Update README.md
Browse files
README.md
CHANGED
@@ -7,4 +7,37 @@ language:
|
|
7 |
library_name: transformers
|
8 |
tags:
|
9 |
- evaluations
|
10 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
library_name: transformers
|
8 |
tags:
|
9 |
- evaluations
|
10 |
+
---
|
11 |
+
|
12 |
+
This is a quantized version merge of [evalistral](https://huggingface.co/rachittshah/evalistral) pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
13 |
+
|
14 |
+
## Merge Details
|
15 |
+
### Merge Method
|
16 |
+
|
17 |
+
This model was merged using the [TIES](https://arxiv.org/abs/2306.01708) merge method using [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) as a base.
|
18 |
+
|
19 |
+
### Models Merged
|
20 |
+
|
21 |
+
The following models were included in the merge:
|
22 |
+
* [kaist-ai/prometheus-7b-v1.0](https://huggingface.co/kaist-ai/prometheus-7b-v1.0)
|
23 |
+
* [Mistral-7b-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1)
|
24 |
+
|
25 |
+
### Configuration
|
26 |
+
|
27 |
+
The following YAML configuration was used to produce this model:
|
28 |
+
|
29 |
+
```yaml
|
30 |
+
models:
|
31 |
+
- model: mistralai/Mistral-7B-v0.1
|
32 |
+
# no parameters necessary for base model
|
33 |
+
- model: kaist-ai/prometheus-7b-v1.0
|
34 |
+
parameters:
|
35 |
+
density: 0.5
|
36 |
+
weight: 0.5
|
37 |
+
merge_method: ties
|
38 |
+
base_model: mistralai/Mistral-7B-v0.1
|
39 |
+
parameters:
|
40 |
+
normalize: true
|
41 |
+
dtype: float16
|
42 |
+
|
43 |
+
```
|