AlekseiPravdin commited on
Commit
fadf3c4
1 Parent(s): ef965bb

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +70 -0
README.md ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - merge
5
+ - mergekit
6
+ - lazymergekit
7
+ - AlekseiPravdin/KukulStanta-InfinityRP-7B-slerp
8
+ - AlekseiPravdin/NSK-128k-7B-slerp
9
+ - gguf
10
+ - Q2_K
11
+ - Q3_K_L
12
+ - Q3_K_M
13
+ - Q3_K_S
14
+ - Q4_0
15
+ - Q4_1
16
+ - Q4_K_S
17
+ - Q4_k_m
18
+ - Q5_0
19
+ - Q5_1
20
+ - Q6_K
21
+ - Q5_K_S
22
+ - Q5_k_m
23
+ - Q8_0
24
+ - 128k
25
+ language:
26
+ - en
27
+ - ru
28
+ - th
29
+ ---
30
+
31
+ # KSI-RP-NSK-128k-7B
32
+
33
+ KSI-RP-NSK-128k-7B is a merge of the following models using [mergekit](https://github.com/cg123/mergekit):
34
+ * [AlekseiPravdin/KukulStanta-InfinityRP-7B-slerp](https://huggingface.co/AlekseiPravdin/KukulStanta-InfinityRP-7B-slerp)
35
+ * [AlekseiPravdin/NSK-128k-7B-slerp](https://huggingface.co/AlekseiPravdin/NSK-128k-7B-slerp)
36
+
37
+ ## 🧩 Configuration
38
+
39
+ ```yaml
40
+ slices:
41
+ - sources:
42
+ - model: AlekseiPravdin/KukulStanta-InfinityRP-7B-slerp
43
+ layer_range: [0, 32]
44
+ - model: AlekseiPravdin/NSK-128k-7B-slerp
45
+ layer_range: [0, 32]
46
+ merge_method: slerp
47
+ base_model: AlekseiPravdin/NSK-128k-7B-slerp
48
+ parameters:
49
+ t:
50
+ - filter: self_attn
51
+ value: [0, 0.5, 0.3, 0.7, 1]
52
+ - filter: mlp
53
+ value: [1, 0.5, 0.7, 0.3, 0]
54
+ - value: 0.5
55
+ dtype: bfloat16
56
+
57
+ ```
58
+
59
+ Eval embedding benchmark (with 70 specific quesions):
60
+
61
+ ![inf.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/UbeMfW28pMHSRLsSbEsJB.jpeg)
62
+ ![md28g.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/6UNV3CaKdofeAUr7C7x9k.jpeg)
63
+ ![SK.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/uSnHhxDCqo9DP9oSb_l6j.jpeg)
64
+ ![ks-inf.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/1ekTvK84ZlEsFFOYWOHE4.jpeg)
65
+ ![command-r.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/5lVz28EK07RmrUe49y4jn.jpeg)
66
+ ![NSK.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/aNdIdS5MnkwJ9YhprGznw.jpeg)
67
+ ![NSMv2.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/vk2GpfnJnYS5u1_wA1Nhr.jpeg)
68
+ ![aura.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/A3m0DC5E2x7V7UCbS1iCf.jpeg)
69
+ ![ivanDrogo.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/DaQIw6z8c-SupynTm9qos.jpeg)
70
+ ![KSI.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/EfEHDxVcAypb5YLDk_rQJ.jpeg)