| license: apache-2.0 | |
| tags: | |
| - merge | |
| - mergekit | |
| - lazymergekit | |
| - aifeifei798/llama3-8B-DarkIdol-2.1-Uncensored-32K | |
| - DevsDoCode/LLama-3-8b-Uncensored | |
| # Llama-uncensored-task-arithmatic | |
| Llama-uncensored-task-arithmatic is a merge of the following models using [mergekit](https://github.com/cg123/mergekit): | |
| * [aifeifei798/llama3-8B-DarkIdol-2.1-Uncensored-32K](https://huggingface.co/aifeifei798/llama3-8B-DarkIdol-2.1-Uncensored-32K) | |
| * [DevsDoCode/LLama-3-8b-Uncensored](https://huggingface.co/DevsDoCode/LLama-3-8b-Uncensored) | |
| ## 🧩 Configuration | |
| \```yaml | |
| models: | |
| - model: aifeifei798/llama3-8B-DarkIdol-2.1-Uncensored-32K | |
| parameters: | |
| density: [1, 0.7, 0.1] # density gradient | |
| weight: 1.0 | |
| - model: DevsDoCode/LLama-3-8b-Uncensored | |
| parameters: | |
| density: 0.5 | |
| weight: [0, 0.3, 0.7, 1] # weight gradient | |
| merge_method: ties | |
| base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored | |
| parameters: | |
| normalize: true | |
| int8_mask: true | |
| dtype: float16 | |
| \``` |