Crystalcareai commited on
Commit
9ec5973
1 Parent(s): e9e3537

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -46
README.md CHANGED
@@ -1,46 +1 @@
1
- ---
2
- base_model: []
3
- library_name: transformers
4
- tags:
5
- - mergekit
6
- - merge
7
-
8
- ---
9
- # final
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
- * ./commerge1
22
- * ./commerge2
23
-
24
- ### Configuration
25
-
26
- The following YAML configuration was used to produce this model:
27
-
28
- ```yaml
29
- slices:
30
- - sources:
31
- - model: ./commerge1
32
- layer_range: [0, 28]
33
- - model: ./commerge2
34
- layer_range: [0, 28]
35
-
36
- merge_method: slerp
37
- base_model: ./commerge1
38
- parameters:
39
- t:
40
- - filter: self_attn
41
- value: [0, 0.5, 0.3, 0.7, 1]
42
- - filter: mlp
43
- value: [1, 0.5, 0.7, 0.3, 0]
44
- - value: 0.5
45
- dtype: bfloat16
46
- ```
 
1
+ This is simply a base gemma with the bug fixes I implemented in GemMoE applied to this model. Be sure to use ```trust_remote_code:/= True``` when training or running inference.