File size: 2,149 Bytes
ca74f73 d78a926 ca74f73 1be8699 a15977f ca74f73 a15977f 1be8699 a15977f ca74f73 |
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 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 |
---
license: llama3
tags:
- merge
- mergekit
- lazymergekit
---
# Cognitron-8B
Cognitron-8B is an experimental large language model (LLM) created by combining three pre-existing models: Llama-3-8B-Lexi-Uncensored, Einstein-v6.1-Llama3-8B, and dolphin-2.9-llama3-8b. This combination aims to achieve a unique blend of capabilities:
* Uncensored Knowledge: By incorporating Llama-3-8B-Lexi-Uncensored, Cognitron-8B has access to a wider range of information without filtering.
* Enhanced Intelligence: The inclusion of Einstein-v6.1-Llama3-8B is intended to boost Cognitron-8B's reasoning and problem-solving abilities.
* Creative Fluency: The dolphin-2.9-llama3-8b component is designed to contribute creativity and unconventional thinking to Cognitron-8B's responses.
It is important to note that combining these models is an experiment, and the resulting performance is unknown.
GGUF: https://huggingface.co/mradermacher/Cognitron-8B-GGUF
Cognitron-8B is a merge of the following models using [mergekit](https://github.com/cg123/mergekit):
## Potential Biases and Limitations
Uncensored Content: Due to the inclusion of uncensored models, Cognitron-8B may generate outputs containing biases, hate speech, or offensive language.
## Importance of Uncensored Models
The inclusion of an uncensored model in Cognitron-8B reflects a growing interest in exploring the potential benefits of unfiltered information for LLMs. Here's why uncensored models are important:
* Comprehensiveness: Unrestricted access to information allows LLMs to capture a more complete picture of the world, even if it includes controversial or sensitive topics.
* Real-World Applicability: In situations where internet access is limited, uncensored LLMs could serve as a valuable source of unfiltered knowledge, allowing users to make informed decisions based on the available data.
## 🧩 Configuration
```yaml
models:
- model: Orenguteng/Llama-3-8B-Lexi-Uncensored
- model: Weyaxi/Einstein-v6.1-Llama3-8B
- model: cognitivecomputations/dolphin-2.9-llama3-8b
merge_method: model_stock
base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored
dtype: bfloat16
``` |