Hugging Face's logo

Casual-Autopsy
/
Llama-3-Yollow-SCE

Text Generation
Transformers
Safetensors
llama
mergekit
Merge
text-generation-inference
Inference Endpoints
Model card Files Community
Casual-Autopsy commited on
Commit
6285eb8
·
1 Parent(s): 74fbbab

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. README.md +3 -7
  2. mergekit_config.yml +2 -6
  3. model-00001-of-00004.safetensors +3 -0
  4. model-00002-of-00004.safetensors +3 -0
  5. model-00003-of-00004.safetensors +3 -0
  6. model-00004-of-00004.safetensors +3 -0
README.md CHANGED
@@ -18,8 +18,8 @@ This model was merged using the [SCE](https://arxiv.org/abs/2408.07990) merge me
18
  ### Models Merged
19
 
20
  The following models were included in the merge:
21
- * /kaggle/input/llama-3-swallow-8b-v0.1/transformers/hf/1
22
  * /kaggle/input/llama-3-youko-8b/transformers/hf/1
 
23
  * /kaggle/input/meta-llama-3-8b-instruct/transformers/hf/1
24
 
25
  ### Configuration
@@ -33,16 +33,12 @@ models:
33
  - model: /kaggle/input/meta-llama-3-8b/transformers/hf/1
34
  # Target models
35
  - model: /kaggle/input/meta-llama-3-8b-instruct/transformers/hf/1
36
- parameters:
37
- select_topk: 0.8
38
  - model: /kaggle/input/llama-3-youko-8b/transformers/hf/1
39
- parameters:
40
- select_topk: 0.65
41
  - model: /kaggle/input/llama-3-swallow-8b-v0.1/transformers/hf/1
42
- parameters:
43
- select_topk: 0.65
44
  merge_method: sce
45
  base_model: /kaggle/input/meta-llama-3-8b/transformers/hf/1
 
 
46
  dtype: bfloat16
47
 
48
  ```
 
18
  ### Models Merged
19
 
20
  The following models were included in the merge:
 
21
  * /kaggle/input/llama-3-youko-8b/transformers/hf/1
22
+ * /kaggle/input/llama-3-swallow-8b-v0.1/transformers/hf/1
23
  * /kaggle/input/meta-llama-3-8b-instruct/transformers/hf/1
24
 
25
  ### Configuration
 
33
  - model: /kaggle/input/meta-llama-3-8b/transformers/hf/1
34
  # Target models
35
  - model: /kaggle/input/meta-llama-3-8b-instruct/transformers/hf/1
 
 
36
  - model: /kaggle/input/llama-3-youko-8b/transformers/hf/1
 
 
37
  - model: /kaggle/input/llama-3-swallow-8b-v0.1/transformers/hf/1
 
 
38
  merge_method: sce
39
  base_model: /kaggle/input/meta-llama-3-8b/transformers/hf/1
40
+ parameters:
41
+ select_topk: 0.65
42
  dtype: bfloat16
43
 
44
  ```
mergekit_config.yml CHANGED
@@ -4,14 +4,10 @@ models:
4
  - model: /kaggle/input/meta-llama-3-8b/transformers/hf/1
5
  # Target models
6
  - model: /kaggle/input/meta-llama-3-8b-instruct/transformers/hf/1
7
- parameters:
8
- select_topk: 0.8
9
  - model: /kaggle/input/llama-3-youko-8b/transformers/hf/1
10
- parameters:
11
- select_topk: 0.65
12
  - model: /kaggle/input/llama-3-swallow-8b-v0.1/transformers/hf/1
13
- parameters:
14
- select_topk: 0.65
15
  merge_method: sce
16
  base_model: /kaggle/input/meta-llama-3-8b/transformers/hf/1
 
 
17
  dtype: bfloat16
 
4
  - model: /kaggle/input/meta-llama-3-8b/transformers/hf/1
5
  # Target models
6
  - model: /kaggle/input/meta-llama-3-8b-instruct/transformers/hf/1
 
 
7
  - model: /kaggle/input/llama-3-youko-8b/transformers/hf/1
 
 
8
  - model: /kaggle/input/llama-3-swallow-8b-v0.1/transformers/hf/1
 
 
9
  merge_method: sce
10
  base_model: /kaggle/input/meta-llama-3-8b/transformers/hf/1
11
+ parameters:
12
+ select_topk: 0.65
13
  dtype: bfloat16
model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4e5e358de8c0553d792d7632ed81ef59f800b6b201eff0fb20fb0c013c657b3
3
+ size 4953586384
model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc17beb40559194f0ad3b9446fe3338c4ec4d704cb06faf3437e89803a789882
3
+ size 4999819336
model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b233cdafea1798c767ec2b549a8007c3aa8d8473adddb61501b97a919364fa33
3
+ size 4915916144
model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24de7c106f6c73c8ba75b8f10af6ad776b242e541cc9a61cb2ccebe01e778198
3
+ size 1191234472