|
--- |
|
base_model: |
|
- Sao10K/L3-70B-Euryale-v2.1 |
|
- NousResearch/Hermes-2-Theta-Llama-3-70B |
|
library_name: transformers |
|
tags: |
|
- mergekit |
|
- merge |
|
|
|
--- |
|
# Hermes-2-Theta-L3-Euryale-Ties-0.8-70B |
|
|
|
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit). |
|
I wanted to have enhanced roleplay performance and merged two of my favorite models together. I maybe release some merges like this with different methods and biases. This config seemed to be the best mix of both world on first glance. |
|
I will update my experiences further in the future. |
|
|
|
## Merge Details |
|
### Merge Method |
|
|
|
This model was merged using the [TIES](https://arxiv.org/abs/2306.01708) merge method using [NousResearch/Hermes-2-Theta-Llama-3-70B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-70B) as a base. |
|
|
|
### Models Merged |
|
|
|
The following models were included in the merge: |
|
* [Sao10K/L3-70B-Euryale-v2.1](https://huggingface.co/Sao10K/L3-70B-Euryale-v2.1) |
|
|
|
### Configuration |
|
|
|
The following YAML configuration was used to produce this model: |
|
|
|
```yaml |
|
models: |
|
- model: NousResearch/Hermes-2-Theta-Llama-3-70B |
|
parameters: |
|
weight: 0.8 |
|
- model: Sao10K/L3-70B-Euryale-v2.1 |
|
parameters: |
|
weight: 0.2 |
|
merge_method: ties |
|
base_model: NousResearch/Hermes-2-Theta-Llama-3-70B |
|
parameters: |
|
density: 0.7 |
|
normalize: true |
|
dtype: float16 |
|
tokenizer_source: base |
|
|
|
``` |
|
|
