automerger commited on
Commit
bef8b95
1 Parent(s): 1140dfa

Upload folder using huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +20 -21
README.md CHANGED
@@ -6,37 +6,36 @@ tags:
6
  - lazymergekit
7
  - automerger
8
  base_model:
9
- - vicgalle/Configurable-Hermes-2-Pro-Llama-3-8B
10
- - DeepMount00/Llama-3-8b-Ita
11
  ---
12
 
13
  # ConfigurableLlama-7B
14
 
15
  ConfigurableLlama-7B is an automated merge created by [Maxime Labonne](https://huggingface.co/mlabonne) using the following configuration.
16
- * [vicgalle/Configurable-Hermes-2-Pro-Llama-3-8B](https://huggingface.co/vicgalle/Configurable-Hermes-2-Pro-Llama-3-8B)
17
- * [DeepMount00/Llama-3-8b-Ita](https://huggingface.co/DeepMount00/Llama-3-8b-Ita)
18
 
19
  ## 🧩 Configuration
20
 
21
  ```yaml
22
- slices:
23
- - sources:
24
- - model: vicgalle/Configurable-Hermes-2-Pro-Llama-3-8B
25
- layer_range: [0, 32]
26
- - model: DeepMount00/Llama-3-8b-Ita
27
- layer_range: [0, 32]
28
- merge_method: slerp
29
- base_model: mlabonne/Meta-Llama-3-8B
 
 
 
 
 
30
  parameters:
31
- t:
32
- - filter: self_attn
33
- value: [0, 0.5, 0.3, 0.7, 1]
34
- - filter: mlp
35
- value: [1, 0.5, 0.7, 0.3, 0]
36
- - value: 0.5
37
- dtype: bfloat16
38
- random_seed: 0
39
- ```
40
 
41
  ## 💻 Usage
42
 
 
6
  - lazymergekit
7
  - automerger
8
  base_model:
9
+ - NousResearch/Meta-Llama-3-8B-Instruct
10
+ - mlabonne/OrpoLlama-3-8B
11
  ---
12
 
13
  # ConfigurableLlama-7B
14
 
15
  ConfigurableLlama-7B is an automated merge created by [Maxime Labonne](https://huggingface.co/mlabonne) using the following configuration.
16
+ * [NousResearch/Meta-Llama-3-8B-Instruct](https://huggingface.co/NousResearch/Meta-Llama-3-8B-Instruct)
17
+ * [mlabonne/OrpoLlama-3-8B](https://huggingface.co/mlabonne/OrpoLlama-3-8B)
18
 
19
  ## 🧩 Configuration
20
 
21
  ```yaml
22
+ models:
23
+ - model: NousResearch/Meta-Llama-3-8B
24
+ # No parameters necessary for base model
25
+ - model: NousResearch/Meta-Llama-3-8B-Instruct
26
+ parameters:
27
+ density: 0.6
28
+ weight: 0.5
29
+ - model: mlabonne/OrpoLlama-3-8B
30
+ parameters:
31
+ density: 0.55
32
+ weight: 0.05
33
+ merge_method: dare_ties
34
+ base_model: NousResearch/Meta-Llama-3-8B
35
  parameters:
36
+ int8_mask: true
37
+ dtype: float16
38
+ ```
 
 
 
 
 
 
39
 
40
  ## 💻 Usage
41