Hugging Face's logo

Khetterman
/
CursedMatrix-8B-v9

Text Generation
Transformers
Safetensors
English
Russian
llama
mergekit
Merge
bfloat16
8b
chat
creative
roleplay
conversational
Not-For-All-Audiences
text-generation-inference
Inference Endpoints
Model card Files Community
2
CursedMatrix-8B-v9 / README.md
Khetterman's picture
Khetterman
Update README.md
cd1c224 verified
preview code
|
raw
history blame contribute delete
14.8 kB
metadata
base_model:
  - bluuwhale/L3-SthenoMaidBlackroot-8B-V1
  - bunnycore/Llama-3.1-8B-OmniMatrix
  - bunnycore/Llama-3.1-8B-TitanFusion-Mix-2.1
  - Casual-Autopsy/L3-Super-Nova-RP-8B
  - Casual-Autopsy/L3-Umbral-Mind-RP-v3.0-8B
  - d0rj/Llama-3-8B-saiga-suzume-ties
  - DavidAU/L3.1-Dark-Planet-SpinFire-Uncensored-8B
  - DreadPoor/CoolerCoder-8B-Model_Stock
  - DreadPoor/L3.1-BaeZel-8B-Della
  - DreadPoor/Trinas_Nectar-8B-model_stock
  - IlyaGusev/saiga_llama3_8b
  - invisietch/EtherealRainbow-v0.3-8B
  - invisietch/L3.1-EtherealRainbow-v1.0-rc1-8B
  - jeiku/Chaos_RP_l3_8B
  - mlabonne/Daredevil-8B
  - MrRobotoAI/Loki-.Epic_Fiction.-8b
  - PJMixers/LLaMa-3-CursedStock-v2.0-8B
  - ResplendentAI/Nymph_8B
  - rityak/L3.1-DarkStock-8B
  - saishf/Neural-SOVLish-Devil-8B-L3
  - saishf/SOVL-Mega-Mash-V2-L3-8B
  - sethuiyer/Dr.Samantha-8B
  - SicariusSicariiStuff/LLAMA-3_8B_Unaligned_BETA
  - v000000/L3-8B-BlueSerpentine
  - v000000/L3.1-Storniitova-8B
  - win10/ArliAI-RPMax-v1.3-merge-8B
  - ZeroXClem/Llama-3-8B-ProLong-SAO-Roleplay-512k
library_name: transformers
tags:
  - mergekit
  - merge
  - bfloat16
  - safetensors
  - 8b
  - chat
  - creative
  - roleplay
  - conversational
  - not-for-all-audiences
language:
  - en
  - ru

CursedMatrix-8B-v9

The long journey from despair to acceptable perfection.

CursedMatrixLogo256.png

This is an interesting merge of 27 cool models, created using mergekit. Enjoy exploring :)

Merge Details

Method

This model was merged using the multistep process and remerge with some model variations for best result.

Models

The following models were included in the merge:

Configuration

The following YAML configurations was used to produce this model:

### ::: Generation 1 merges :

# CursedMatrix-8B-v1
models:
  - model: bunnycore/Llama-3.1-8B-OmniMatrix
    parameters:
      density: [0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1]
      weight:  [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]
  - model: PJMixers/LLaMa-3-CursedStock-v2.0-8B
    parameters:
      density: [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]
      weight:  [0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1]
merge_method: ties
base_model: saishf/SOVL-Mega-Mash-V2-L3-8B
dtype: bfloat16

# TitanPlanet-8B-v1
models:
  - model: bunnycore/Llama-3.1-8B-TitanFusion-Mix-2.1
    parameters:
      density: [0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1]
      weight:  [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]
  - model: MrRobotoAI/Loki-.Epic_Fiction.-8b
    parameters:
      density: [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]
      weight:  [0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1]
merge_method: ties
base_model: DavidAU/L3.1-Dark-Planet-SpinFire-Uncensored-8B
dtype: bfloat16

# NeuralCoder-8B-v1
models:
  - model: sethuiyer/Dr.Samantha-8B
    parameters:
      density: [0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1]
      weight:  [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]
  - model: DreadPoor/CoolerCoder-8B-Model_Stock
    parameters:
      density: [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]
      weight:  [0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1]
merge_method: ties
base_model: saishf/Neural-SOVLish-Devil-8B-L3
dtype: bfloat16

# EtherealNymph-8B-v1
models:
  - model: invisietch/EtherealRainbow-v0.3-8B
  - model: ResplendentAI/Nymph_8B
merge_method: slerp
base_model: invisietch/EtherealRainbow-v0.3-8B
dtype: bfloat16
parameters:
  t: [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.9, 0.1,  0.9, 0.1,  0.9, 0.1, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]

# UmbralDevil-8B-v1
models:
  - model: Casual-Autopsy/L3-Umbral-Mind-RP-v3.0-8B
  - model: mlabonne/Daredevil-8B
merge_method: slerp
base_model: Casual-Autopsy/L3-Umbral-Mind-RP-v3.0-8B
dtype: bfloat16
parameters:
  t: [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.9, 0.1,  0.9, 0.1,  0.9, 0.1, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]

