--- base_model: - Vortex5/Astral-Noctra-12B - Vortex5/Azure-Starlight-12B - Vortex5/Crimson-Constellation-12B - Vortex5/Red-Synthesis-12B - Vortex5/Shining-Seraph-12B - Vortex5/Starlit-Shadow-12B - Vortex5/Vermilion-Sage-12B - Vortex5/Scarlet-Seraph-12B - Vortex5/Maroon-Sunset-12B - Vortex5/Amber-Starlight-12B language: - en library_name: transformers license: apache-2.0 tags: - karcher - merge - mergekit - mistral - nemo widget: - text: "Riemannian-Redshift-12B-v1" output: url: https://cdn-uploads.huggingface.co/production/uploads/68e840caa318194c44ec2a04/-5OsIfrWlUxoJZP4U83uq.png --- > [!NOTE] > ⚠️ Note: This model requires **Mistral Tekken** chat template. > # 🌌 Riemannian Redshift 12B v1 This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit). ![Redshift](https://cdn-uploads.huggingface.co/production/uploads/68e840caa318194c44ec2a04/-5OsIfrWlUxoJZP4U83uq.png) ## Merge Details ### Merge Method This is an experimental `karcher` merge of several high quality [Vortex5](https://huggingface.co/Vortex5) models. I used `float32` precision and `max_iter: 1000` to ensure the best bits were chosen for the Riemannian center. This merge took 5 hours using [graph_v18](https://huggingface.co/spaces/Naphula/model_tools/blob/main/graph_v18.py) as an accelerant with 8GB VRAM. This model was merged using the [Karcher Mean](https://en.wikipedia.org/wiki/Karcher_mean) merge method. ### Models Merged The following models were included in the merge: * [Vortex5/Maroon-Sunset-12B](https://huggingface.co/Vortex5/Maroon-Sunset-12B) * [Vortex5/Azure-Starlight-12B](https://huggingface.co/Vortex5/Azure-Starlight-12B) * [Vortex5/Scarlet-Seraph-12B](https://huggingface.co/Vortex5/Scarlet-Seraph-12B) * [Vortex5/Amber-Starlight-12B](https://huggingface.co/Vortex5/Amber-Starlight-12B) * [Vortex5/Shining-Seraph-12B](https://huggingface.co/Vortex5/Shining-Seraph-12B) * [Vortex5/Red-Synthesis-12B](https://huggingface.co/Vortex5/Red-Synthesis-12B) * [Vortex5/Starlit-Shadow-12B](https://huggingface.co/Vortex5/Starlit-Shadow-12B) * [Vortex5/Crimson-Constellation-12B](https://huggingface.co/Vortex5/Crimson-Constellation-12B) * [Vortex5/Vermilion-Sage-12B](https://huggingface.co/Vortex5/Vermilion-Sage-12B) * [Vortex5/Astral-Noctra-12B](https://huggingface.co/Vortex5/Astral-Noctra-12B) ### Configuration The following YAML configuration was used to produce this model: ```yaml models: - model: B:/12B/models--Vortex5--Astral-Noctra-12B - model: B:/12B/models--Vortex5--Azure-Starlight-12B - model: B:/12B/models--Vortex5--Crimson-Constellation-12B - model: B:/12B/models--Vortex5--Red-Synthesis-12B - model: B:/12B/models--Vortex5--Shining-Seraph-12B - model: B:/12B/models--Vortex5--Starlit-Shadow-12B - model: B:/12B/models--Vortex5--Vermilion-Sage-12B - model: B:/12B/models--Vortex5--Scarlet-Seraph-12B - model: B:/12B/models--Vortex5--Maroon-Sunset-12B - model: B:/12B/models--Vortex5--Amber-Starlight-12B merge_method: karcher parameters: max_iter: 1000 tol: 1.0e-9 dtype: float32 out_dtype: bfloat16 tokenizer: source: union chat_template: auto name: 🌌 Riemannian-Redshift-12B-v1 ```