add measurement.json

README.md

@@ -0,0 +1,44 @@
+---
+license: apache-2.0
+license_link: https://huggingface.co/huihui-ai/QwQ-32B-Coder-Fusion-8020/blob/main/LICENSE
+language:
+- en
+pipeline_tag: text-generation
+base_model: 
+ - huihui-ai/QwQ-32B-Coder-Fusion-8020
+tags:
+- chat
+- abliterated
+- uncensored
+library_name: transformers
+---
+
+# huihui-ai/QwQ-32B-Coder-Fusion-8020
+
+
+## Overview
+`QwQ-32B-Coder-Fusion-8020` 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 
+[huihui-ai/Qwen2.5-Coder-32B-Instruct-abliterated](https://huggingface.co/huihui-ai/Qwen2.5-Coder-32B-Instruct-abliterated).   
+The weights are blended in a 8:2 ratio, with 80% of the weights from QwQ-32B-Preview-abliterated and 20% from the abliterated Qwen2.5-Coder-32B-Instruct-abliterated model.
+**Although it's a simple mix, the model is usable, and no gibberish has appeared**.
+This is an experiment. I test the [9:1](https://huggingface.co/huihui-ai/QwQ-32B-Coder-Fusion-9010), 
+[8:2](https://huggingface.co/huihui-ai/QwQ-32B-Coder-Fusion-8020), 
+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.
+
+## Model Details
+- **Base Models:**
+  - [huihui-ai/QwQ-32B-Preview-abliterated](https://huggingface.co/huihui-ai/QwQ-32B-Preview-abliterated) (80%)
+  - [huihui-ai/Qwen2.5-Coder-32B-Instruct-abliterated](https://huggingface.co/huihui-ai/Qwen2.5-Coder-32B-Instruct-abliterated) (20%)
+- **Model Size:** 32B parameters
+- **Architecture:** Qwen 2.5
+- **Mixing Ratio:** 8:2 (QwQ-32B-Preview-abliterated:Qwen2.5-Coder-32B-Instruct-abliterated)
+
+
+## ollama
+
+You can use [huihui_ai/qwq-fusion:32b-8020](https://ollama.com/huihui_ai/qwq-fusion:32b-8020) directly, 
+```
+ollama run huihui_ai/qwq-fusion:32b-8020
+```
+
+Other proportions can be obtained by visiting [huihui_ai/qwq-fusion](https://ollama.com/huihui_ai/qwq-fusion/tags).