RicardoEstep/RPBizkit-v11-12B-Base

"Experimental Mistral Nemo Mix" using... "Model Stock merge with Mergekit".

This is a mix of some known "RP Uncensored" models. I'm not responsible of the use of this.

I did it all for the nookie, so you can take that cookie, and stick it up your...

After noticing a kind of "Assistant Tone" in V10, i remade the mix using "a cleaner" list... It works pretty nice and stable, and it's the "Base" for something even more experimental.

Models Merged:

• TheDrummer/Rocinante-X-12B-v1 as base.
• ReadyArt/Dark-Nexus-12B-v2.0
• anthracite-org/magnum-v2-12b
• NeverSleep/Lumimaid-v0.2-12B
• MuXodious/Wayfarer-2-12B-absolute-heresy
• SicariusSicariiStuff/Impish_Bloodmoon_12B
• allura-org/MN-12b-RP-Ink
• DreadPoor/Krix-12B-Model_Stock
• ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2
• ChaoticNeutrals/Nera_Noctis-12B

Configuration Used:

merge_method: model_stock
base_model: TheDrummer/Rocinante-X-12B-v1
tokenizer_source: base
parameters:
  normalize: true
  int8_mask: false
  filter_wise: true
models:
  - model: anthracite-org/magnum-v2-12b
  - model: MuXodious/Wayfarer-2-12B-absolute-heresy
  - model: NeverSleep/Lumimaid-v0.2-12B
  - model: ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2
  - model: allura-org/MN-12b-RP-Ink
  - model: ReadyArt/Dark-Nexus-12B-v2.0
  - model: SicariusSicariiStuff/Impish_Bloodmoon_12B
  - model: ChaoticNeutrals/Nera_Noctis-12B
  - model: DreadPoor/Krix-12B-Model_Stock
dtype: bfloat16

Final Notes;

The Tokenizer & Embedding Sizes (131072) are clean + It's "supposed to" support now a full "128K Context Size" (patched at 1M).

The chat template is added with "full mistral" support.

Rapid Benchmarking - "64-Shot + 5318008 Seed" Evaluation.

These are the results of running "lm-evaluation-harness", using the "Q8_0 Model", with an "8K Context Window". Executed on a free "Google Collab - CUDA-accelerated backend (T4)" instance.

• Quantized Model, Here RicardoEstep/RPBizkit-v11-12B-Base-GGUF.