solar-merge-slerp / README.md
nebchi's picture
Update README.md
e62d54f verified
|
raw
history blame
2.92 kB
---
base_model:
- davidkim205/nox-solar-10.7b-v2
- chihoonlee10/T3Q-ko-solar-dpo-v6.0
library_name: transformers
tags:
- mergekit
- merge
---
# model_storage
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 SLERP merge method.
### Models Merged
The following models were included in the merge:
* [davidkim205/nox-solar-10.7b-v2](https://huggingface.co/davidkim205/nox-solar-10.7b-v2)
* [chihoonlee10/T3Q-ko-solar-dpo-v6.0](https://huggingface.co/chihoonlee10/T3Q-ko-solar-dpo-v6.0)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
base_model:
model:
path: chihoonlee10/T3Q-ko-solar-dpo-v6.0
dtype: float16
merge_method: slerp
parameters:
t:
- filter: self_attn
value: [0.0, 0.5, 0.3, 0.7, 1.0]
- filter: mlp
value: [1.0, 0.5, 0.7, 0.3, 0.0]
- value: 0.5
slices:
- sources:
- layer_range: [0, 47]
model:
model:
path: chihoonlee10/T3Q-ko-solar-dpo-v6.0
- layer_range: [0, 47]
model:
model:
path: davidkim205/nox-solar-10.7b-v2
```
<!DOCTYPE html>
<html lang="ko">
<head>
<meta charset="UTF-8">
<title>Evaluation Results</title>
<style>
table {
width: 100%;
border-collapse: collapse;
margin: 25px 0;
font-size: 18px;
text-align: left;
}
th, td {
padding: 12px 15px;
}
th {
background-color: #f2f2f2;
}
tr:nth-of-type(even) {
background-color: #f9f9f9;
}
tr:hover {
background-color: #f1f1f1;
}
</style>
</head>
<body>
<table border="1">
<thead>
<tr>
<th>Model</th>
<th>글쓰기</th>
<th>이해</th>
<th>문법</th>
</tr>
</thead>
<tbody>
<tr>
<td>HyperClovaX</td>
<td>8.50</td>
<td>9.50</td>
<td><b>8.50</b></td>
</tr>
<tr>
<td>solar-1-mini-chat</td>
<td>8.50</td>
<td>7.00</td>
<td>5.21</td>
</tr>
<tr>
<td>allganize/Llama-3-Alpha-Ko-8B-Instruct</td>
<td>8.50</td>
<td>8.35</td>
<td>4.92</td>
</tr>
<tr>
<td>Synatra-kiqu-7B</td>
<td>4.42</td>
<td>5.71</td>
<td>4.50</td>
</tr>
<tr>
<td><b>Ocelot-ko-10.8B</b></td>
<td><b>8.57</b></td>
<td>7.00</td>
<td>6.57</td>
</tr>
</tbody>
</table>
</body>
</html>