ABX-AI/Cerebral-Infinity-7B

1---2base_model:3- AetherResearch/Cerebrum-1.0-7b4- Endevor/InfinityRP-v1-7B5library_name: transformers6tags:7- mergekit8- merge9 10---11 12![image/png](https://cdn-uploads.huggingface.co/production/uploads/65d936ad52eca001fdcd3245/FujeSOGQBUSCz1KbvbyUt.png)13 14# Cerebral-Infinity-7B15 16The concept behind this merge is to create a "smarter" RP model, based on the already great InfinityRP, with the added logical reasoning qualities of Cerebrum.17 18I believe it has achieved that based on testing I did, however it may still offer refusals, but they are easy to work around with character cards and good prompting.19 20 21## Merge Details22 23This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).24 25### Merge Method26 27This model was merged using the SLERP merge method.28 29### Models Merged30 31The following models were included in the merge:32* [AetherResearch/Cerebrum-1.0-7b](https://huggingface.co/AetherResearch/Cerebrum-1.0-7b)33* [Endevor/InfinityRP-v1-7B](https://huggingface.co/Endevor/InfinityRP-v1-7B)34 35### Configuration36 37The following YAML configuration was used to produce this model:38 39```yaml40slices:41  - sources:42      - model: Endevor/InfinityRP-v1-7B43        layer_range: [0, 32]44      - model: AetherResearch/Cerebrum-1.0-7b45        layer_range: [0, 32]46merge_method: slerp47base_model: Endevor/InfinityRP-v1-7B48parameters:49  t:50    - filter: self_attn51      value: [0.7, 0.3, 0.6, 0.2, 0.5]52    - filter: mlp53      value: [0.3, 0.7, 0.4, 0.8, 0.5]54    - value: 0.555dtype: bfloat1656 57```58