Apel-sin commited on
Commit
cb2e451
·
1 Parent(s): f80a800

add measurement.json

Browse files
Files changed (2) hide show
  1. README.md +44 -0
  2. measurement.json +0 -0
README.md ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ license_link: https://huggingface.co/huihui-ai/QwQ-32B-Coder-Fusion-9010/blob/main/LICENSE
4
+ language:
5
+ - en
6
+ pipeline_tag: text-generation
7
+ base_model:
8
+ - huihui-ai/QwQ-32B-Coder-Fusion-9010
9
+ tags:
10
+ - chat
11
+ - abliterated
12
+ - uncensored
13
+ library_name: transformers
14
+ ---
15
+
16
+ # huihui-ai/QwQ-32B-Coder-Fusion-9010
17
+
18
+
19
+ ## Overview
20
+ `QwQ-32B-Coder-Fusion-9010` is a mixed model that combines the strengths of two powerful Qwen-based models: [huihui-ai/QwQ-32B-Preview-abliterated](https://huggingface.co/huihui-ai/QwQ-32B-Preview-abliterated) and
21
+ [huihui-ai/Qwen2.5-Coder-32B-Instruct-abliterated](https://huggingface.co/huihui-ai/Qwen2.5-Coder-32B-Instruct-abliterated).
22
+ The weights are blended in a 9:1 ratio, with 90% of the weights from the abliterated QwQ-32B-Preview-abliterated and 10% from the abliterated Qwen2.5-Coder-32B-Instruct-abliterated model.
23
+ **Although it's a simple mix, the model is usable, and no gibberish has appeared**.
24
+ This is an experiment. I test the [9:1](https://huggingface.co/huihui-ai/QwQ-32B-Coder-Fusion-9010),
25
+ [8:2](https://huggingface.co/huihui-ai/QwQ-32B-Coder-Fusion-8020),
26
+ and [7:3](https://huggingface.co/huihui-ai/QwQ-32B-Coder-Fusion-7030) ratios separately to see how much impact they have on the model.
27
+ Now the effective ratios are 9:1, 8:2, and 7:3. Any other ratios (6:4,5:5) would result in mixed or unclear expressions.
28
+
29
+ ## Model Details
30
+ - **Base Models:**
31
+ - [huihui-ai/QwQ-32B-Preview-abliterated](https://huggingface.co/huihui-ai/QwQ-32B-Preview-abliterated) (90%)
32
+ - [huihui-ai/Qwen2.5-Coder-32B-Instruct-abliterated](https://huggingface.co/huihui-ai/Qwen2.5-Coder-32B-Instruct-abliterated) (10%)
33
+ - **Model Size:** 32B parameters
34
+ - **Architecture:** Qwen 2.5
35
+ - **Mixing Ratio:** 9:1 (QwQ-32B-Preview-abliterated:Qwen2.5-Coder-32B-Instruct-abliterated)
36
+
37
+ ## ollama
38
+
39
+ You can use [huihui_ai/qwq-fusion](https://ollama.com/huihui_ai/qwq-fusion) directly,
40
+ ```
41
+ ollama run huihui_ai/qwq-fusion
42
+ ```
43
+
44
+ Other proportions can be obtained by visiting [huihui_ai/qwq-fusion](https://ollama.com/huihui_ai/qwq-fusion/tags).
measurement.json ADDED
The diff for this file is too large to render. See raw diff