KaraKaraWitch commited on
Commit
35708d9
·
verified ·
1 Parent(s): e367c36

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +80 -0
README.md ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ license_name: qwen
4
+ license_link: https://huggingface.co/Qwen/Qwen2.5-72B/blob/main/LICENSE
5
+ base_model:
6
+ - rombodawg/Rombos-LLM-V2.5-Qwen-72b
7
+ - abacusai/Dracarys2-72B-Instruct
8
+ - EVA-UNIT-01/EVA-Qwen2.5-72B-v0.0
9
+ - ZeusLabs/Chronos-Platinum-72B
10
+ - Qwen/Qwen2.5-72B
11
+ - m8than/banana-2-b-72b
12
+ language:
13
+ - en
14
+ pipeline_tag: text-generation
15
+ library_name: transformers
16
+ tags:
17
+ - mergekit
18
+ - merge
19
+
20
+ ---
21
+
22
+ # LLENN-v0.75-Qwen2.5-72b
23
+
24
+ [![image/png](https://cdn-uploads.huggingface.co/production/uploads/633e85093a17ab61de8d9073/mYiG-Ndxzqu8ofaBGbOIZ.png)](https://www.youtube.com/watch?v=PaEPo1sUc4Y "Cute Girl with a gun!")
25
+
26
+ I liked the previous model, but didn't *exactly* liked the claude vibes it's giving me. So I removed magnum. Other than that, there isn't any new model to merge in so the rest is kept as-is.
27
+
28
+ **Please do not ask for quants, contact others instead.**
29
+
30
+ *All models are ready for testing on [featherless.ai](https://featherless.ai) as soon as it goes live.*
31
+
32
+ ### Models Merged
33
+
34
+ The following models were included in the merge:
35
+ * [rombodawg/Rombos-LLM-V2.5-Qwen-72b](https://huggingface.co/rombodawg/Rombos-LLM-V2.5-Qwen-72b)
36
+ * [abacusai/Dracarys2-72B-Instruct](https://huggingface.co/abacusai/Dracarys2-72B-Instruct)
37
+ * [EVA-UNIT-01/EVA-Qwen2.5-72B-v0.0](https://huggingface.co/EVA-UNIT-01/EVA-Qwen2.5-72B-v0.0)
38
+ * [ZeusLabs/Chronos-Platinum-72B](https://huggingface.co/ZeusLabs/Chronos-Platinum-72B)
39
+ * [m8than/banana-2-b-72b](https://huggingface.co/m8than/banana-2-b-72b)
40
+
41
+ ### Configuration
42
+
43
+ The following YAML configuration was used to produce this model:
44
+
45
+ ```yaml
46
+ models:
47
+ - model: EVA-UNIT-01/EVA-Qwen2.5-72B-v0.0
48
+ - model: ZeusLabs/Chronos-Platinum-72B
49
+ - model: abacusai/Dracarys2-72B-Instruct
50
+ - model: rombodawg/Rombos-LLM-V2.5-Qwen-72b
51
+ - model: m8than/banana-2-b-72b
52
+
53
+ merge_method: model_stock
54
+ base_model: Qwen/Qwen2.5-72B
55
+ parameters:
56
+ normalize: true
57
+ dtype: bfloat16
58
+ ```
59
+
60
+ ### Prompt Format
61
+
62
+ ChatML works for the most part.
63
+
64
+ ### Sampler Settings
65
+
66
+ Personally I use the following:
67
+
68
+ ```
69
+ Temp: 1.2
70
+ Min P: 0.07
71
+ Rep Pen: 1.1
72
+ ```
73
+
74
+ Others have suggested the following:
75
+
76
+ ```
77
+ Temp: 1.1
78
+ Top P: 0.98
79
+ Min P: 0.05
80
+ ```