# EvilMind-8B-v1
models:
  - model: mlabonne/Daredevil-8B
    parameters:
      weight:  [1.0, 0.3, 0.1, 0.0]
      density: [0.7, 0.2]
  - model: Casual-Autopsy/L3-Umbral-Mind-RP-v3.0-8B
    parameters:
      weight:  [0.1, 0.9, 0.1]
      density: 0.5
  - model: invisietch/EtherealRainbow-v0.3-8B
    parameters:
      weight:  [0.0, 0.1, 0.3, 1.0]
      density: [0.2, 0.7]
merge_method: della_linear
parameters:
  epsilon: 0.15
  lambda: 1
base_model: ResplendentAI/Nymph_8B
dtype: bfloat16

### ::: Generation 2 merges :

# DevilMind-8B-v1
models:
  - model: F:/EvilMind-8B-v1
  - model: F:/UmbralDevil-8B-v1
merge_method: slerp
base_model: F:/EvilMind-8B-v1
dtype: bfloat16
parameters:
  t: [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.9, 0.1,  0.9, 0.1,  0.9, 0.1, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]

# TitanNymph-8B-v1
models:
  - model: F:/TitanPlanet-8B-v1
  - model: F:/EtherealNymph-8B-v1
merge_method: slerp
base_model: F:/TitanPlanet-8B-v1
dtype: bfloat16
parameters:
  t: [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.9, 0.1,  0.9, 0.1,  0.9, 0.1, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]

# CursedMatrix-8B-v2
models:
  - model: F:/TitanPlanet-8B-v1
    parameters:
      density: [0.5, 0.4, 0.6, 0.3, 0.7, 0.2, 0.8, 0.1, 0.9,  0.1, 0.9, 0.1, 0.9, 0.1, 0.9,  0.1, 0.9, 0.2, 0.8, 0.3, 0.7, 0.4, 0.6, 0.5]
      weight:  [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.9, 0.1,  0.9, 0.1, 0.9, 0.1, 0.9, 0.1,  0.9, 0.1, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]
  - model: F:/NeuralCoder-8B-v1
    parameters:
      density: [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.9, 0.1,  0.9, 0.1, 0.9, 0.1, 0.9, 0.1,  0.9, 0.1, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]
      weight:  [0.5, 0.4, 0.6, 0.3, 0.7, 0.2, 0.8, 0.1, 0.9,  0.1, 0.9, 0.1, 0.9, 0.1, 0.9,  0.1, 0.9, 0.2, 0.8, 0.3, 0.7, 0.4, 0.6, 0.5]
merge_method: dare_ties
base_model: F:/CursedMatrix-8B-v1
dtype: bfloat16

# CursedMatrix-8B-v3
models:
  - model: F:/TitanPlanet-8B-v1
  - model: F:/CursedMatrix-8B-v1
  - model: F:/EtherealNymph-8B-v1
merge_method: model_stock
base_model: F:/CursedMatrix-8B-v2
dtype: bfloat16

### ::: Generation 3 merges :

# CursedMatrix-8B-v4
models:
  - model: F:/CursedMatrix-8B-v3
    parameters:
      weight: 0.8
  - model: F:/TitanNymph-8B-v1
    parameters:
      weight: 0.4
  - model: DreadPoor/Trinas_Nectar-8B-model_stock
    parameters:
      weight: 0.3
  - model: F:/DevilMind-8B-v1
    parameters:
      weight: 0.2
merge_method: task_arithmetic
base_model: F:/CursedMatrix-8B-v3
dtype: bfloat16

# CursedMatrix-8B-v4-rev2
models:
  - model: F:/CursedMatrix-8B-v3
    parameters:
      weight: 0.8
  - model: F:/TitanNymph-8B-v1
    parameters:
      weight: 0.4
  - model: DreadPoor/Trinas_Nectar-8B-model_stock
    parameters:
      weight: 0.3
  - model: F:/DevilMind-8B-v1
    parameters:
      weight: 0.2
merge_method: task_arithmetic
base_model: F:/CursedMatrix-8B-v1
dtype: bfloat16

# CursedMatrix-8B-v5
models:
  - model: F:/CursedMatrix-8B-v4-rev2
merge_method: slerp
base_model: F:/CursedMatrix-8B-v4
dtype: bfloat16
parameters:
  t: [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]

### ::: Generation 4 merges :

# CursedMatrix-8B-v6
models:
  - model: jeiku/Chaos_RP_l3_8B
  - model: ZeroXClem/Llama-3-8B-ProLong-SAO-Roleplay-512k
merge_method: model_stock
base_model: F:/CursedMatrix-8B-v5
dtype: bfloat16

# CursedMatrix-8B-v6-rev2
models:
  - model: win10/ArliAI-RPMax-v1.3-merge-8B
  - model: v000000/L3.1-Storniitova-8B
  - model: d0rj/Llama-3-8B-saiga-suzume-ties
merge_method: model_stock
base_model: F:/CursedMatrix-8B-v5
dtype: bfloat16

