| license: apache-2.0 | |
| tags: | |
| - merge | |
| - mergekit | |
| - lazymergekit | |
| - ProbeMedicalYonseiMAILab/medllama3-v20 | |
| - Llama-3.1-8B-instruction5_r256_e4-slerp-0.5 | |
| # Llama-3.1-EIRAI-8B-Prob | |
| Llama-3.1-EIRAI-8B-Prob is a merge of the following models using [mergekit](https://github.com/cg123/mergekit): | |
| * [ProbeMedicalYonseiMAILab/medllama3-v20](https://huggingface.co/ProbeMedicalYonseiMAILab/medllama3-v20) | |
| * [Llama-3.1-8B-instruction5_r256_e4-slerp-0.5](https://huggingface.co/Llama-3.1-8B-instruction5_r256_e4-slerp-0.5) | |
| ## 🧩 Configuration | |
| \```yaml | |
| slices: | |
| - sources: | |
| - model: ProbeMedicalYonseiMAILab/medllama3-v20 | |
| layer_range: | |
| - 0 | |
| - 32 | |
| - model: Llama-3.1-8B-instruction5_r256_e4-slerp-0.5 | |
| layer_range: | |
| - 0 | |
| - 32 | |
| merge_method: slerp | |
| base_model: ProbeMedicalYonseiMAILab/medllama3-v20 | |
| parameters: | |
| t: | |
| - value: 0.5 | |
| dtype: bfloat16 | |
| \``` |