|
--- |
|
license: apache-2.0 |
|
pipeline_tag: text-generation |
|
tags: |
|
- safetensors |
|
- mergekit |
|
- merge |
|
- mistral |
|
- not-for-all-audiences |
|
- nsfw |
|
- rp |
|
- roleplay |
|
language: |
|
- en |
|
--- |
|
# This model is recommended for RP, but you can use it as assistant as well. Please, give it a try. This version has lees GPTism. |
|
--- |
|
### Prompt Format: |
|
- **Extended Alpaca Format** As for exemple from [lemonilia/LimaRP-Mistral-7B-v0.1](https://huggingface.co/lemonilia/LimaRP-Mistral-7B-v0.1). |
|
Use *### Response: (length = huge)* for exemple, to increase length. |
|
|
|
### Configuration |
|
|
|
The following YAML configuration was used to produce this model: |
|
|
|
```yaml |
|
models: |
|
- model: .\Endevor_EndlessRP_v1 |
|
- model: kubernetes-bad/good-robot+.\toxic-dpo-v0.1-NoWarning-lora # This removes most of GPTism. The Toxic DPO is a lora I finetuned myself. |
|
parameters: |
|
weight: 0.6 |
|
density: 0.53 |
|
- model: rwitz/go-bruins-v2+Undi95/Mistral-7B-smoll_pippa-lora # Maintein RP stability. |
|
parameters: |
|
weight: 0.4 |
|
density: 0.53 |
|
merge_method: dare_ties |
|
base_model: .\Endevor_EndlessRP_v1 |
|
parameters: |
|
normalize: false |
|
int8_mask: true |
|
dtype: bfloat16 |
|
``` |
|
As this mostly focuses on RP, please don't expect it being smart with riddles or logical tests. |