# CursedMatrix-8B-v6-rev3
models:
  - model: Casual-Autopsy/L3-Super-Nova-RP-8B
  - model: IlyaGusev/saiga_llama3_8b
merge_method: model_stock
base_model: F:/CursedMatrix-8B-v5
dtype: bfloat16

# CursedMatrix-8B-v7
models:
  - model: F:/CursedMatrix-8B-v6
    parameters:
      weight:  [0.5, 0.4, 0.6, 0.3, 0.7, 0.2, 0.8, 0.2, 0.8, 0.3, 0.7, 0.4, 0.6, 0.5]
      density: [0.05, 0.25]
  - model: F:/CursedMatrix-8B-v6-rev3
    parameters:
      weight:  [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]
      density: [0.25, 0.05]
merge_method: ties
base_model: F:/CursedMatrix-8B-v6-rev2
dtype: bfloat16

# CursedMatrix-8B-v8
models:
  - model: F:/CursedMatrix-8B-v6
  - model: F:/CursedMatrix-8B-v6-rev2
  - model: F:/CursedMatrix-8B-v6-rev3
merge_method: model_stock
base_model: F:/CursedMatrix-8B-v7
dtype: bfloat16

### ::: Generation 5 merges :

# Cursed-DarkRainbow-8B-v1
models:
  - model: invisietch/L3.1-EtherealRainbow-v1.0-rc1-8B
    parameters:
      weight:  [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]
      density: [0.5, 0.4, 0.6, 0.3, 0.7, 0.2, 0.8, 0.2, 0.8, 0.3, 0.7, 0.4, 0.6, 0.5]
  - model: rityak/L3.1-DarkStock-8B
    parameters:
      weight:  [0.5, 0.4, 0.6, 0.3, 0.7, 0.2, 0.8, 0.2, 0.8, 0.3, 0.7, 0.4, 0.6, 0.5]
      density: [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]
merge_method: della
parameters:
  epsilon: 0.123456789
  lambda:  0.987654321
base_model: F:/CursedMatrix-8B-v8
dtype: bfloat16

# Cursed-BlueBaezel-8B-v1
models:
  - model: v000000/L3-8B-BlueSerpentine
    parameters:
      weight:  [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]
      density: [0.5, 0.4, 0.6, 0.3, 0.7, 0.2, 0.8, 0.2, 0.8, 0.3, 0.7, 0.4, 0.6, 0.5]
  - model: DreadPoor/L3.1-BaeZel-8B-Della
    parameters:
      weight:  [0.5, 0.4, 0.6, 0.3, 0.7, 0.2, 0.8, 0.2, 0.8, 0.3, 0.7, 0.4, 0.6, 0.5]
      density: [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]
merge_method: della
parameters:
  epsilon: 0.123456789
  lambda:  0.987654321
base_model: F:/CursedMatrix-8B-v8
dtype: bfloat16

# Cursed-SuzumeMaid-8B-v1
models:
  - model: d0rj/Llama-3-8B-saiga-suzume-ties
    parameters:
      weight:  [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]
      density: [0.5, 0.4, 0.6, 0.3, 0.7, 0.2, 0.8, 0.2, 0.8, 0.3, 0.7, 0.4, 0.6, 0.5]
  - model: bluuwhale/L3-SthenoMaidBlackroot-8B-V1
    parameters:
      weight:  [0.5, 0.4, 0.6, 0.3, 0.7, 0.2, 0.8, 0.2, 0.8, 0.3, 0.7, 0.4, 0.6, 0.5]
      density: [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]
merge_method: della
parameters:
  epsilon: 0.123456789
  lambda:  0.987654321
base_model: F:/CursedMatrix-8B-v8
dtype: bfloat16

# Cursed-UnalignedSaiga-8B-v1
models:
  - model: SicariusSicariiStuff/LLAMA-3_8B_Unaligned_BETA
    parameters:
      weight:  [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]
      density: [0.5, 0.4, 0.6, 0.3, 0.7, 0.2, 0.8, 0.2, 0.8, 0.3, 0.7, 0.4, 0.6, 0.5]
  - model: IlyaGusev/saiga_llama3_8b
    parameters:
      weight:  [0.5, 0.4, 0.6, 0.3, 0.7, 0.2, 0.8, 0.2, 0.8, 0.3, 0.7, 0.4, 0.6, 0.5]
      density: [0.5, 0.6, 0.4, 0.7, 0.3, 0.8, 0.2, 0.8, 0.2, 0.7, 0.3, 0.6, 0.4, 0.5]
merge_method: della
parameters:
  epsilon: 0.123456789
  lambda:  0.987654321
base_model: F:/CursedMatrix-8B-v8
dtype: bfloat16

# CursedMatrix-8B-v9
# Final model...
models:
  - model: F:/Cursed-UnalignedSaiga-8B-v1
  - model: F:/Cursed-DarkRainbow-8B-v1
  - model: F:/Cursed-BlueBaezel-8B-v1
  - model: F:/Cursed-SuzumeMaid-8B-v1
merge_method: model_stock
base_model: F:/CursedMatrix-8B-v8
dtype: bfloat16

My thanks to the authors of the original models, your work is incredible. Have a good time 🖤