--- license: cc-by-nc-4.0 tags: - not-for-all-audiences - nsfw language: - en --- ## Miqu PlayMaid 70B v0.1
Image to be replaced :P
This model uses the Alpaca **prompting format**. Unlike the base model this one is tuned on **16384 sequence length**. This was due to memory constraints. This is a roleplay focused finetune using MiquMaid v2 70B DPO by [Undi95](https://huggingface.co/Undi95) and [IkariDev](https://huggingface.co/IkariDev) as base. My goal was to further push MiquMaid into creative writing and roleplay, hopefully adding some spice to it. Ultimately I want to follow Undi's and Ikari's approach and combine it into a 2x70B alongside their Miqu 70B DPO model. ## Credits: - Netrve - Special thanks to everyone on TheBloke's Discord who helped me! Credits for MiquMaid: - Undi - IkariDev ## Description This repo contains BFP16 files of Miqu-PlayMaid-70B-v0.1 ## Training data used: - [AndrijDavid Roleplay](https://huggingface.co/datasets/Undi95/andrijdavid_roleplay-conversation-sharegpt) - [LimaRP Augmented](https://huggingface.co/datasets/grimulkan/LimaRP-augmented) - [Airoboros 3.2](https://huggingface.co/datasets/jondurbin/airoboros-3.2) - [Bluemoon Fandom RP](https://huggingface.co/datasets/Squish42/bluemoon-fandom-1-1-rp-cleaned) ### Format: I can only guarantee that it works with a maximum context length of 16384. Might work with 32768, but no promises. ``` ### Instruction: {system prompt} ### Input: {input} ### Response: {reply} ``` ## Support - Netrve: If you like what I did, feel free to support future work [here](https://ko-fi.com/netrve) Don't forget to send some love to the original masterminds behind MiquMaid: - Undi: If you want to support us, you can [here](https://ko-fi.com/undiai) - IkariDev: Visit my [retro/neocities style website](https://ikaridevgit.github.io/) please kek