Update README.md
Browse files
README.md
CHANGED
@@ -1,43 +1,14 @@
|
|
1 |
---
|
2 |
base_model: []
|
3 |
-
|
4 |
-
|
5 |
-
-
|
6 |
-
-
|
7 |
-
|
|
|
|
|
|
|
8 |
---
|
9 |
-
# Faro-34B-
|
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 SLERP merge method.
|
17 |
-
|
18 |
-
### Models Merged
|
19 |
-
|
20 |
-
The following models were included in the merge:
|
21 |
-
* /cephfs/panwenbo/work/models/Faro-34B-DPO
|
22 |
-
* /cephfs/panwenbo/work/models/Faro-34B
|
23 |
-
|
24 |
-
### Configuration
|
25 |
-
|
26 |
-
The following YAML configuration was used to produce this model:
|
27 |
-
|
28 |
-
```yaml
|
29 |
-
models:
|
30 |
-
- model: /cephfs/panwenbo/work/models/Faro-34B-DPO
|
31 |
-
- model: /cephfs/panwenbo/work/models/Faro-34B
|
32 |
-
merge_method: slerp
|
33 |
-
base_model: /cephfs/panwenbo/work/models/Faro-34B
|
34 |
-
parameters:
|
35 |
-
t:
|
36 |
-
- filter: self_attn
|
37 |
-
value: [0, 0.5, 0.3, 0.7, 1]
|
38 |
-
- filter: mlp
|
39 |
-
value: [1, 0.5, 0.7, 0.3, 0]
|
40 |
-
- value: 0.5 # fallback for rest of tensors
|
41 |
-
dtype: bfloat16
|
42 |
|
43 |
-
|
|
|
1 |
---
|
2 |
base_model: []
|
3 |
+
license: mit
|
4 |
+
datasets:
|
5 |
+
- wenbopan/Chinese-dpo-pairs
|
6 |
+
- Intel/orca_dpo_pairs
|
7 |
+
language:
|
8 |
+
- en
|
9 |
+
- zh
|
10 |
+
pipeline_tag: text-generation
|
11 |
---
|
12 |
+
# Faro-Yi-34B-DPO
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
|
14 |
+
Faro-Yi-34B-DPO is the DPO version of Faro-Yi-34B.
|