Mistral 12B merges
Collection
5 items
•
Updated
merge
This is a merge of pre-trained language models
What if we give a badman an education? We'll get a villain.
While testing Badman_12B, it seemed to me that its "intellectual" capabilities is not enough.
This merge aims to inherit understanding of small details from rubenroy/Geneva-12B-GCv2-5m while preserving the features of Badman_12B.
It seems that responces in this merge is slightly shorter.
Purposes - uncensored unalinged chat/roleplay, assistant, non strict instruction folloving.
Tested on T1.01 Use ChatML
- Downloads last month
- 8
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.