--- license: apache-2.0 base_model: - TheDrummer/Rocinante-12B-v1 - TheDrummer/Rocinante-12B-v1.1 - TheDrummer/Rocinante-X-12B-v1 - TheDrummer/UnslopNemo-12B-v4.1 - TheDrummer/Rivermind-Lux-12B-v1 tags: - merge - mergekit - lazymergekit - TheDrummer/Rocinante-12B-v1 - TheDrummer/Rocinante-12B-v1.1 - TheDrummer/Rocinante-X-12B-v1 - TheDrummer/UnslopNemo-12B-v4.1 - TheDrummer/Rivermind-Lux-12B-v1 --- # Riverfish-Rocinante-12B Riverfish-Rocinante-12B is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing): * [TheDrummer/Rocinante-12B-v1](https://huggingface.co/TheDrummer/Rocinante-12B-v1) * [TheDrummer/Rocinante-12B-v1.1](https://huggingface.co/TheDrummer/Rocinante-12B-v1.1) * [TheDrummer/Rocinante-X-12B-v1](https://huggingface.co/TheDrummer/Rocinante-X-12B-v1) * [TheDrummer/UnslopNemo-12B-v4.1](https://huggingface.co/TheDrummer/UnslopNemo-12B-v4.1) * [TheDrummer/Rivermind-Lux-12B-v1](https://huggingface.co/TheDrummer/Rivermind-Lux-12B-v1) ## 🧩 Configuration ```yamlmerge_method: sequence sequence: - merge_method: model_stock base_model: TheDrummer/Rocinante-12B-v1 models: - model: TheDrummer/Rocinante-12B-v1.1 - model: TheDrummer/Rocinante-X-12B-v1 parameters: normalize: true dtype: bfloat16 output: ./merge-1 # intermediate output, used later - merge_method: karcher base_model: ./merge-1 # reference previous step's output models: - model: ./merge-1 - model: TheDrummer/UnslopNemo-12B-v4.1 parameters: max_iter: 500 tol: 1e-6 normalize: true dtype: bfloat16 output: ./merge-2 - merge_method: arcee_fusion base_model: ./merge-2 models: - model: ./merge-2 - model: TheDrummer/Rivermind-Lux-12B-v1 dtype: bfloat16 output: ./Riverfish-Rocinante-12B # final model directory