Update README.md
Browse files
README.md
CHANGED
@@ -13,9 +13,7 @@ tags:
|
|
13 |
|
14 |
</figure>
|
15 |
|
16 |
-
This is a merge of pre-trained language models
|
17 |
-
|
18 |
-
|
19 |
|
20 |
|
21 |
## Merge Details
|
@@ -26,7 +24,7 @@ This model was merged using the SLERP merge method.
|
|
26 |
### Models Merged
|
27 |
|
28 |
The following models were included in the merge:
|
29 |
-
*
|
30 |
* cognitivecomputations_dolphin-2.9-llama3-8b
|
31 |
|
32 |
### Configuration
|
@@ -36,12 +34,12 @@ The following YAML configuration was used to produce this model:
|
|
36 |
```yaml
|
37 |
slices:
|
38 |
- sources:
|
39 |
-
- model:
|
40 |
layer_range: [0, 32]
|
41 |
-
- model:
|
42 |
layer_range: [0, 32]
|
43 |
merge_method: slerp
|
44 |
-
base_model:
|
45 |
parameters:
|
46 |
t:
|
47 |
- filter: self_attn
|
|
|
13 |
|
14 |
</figure>
|
15 |
|
16 |
+
This is a merge of pre-trained language models abacusai/Llama-3-Smaug-8B and cognitivecomputations/dolphin-2.9-llama3-8b created using [mergekit](https://github.com/cg123/mergekit).
|
|
|
|
|
17 |
|
18 |
|
19 |
## Merge Details
|
|
|
24 |
### Models Merged
|
25 |
|
26 |
The following models were included in the merge:
|
27 |
+
* abacusai/Llama-3-Smaug-8B
|
28 |
* cognitivecomputations_dolphin-2.9-llama3-8b
|
29 |
|
30 |
### Configuration
|
|
|
34 |
```yaml
|
35 |
slices:
|
36 |
- sources:
|
37 |
+
- model: cognitivecomputations/dolphin-2.9-llama3-8b
|
38 |
layer_range: [0, 32]
|
39 |
+
- model: abacusai/Llama-3-Smaug-8B
|
40 |
layer_range: [0, 32]
|
41 |
merge_method: slerp
|
42 |
+
base_model: cognitivecomputations/dolphin-2.9-llama3-8b
|
43 |
parameters:
|
44 |
t:
|
45 |
- filter: self_attn
|