Elfrino/AndroidPrincess-20B
016
1---2base_model:3- Elfrino/XwinXtended-20B4- seb-c/Psydestroyer-20B5library_name: transformers6tags:7- mergekit8- merge9 10---11 12**NOTES:**13**LOOKS PROMISING SO FAR..... IN TESTING......***14 15 16 17 18***Some findings: Articulate, long flowy sentences and can be pushed to high temps. Lacks the zany outlandish creativity***19***of PsymedRP-20B or Xwin-MLewd-20B but it's still a solid story weaver..***20 21 22 23##################################################################################################################24 25# merge26 27This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).28 29## Merge Details30### Merge Method31 32This model was merged using the SLERP merge method.33 34### Models Merged35 36The following models were included in the merge:37* [Elfrino/XwinXtended-20B](https://huggingface.co/Elfrino/XwinXtended-20B)38* [seb-c/Psydestroyer-20B](https://huggingface.co/seb-c/Psydestroyer-20B)39 40### Configuration41 42The following YAML configuration was used to produce this model:43 44```yaml45NEXT TRIAL: make WizardLM into a 20B and SLERP it with XwinXtended? maybe....46 47 48 49slices:50 - sources:51 - model: seb-c/Psydestroyer-20B52 layer_range: [0, 62]53 - model: Elfrino/XwinXtended-20B54 layer_range: [0, 62]55merge_method: slerp56base_model: seb-c/Psydestroyer-20B57parameters:58 t:59 - filter: self_attn60 value: [0.8, 0.8, 0.9, 0.7, .8]61 - filter: mlp62 value: [.8, 0.8, 0.9, 0.8, .7]63 - value: 0.336964dtype: bfloat1665```66