Upload README.md with huggingface_hub
Browse files
README.md
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model: []
|
3 |
+
library_name: transformers
|
4 |
+
tags:
|
5 |
+
- mergekit
|
6 |
+
- merge
|
7 |
+
|
8 |
+
---
|
9 |
+
# PapyLlamaMerged
|
10 |
+
|
11 |
+
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
12 |
+
|
13 |
+
## Merge Details
|
14 |
+
### Merge Method
|
15 |
+
|
16 |
+
This model was merged using the [TIES](https://arxiv.org/abs/2306.01708) merge method using /mimer/NOBACKUP/groups/naiss2024-22-361/Eric_Pap/Llama-3.1-8B-Instruct as a base.
|
17 |
+
|
18 |
+
### Models Merged
|
19 |
+
|
20 |
+
The following models were included in the merge:
|
21 |
+
* /mimer/NOBACKUP/groups/naiss2024-22-201/PapInsc3/Papyllama2
|
22 |
+
|
23 |
+
### Configuration
|
24 |
+
|
25 |
+
The following YAML configuration was used to produce this model:
|
26 |
+
|
27 |
+
```yaml
|
28 |
+
models:
|
29 |
+
- model: /mimer/NOBACKUP/groups/naiss2024-22-361/Eric_Pap/Llama-3.1-8B-Instruct
|
30 |
+
- model: /mimer/NOBACKUP/groups/naiss2024-22-201/PapInsc3/Papyllama2
|
31 |
+
parameters:
|
32 |
+
density: 1.1 # Fixed density, slightly more sparse than the original
|
33 |
+
weight: 0.6 # Fixed weight to keep the fine-tuned model's influence high
|
34 |
+
merge_method: ties
|
35 |
+
base_model: /mimer/NOBACKUP/groups/naiss2024-22-361/Eric_Pap/Llama-3.1-8B-Instruct
|
36 |
+
parameters:
|
37 |
+
normalize: true
|
38 |
+
dtype: bfloat16
|
39 |
+
|
40 |
+
|
41 |
+
|
42 |
+
|
43 |
+
```
|