Rose-Engine

This is a merge of pre-trained language models created using mergekit.

Dans-PersonalityEngine-V1.3.0 merged with a highly furry and NSFW-biased model. See Rosier's datasets to see exactly how cursed this is gonna be.

This should retain most of what makes Dans-PersonalityEngine good while understanding furry content better.

Very strong weight is used with Rosier because there should be little to no overlap between models.

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using mistralai/Mistral-Nemo-Base-2407 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

name: rose-engine-dt
merge_method: dare_ties
base_model: mistralai/Mistral-Nemo-Base-2407
models:
  # Base model
  - model: PocketDoc/Dans-PersonalityEngine-V1.3.0-12b
    parameters:
      weight: 1.0
  # Furry knowledge
  - model: Fizzarolli/MN-12b-Rosier-v1
    parameters:
      weight: 0.8
      density: 0.4
tokenizer:
  source: PocketDoc/Dans-PersonalityEngine-V1.3.0-12b
Downloads last month
8
Safetensors
Model size
12B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Snqwii/Rose-Engine-12B

Collection including Snqwii/Rose-Engine-12B

Paper for Snqwii/Rose-Engine-12B