-8B-v2.0
- Undi95/Llama-3-Unholy-8B
- Undi95/Llama-3-LewdPlay-8B
- UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3
- VAGOsolutions/Llama-3-SauerkrautLM-8b-Instruct
- MrRobotoAI/llama3-8B-Special-Dark-v2.0
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 [MrRobotoAI/llama3-8B-Special-Dark-v2.0](https://huggingface.co./MrRobotoAI/llama3-8B-Special-Dark-v2.0) as a base.
### Models Merged
The following models were included in the merge:
* [WhiteRabbitNeo/Llama-3-WhiteRabbitNeo-8B-v2.0](https://huggingface.co./WhiteRabbitNeo/Llama-3-WhiteRabbitNeo-8B-v2.0)
* [Undi95/Llama-3-Unholy-8B](https://huggingface.co./Undi95/Llama-3-Unholy-8B)
* [Undi95/Llama-3-LewdPlay-8B](https://huggingface.co./Undi95/Llama-3-LewdPlay-8B)
* [UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3](https://huggingface.co./UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3)
* [VAGOsolutions/Llama-3-SauerkrautLM-8b-Instruct](https://huggingface.co./VAGOsolutions/Llama-3-SauerkrautLM-8b-Instruct)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
models:
- model: UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3
parameters:
weight: 0.1667
density: 0.9
- model: Undi95/Llama-3-LewdPlay-8B
parameters:
weight: 0.1667
density: 0.9
- model: Undi95/Llama-3-Unholy-8B
parameters:
weight: 0.1667
density: 0.9
- model: VAGOsolutions/Llama-3-SauerkrautLM-8b-Instruct
parameters:
weight: 0.1667
density: 0.9
- model: WhiteRabbitNeo/Llama-3-WhiteRabbitNeo-8B-v2.0
parameters:
weight: 0.1667
density: 0.9
- model: MrRobotoAI/llama3-8B-Special-Dark-v2.0
parameters:
weight: 0.1667
density: 0.9
merge_method: dare_ties
base_model: MrRobotoAI/llama3-8B-Special-Dark-v2.0
parameters:
normalize: true
int8_mask: true
dtype: float16
```