| license: apache-2.0 | |
| tags: | |
| - merge | |
| - mergekit | |
| - lazymergekit | |
| - Orenguteng/Llama-3-8B-Lexi-Uncensored | |
| - aifeifei798/llama3-8B-DarkIdol-2.3-Uncensored-32K | |
| # Llama-3.1-8B-Uncensored | |
| Llama-3.1-8B-Uncensored is a merge of the following models using [mergekit](https://github.com/cg123/mergekit): | |
| * [Orenguteng/Llama-3-8B-Lexi-Uncensored](https://huggingface.co/Orenguteng/Llama-3-8B-Lexi-Uncensored) | |
| * [aifeifei798/llama3-8B-DarkIdol-2.3-Uncensored-32K](https://huggingface.co/aifeifei798/llama3-8B-DarkIdol-2.3-Uncensored-32K) | |
| ## 🧩 Configuration | |
| \```yaml | |
| models: | |
| - model: meta-llama/Meta-Llama-3.1-8B-Instruct | |
| - model: Orenguteng/Llama-3-8B-Lexi-Uncensored | |
| parameters: | |
| density: 0.53 | |
| weight: 0.6 | |
| - model: aifeifei798/llama3-8B-DarkIdol-2.3-Uncensored-32K | |
| parameters: | |
| density: 0.53 | |
| weight: 0.4 | |
| merge_method: dare_ties | |
| base_model: meta-llama/Meta-Llama-3.1-8B-Instruct | |
| parameters: | |
| int8_mask: true | |
| dtype: bfloat16 | |
| \``` |