base_model:
- nbeerbower/mistral-nemo-bophades-12B
- nbeerbower/mistral-nemo-gutenberg-12B-v3
- nbeerbower/Lyra-Gutenberg-mistral-nemo-12B
- mistralai/Mistral-Nemo-Instruct-2407
- intervitens/mini-magnum-12b-v1.1
- Sao10K/MN-12B-Lyra-v1
license: apache-2.0
library_name: transformers
tags:
- merge
- roleplay
- not-for-all-audiences
Instant-RP-Noodles-12B-v1.2
An RP/Story writing merge of 3 models that all use the mistral prompt and have seen some success at higher context levels. The original models used were Mistral Instruct, Mini-Magnum, and Lyra-v1. These models then had extra dpo/orpo training done on top to enhance there already good creative capabilities. This 1.2 version should fix some problems with formatting that occurred in the initial merge, as well as increase it's smarts a bit too.
Big thanks to the MistralAI team, intervitens and his Anthracite team, along with Sao10K for the original models used, plus nbeerbower for the extra training done as well!
Settings
Temperature @ 0.7
Min-P @ 0.02
Smoothing Factor @ 0.3
DRY Multiplier (plus standard DRY settings) @ 0.8
Skip Special Tokens @ On
Everything else @ Off
Prompt Format: Nemo-Mistral
[INST] user prompt[/INST] character response</s>[INST] user prompt[/INST]
Models Merged
The following models were included in the merge:
https://huggingface.co/nbeerbower/mistral-nemo-bophades-12B
https://huggingface.co/nbeerbower/mistral-nemo-gutenberg-12B-v3
https://huggingface.co/nbeerbower/Lyra-Gutenberg-mistral-nemo-12B