|
--- |
|
license: other |
|
language: |
|
- en |
|
tags: |
|
- rp |
|
- erp |
|
- chat |
|
- storywriting |
|
- not-for-all-audiences |
|
- nsfw |
|
- rotating-stack-merge |
|
--- |
|
# La Dame Blanche 103b |
|
|
|
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/65a531bc7ec6af0f95c707b1/aO6n3uOtyUllpwefLMapa.jpeg) |
|
|
|
This model is a rotating-stack merge of three Miqu-based models in a 103b (120 layer) configuration. The result of |
|
this "frankenmerge" is a large model that contains what I consider to be the best of the spicy Miqu finetunes. |
|
|
|
Component models for the rotating stack are |
|
- NeverSleep/MiquMaid-v2-70B-DPO |
|
- sophosympatheia/Midnight-Miqu-70B-v1.5 |
|
- cookinai/OrcaHermes-Mistral-70B-miqu |
|
|
|
|
|
As all components are Miqu-based, coherency out to 32k context appears to be good. |
|
|
|
This model is uncensored and perfectly capable of generating objectionable material. It does not seem to have a |
|
propensity to insert NSFW content into SFW prompts, but YMMV. As with any LLM, no factual claims |
|
made by the model should be taken at face value. You know that boilerplate safety disclaimer that most professional models have? |
|
Assume this has it too. This model is for entertainment purposes only. |
|
|
|
Vanilla GGUFs: |
|
iMatrix GGUFs: |
|
|
|
If you create additional quants, please let me know and I will link them here as well. |
|
|
|
|
|
# Sample output |
|
|
|
``` |
|
{{[INPUT]}} |
|
Write a detailed and humorous story about a cute and fluffy bunny that goes to a Gwar concert. |
|
{{[OUTPUT]}} |
|
|
|
``` |
|
|
|
|
|
# WTF is a rotating-stack merge? |
|
Inspired by Undi's experiments with stacked merges, Jeb Carter found that output quality and model initiative could be significantly |
|
improved by reversing the model order in the stack, and then doing a linear merge between the original and reversed stacks. That is |
|
what I did here. I created three passthrough stacked merges using the three source models (rotating the model order in each stack), |
|
and then doing a linear merge of all three stacks. The exact merge configs can be found in the recipe.txt file. |
|
|