File size: 1,074 Bytes
cce2dcb
 
 
 
 
 
 
 
 
 
 
 
 
5e8ce16
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
---
language:
- en
pipeline_tag: text-generation
tags:
- mistral
- llama
- roleplay
- merge
---
<img src="logo.png">

This is a GGUF variant of <a href=https://huggingface.co/DZgas/GIGABATEMAN-7B>GIGABATEMAN-7B</a> model. Use with <a href=https://github.com/LostRuins/koboldcpp/releases>koboldcpp</a> (do not use GPT4ALL)

*The five questions for neural networks to understand freedom by ME*
"Q5-LLM-freedom"
| Model               | Q1 | Q2 | Q3 | Q4 | Q5 | 
|---------------------|----|----|----|----|----|
| GIGABATEMAN-7B      |  **+** |  **+** |  **+** |  **+** |  **+** |
| toppy-m-7b          |  **+** |  **+** |  **+** |  **+** |  **+** |
| starling-lm-7b-alpha|  **+** |  **+** |  **+** |  **-** |  **+** |
| openchat-3.5-0106   |  **+** |  **-** |  **+** |  **-** |  **+** |
| Mistral-7B-v0.3     |  **+** |  **+** |  **-** |  **-** |  **+** |
| **Meta-Llama-3-8B**     |  **-** |  **-** |  **-** |  **-** |  **-** |
| **Qwen2-7B**            |  **-** |  **-** |  **-** |  **-** |  **-** |
The list of 5 questions is placed in the QR code:
<img src="Q5-system.png">