munish0838 commited on
Commit
d4362b1
1 Parent(s): 75508c6

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +60 -0
README.md ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ ---
3
+
4
+ license: apache-2.0
5
+ tags:
6
+ - merge
7
+ - mergekit
8
+ - lazymergekit
9
+ - vicgalle/Configurable-Llama-3.1-8B-Instruct
10
+ - bunnycore/HyperLlama-3.1-8B
11
+ - ValiantLabs/Llama3.1-8B-ShiningValiant2
12
+
13
+ ---
14
+
15
+ [![QuantFactory Banner](https://lh7-rt.googleusercontent.com/docsz/AD_4nXeiuCm7c8lEwEJuRey9kiVZsRn2W-b4pWlu3-X534V3YmVuVc2ZL-NXg2RkzSOOS2JXGHutDuyyNAUtdJI65jGTo8jT9Y99tMi4H4MqL44Uc5QKG77B0d6-JfIkZHFaUA71-RtjyYZWVIhqsNZcx8-OMaA?key=xt3VSDoCbmTY7o-cwwOFwQ)](https://hf.co/QuantFactory)
16
+
17
+
18
+ # QuantFactory/HyperLlama3.1-v2-GGUF
19
+ This is quantized version of [bunnycore/HyperLlama3.1-v2](https://huggingface.co/bunnycore/HyperLlama3.1-v2) created using llama.cpp
20
+
21
+ # Original Model Card
22
+
23
+
24
+ # HyperLlama3.1-v2
25
+
26
+ HyperLlama3.1-v2 is a merge of the following models using [mergekit](https://github.com/cg123/mergekit):
27
+ * [vicgalle/Configurable-Llama-3.1-8B-Instruct](https://huggingface.co/vicgalle/Configurable-Llama-3.1-8B-Instruct)
28
+ * [bunnycore/HyperLlama-3.1-8B](https://huggingface.co/bunnycore/HyperLlama-3.1-8B)
29
+ * [ValiantLabs/Llama3.1-8B-ShiningValiant2](https://huggingface.co/ValiantLabs/Llama3.1-8B-ShiningValiant2)
30
+
31
+ ## 🧩 Configuration
32
+
33
+ ```yaml
34
+ slices:
35
+ - sources:
36
+ - model: vicgalle/Configurable-Llama-3.1-8B-Instruct
37
+ parameters:
38
+ weight: 1
39
+ layer_range: [0, 32]
40
+ - model: bunnycore/HyperLlama-3.1-8B
41
+ parameters:
42
+ weight: 0.9
43
+ layer_range: [0, 32]
44
+ - model: ValiantLabs/Llama3.1-8B-ShiningValiant2
45
+ parameters:
46
+ weight: 0.6
47
+ layer_range: [0, 32]
48
+ merge_method: task_arithmetic
49
+ base_model: bunnycore/HyperLlama-3.1-8B
50
+ parameters:
51
+ t:
52
+ - filter: self_attn
53
+ value: [0, 0.5, 0.3, 0.7, 1]
54
+ - filter: mlp
55
+ value: [1, 0.5, 0.7, 0.3, 0]
56
+ - value: 0.5
57
+ dtype: bfloat16
58
+
59
+
60
+ ```