|
--- |
|
base_model: |
|
- meta-llama/Meta-Llama-3-8B |
|
- shisa-ai/shisa-v1-llama3-8b |
|
- rinna/llama-3-youko-8b |
|
- tokyotech-llm/Llama-3-Swallow-8B-v0.1 |
|
library_name: transformers |
|
tags: |
|
- mergekit |
|
- merge |
|
license: llama3 |
|
language: |
|
- ja |
|
- en |
|
pipeline_tag: text-generation |
|
--- |
|
# merge2 |
|
|
|
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 [SCE](https://arxiv.org/abs/2408.07990) merge method using [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) as a base. |
|
|
|
### Models Merged |
|
|
|
The following models were included in the merge: |
|
* [shisa-ai/shisa-v1-llama3-8b](https://huggingface.co/shisa-ai/shisa-v1-llama3-8b) |
|
* [rinna/llama-3-youko-8b](https://huggingface.co/rinna/llama-3-youko-8b) |
|
* [tokyotech-llm/Llama-3-Swallow-8B-v0.1](https://huggingface.co/tokyotech-llm/Llama-3-Swallow-8B-v0.1) |
|
|
|
### Configuration |
|
|
|
The following YAML configuration was used to produce this model: |
|
|
|
```yaml |
|
|
|
models: |
|
# Pivot model |
|
- model: meta-llama/Meta-Llama-3-8B |
|
# Target models |
|
- model: shisa-ai/shisa-v1-llama3-8b |
|
- model: rinna/llama-3-youko-8b |
|
- model: tokyotech-llm/Llama-3-Swallow-8B-v0.1 |
|
merge_method: sce |
|
base_model: meta-llama/Meta-Llama-3-8B |
|
parameters: |
|
select_topk: 0.65 |
|
int8_mask: true |
|
dtype: bfloat16 |
|
|
|
``` |
