Kaoeiri's picture
Update README.md
4ea69ee verified
metadata
base_model:
  - byroneverson/Mistral-Small-Instruct-2409-abliterated
  - rAIfle/Acolyte-LORA
  - InferenceIllusionist/SorcererLM-22B
  - allura-org/MS-Meadowlark-22B
  - Saxo/Linkbricks-Horizon-AI-Japanese-Superb-V1-22B
  - ArliAI/Mistral-Small-22B-ArliAI-RPMax-v1.1
  - Darkknight535/MS-Moonlight-22B-v3
  - crestf411/MS-sunfall-v0.7.0
  - TheDrummer/Cydonia-22B-v1.2
  - TheDrummer/Cydonia-22B-v1.1
  - >-
    hf-100/Mistral-Small-Instruct-2409-Spellbound-StoryWriter-22B-instruct-0.4-chkpt-336-16bit
  - Gryphe/Pantheon-RP-Pure-1.6.2-22b-Small
  - unsloth/Mistral-Small-Instruct-2409
  - anthracite-org/magnum-v4-22b
  - TheDrummer/Cydonia-22B-v1.3
  - unsloth/Mistral-Small-Instruct-2409
  - Kaoeiri/Moingooistrial-22B-V1-Lora
  - TroyDoesAI/BlackSheep-MermaidMistral-22B
  - spow12/ChatWaifu_v2.0_22B
library_name: transformers
tags:
  - mergekit
  - merge

image/webp


🌟 Model Merge Overview 🌟

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


πŸ”§ Merge Method

This model was merged using the DARE and TIES merge methods, with unsloth/Mistral-Small-Instruct-2409 as the base model.


🧠 Models Merged

The following models were included in the merge, combining the power and capabilities of various high-performance language models:


βš™οΈ Configuration Details

Here is the YAML configuration used to create this powerful merged model:

models:
  - model: anthracite-org/magnum-v4-22b
    parameters:
      weight: 1.0
      density: 0.85
  - model: TheDrummer/Cydonia-22B-v1.3
    parameters:
      weight: 0.24
      density: 0.69
  - model: TheDrummer/Cydonia-22B-v1.2
    parameters:
      weight: 0.14
      density: 0.67
  - model: TheDrummer/Cydonia-22B-v1.1
    parameters:
      weight: 0.16
      density: 0.67
  - model: Gryphe/Pantheon-RP-Pure-1.6.2-22b-Small
    parameters:
      weight: 0.26
      density: 0.75
  - model: allura-org/MS-Meadowlark-22B
    parameters:
      weight: 0.27
      density: 0.71
  - model: spow12/ChatWaifu_v2.0_22B
    parameters:
      weight: 0.27
      density: 0.7
  - model: Saxo/Linkbricks-Horizon-AI-Japanese-Superb-V1-22B
    parameters:
      weight: 0.2
      density: 0.58
  - model: crestf411/MS-sunfall-v0.7.0
    parameters:
      weight: 0.22
      density: 0.71
  - model: byroneverson/Mistral-Small-Instruct-2409-abliterated+rAIfle/Acolyte-LORA
    parameters:
      weight: 0.24
      density: 0.7
  - model: InferenceIllusionist/SorcererLM-22B
    parameters:
      weight: 0.21
      density: 0.72
  - model: unsloth/Mistral-Small-Instruct-2409+Kaoeiri/Moingooistrial-22B-V1-Lora
    parameters:
      weight: 0.32
      density: 0.76
  - model: ArliAI/Mistral-Small-22B-ArliAI-RPMax-v1.1
    parameters:
      weight: 0.12
      density: 0.65
  - model: Darkknight535/MS-Moonlight-22B-v3
    parameters:
      weight: 0.12
      density: 0.62
  - model: hf-100/Mistral-Small-Instruct-2409-Spellbound-StoryWriter-22B-instruct-0.4-chkpt-336-16bit
    parameters:
      weight: 0.28
      density: 0.74
  - model: TroyDoesAI/BlackSheep-MermaidMistral-22B
    parameters:
      weight: 0.24
      density: 0.7

merge_method: dare_ties
base_model: unsloth/Mistral-Small-Instruct-2409
parameters:
  density: 0.85
  epsilon: 0.09
  lambda: 1.22
dtype: bfloat16

πŸ™ Special Thanks

A huge THANK YOU to the authors and creators of these base models, whose work and dedication make this possible. We deeply appreciate their contributions to the machine learning community.

Special acknowledgment to Hugging Face for providing a platform where these models are shared and developed.

Gratitude to the creators behind the DARE and TIES techniques for pushing the boundaries of model merging.

Without their efforts, this project would not have been possible. πŸš€