Update README.md

README.md

@@ -8,53 +8,18 @@ library_name: transformers
 tags:
 - mergekit
 - merge
-
+- roleplay
+- creative
+language:
+- en
+- ru
 ---
 # merge
 
-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 [TIES](https://arxiv.org/abs/2306.01708) merge method using [RefalMachine/RuadaptQwen2.5-32B-Pro-Beta](https://huggingface.co/RefalMachine/RuadaptQwen2.5-32B-Pro-Beta) as a base.
-
-### Models Merged
-
-The following models were included in the merge:
-* [Sao10K/32B-Qwen2.5-Kunou-v1](https://huggingface.co/Sao10K/32B-Qwen2.5-Kunou-v1)
-* [ParasiticRogue/EVA-Instruct-32B-v2](https://huggingface.co/ParasiticRogue/EVA-Instruct-32B-v2)
-* [allura-org/Qwen2.5-32b-RP-Ink](https://huggingface.co/allura-org/Qwen2.5-32b-RP-Ink)
-
-### Configuration
+This is a merge of pre-trained qwen2.5 32B language models 
 
-The following YAML configuration was used to produce this model:
+Goal of this merge was to create big (on my standarts) roleplay model capable to russian.
 
-```yaml
-models:
-  - model: ParasiticRogue/EVA-Instruct-32B-v2
-    parameters:
-      density: 0.3
-      weight: 0.3
-  - model: allura-org/Qwen2.5-32b-RP-Ink
-    parameters:
-      density: 0.3
-      weight: 0.3
-  - model: Sao10K/32B-Qwen2.5-Kunou-v1
-    parameters:
-      density: 0.3
-      weight: 0.3
-  - model: RefalMachine/RuadaptQwen2.5-32B-Pro-Beta
-    parameters:
-      density: 0.8
-      weight: 0.8
+Well, model is good in roleplay and capable to ru text. Also, during my quick tests ( ~100 answers, because generation is too slow on my PC) model seemed to be creative, uncensored, and smart. 
 
-merge_method: ties
-base_model: RefalMachine/RuadaptQwen2.5-32B-Pro-Beta
-parameters:
-  normalize: false
-  int8_mask: true
-dtype: float16
-tokenizer:
- source: ParasiticRogue/EVA-Instruct-32B-v2
-```
+For now i stop merge or test models bigger than 24B, my hardware is not ready. yet.