--- base_model: - TareksLab/Amethyst-SCE-V1-70B - TareksLab/Carnelian-SCE-V1-70B - TareksLab/Sapphire-V1-70B - TareksLab/Citrine-SCE-V1-70B - TareksLab/Emerald-SCE-V1-70B - TareksLab/Ruby-DL-V1-70B library_name: transformers tags: - mergekit - merge --- # MERGE2 This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit). ## Merge Details ### Merge Method This model was merged using the [DARE TIES](https://arxiv.org/abs/2311.03099) merge method using [TareksLab/Amethyst-SCE-V1-70B](https://huggingface.co/TareksLab/Amethyst-SCE-V1-70B) as a base. ### Models Merged The following models were included in the merge: * [TareksLab/Carnelian-SCE-V1-70B](https://huggingface.co/TareksLab/Carnelian-SCE-V1-70B) * [TareksLab/Sapphire-V1-70B](https://huggingface.co/TareksLab/Sapphire-V1-70B) * [TareksLab/Citrine-SCE-V1-70B](https://huggingface.co/TareksLab/Citrine-SCE-V1-70B) * [TareksLab/Emerald-SCE-V1-70B](https://huggingface.co/TareksLab/Emerald-SCE-V1-70B) * [TareksLab/Ruby-DL-V1-70B](https://huggingface.co/TareksLab/Ruby-DL-V1-70B) ### Configuration The following YAML configuration was used to produce this model: ```yaml models: - model: TareksLab/Emerald-SCE-V1-70B parameters: weight: [0.1, 0.1, 0.1, 0.2, 0.5] density: 0.5 - model: TareksLab/Carnelian-SCE-V1-70B parameters: weight: [0.1, 0.1, 0.2, 0.4, 0.2] density: 0.5 - model: TareksLab/Ruby-DL-V1-70B parameters: weight: [0.1, 0.2, 0.4, 0.2, 0.1] density: 0.5 - model: TareksLab/Amethyst-SCE-V1-70B parameters: weight: [0.2, 0.4, 0.2, 0.1, 0.1] density: 0.5 - model: TareksLab/Citrine-SCE-V1-70B parameters: weight: [0.5, 0.2, 0.1, 0.1, 0.1] density: 0.5 - model: TareksLab/Sapphire-V1-70B parameters: weight: 0.10 density: 0.5 merge_method: dare_ties base_model: TareksLab/Amethyst-SCE-V1-70B parameters: normalize: false int8_mask: true dtype: float32 out_dtype: bfloat16 chat_template: llama3 tokenizer: source: TareksLab/Emerald-SCE-V1-70B pad_to_multiple_of: 8 ```