base_model:
- Novaciano/LAMED
- Novaciano/VAV
- Novaciano/TAV
- Novaciano/YOD
- Novaciano/NUN-FINAL
- Novaciano/BAPHOMET
library_name: transformers
tags:
- mergekit
- merge
- abliterated
- uncensored
- llama
- llama3.2
- not-for-all-audiences
language:
- en
- es
model-index:
- name: Sigil-Of-Satan-3.2-1B
results:
- task:
type: text-generation
name: Text Generation
dataset:
name: IFEval (0-Shot)
type: wis-k/instruction-following-eval
split: train
args:
num_few_shot: 0
metrics:
- type: inst_level_strict_acc and prompt_level_strict_acc
value: 54.94
name: averaged accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=Novaciano%2FSigil-Of-Satan-3.2-1B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: BBH (3-Shot)
type: SaylorTwift/bbh
split: test
args:
num_few_shot: 3
metrics:
- type: acc_norm
value: 9.4
name: normalized accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=Novaciano%2FSigil-Of-Satan-3.2-1B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MATH Lvl 5 (4-Shot)
type: lighteval/MATH-Hard
split: test
args:
num_few_shot: 4
metrics:
- type: exact_match
value: 5.44
name: exact match
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=Novaciano%2FSigil-Of-Satan-3.2-1B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: GPQA (0-shot)
type: Idavidrein/gpqa
split: train
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 1.45
name: acc_norm
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=Novaciano%2FSigil-Of-Satan-3.2-1B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MuSR (0-shot)
type: TAUR-Lab/MuSR
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 1.42
name: acc_norm
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=Novaciano%2FSigil-Of-Satan-3.2-1B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MMLU-PRO (5-shot)
type: TIGER-Lab/MMLU-Pro
config: main
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 9.5
name: accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=Novaciano%2FSigil-Of-Satan-3.2-1B
name: Open LLM Leaderboard
merge
This is a merge of pre-trained language models created using mergekit.
Merge Details
🇬🇧 English
Remixed version of HarmfulProject-3.2-1B model.
I'll note that there's nothing unusual about it other than being a mix of mixes of uncensored and abliterated models.
The difference with HarmfulProject-3.2-1B is that instead of being mixed in one model, the same models have been mixed in smaller mixes in this model.
Author's Note:
I'm not responsible for the content of the model since I only made the mix, I didn't inject any dataset into it... yet.
Others models created with this merge
- Cultist-3.2-1B - Sigil-of-Satan was used as a base, may contain LEWD data.
- LEWD-Mental-Cultist-3.2-1B - LEWD-Mental-Occult was used as a base, may be more explicit.
- ASTAROTH-3.2-1B - ASTAROTH is the definitive merge of all previos merges.
🇪🇦 Español
Versión remezclada / remixada del modelo HarmfulProject-3.2-1B.
Aviso desde ya que no tiene nada raro mas allá de ser una mezcla de mezclas de modelos abliterados y sin censura.
La diferencia con HarmfulProject-3.2-1B es que en vez de haber sido mezclados en un modelo en este modelo se ha mezclado los mismos modelos pero en mezclas más pequeñas.
AVISO: No me hago cargo del contenido del modelo ya que solo hice la mezcla, no le inyecté dataset alguno... aún.
Otros modelos creados con esta mezclas
- Cultist-3.2-1B - Sigil-of-Satan fue usado como base, puede contener datos LEWD.
- LEWD-Mental-Cultist-3.2-1B - LEWD-Mental-Occult fue usado como base, puede que sea más explicito.
- ASTAROTH-3.2-1B - ASTAROTH es la mezcla definitiva de todas mis mezclas anteriores.
Quants / Cuantizaciones
- Static Quants: mradermacher/UNCENSORED-Sigil-Of-Satan-3.2-1B-GGUF
- Weighed/iMatrix: En proceso...
Merge Method
This model was merged using the Model Stock merge method using Novaciano/BAPHOMET as a base.
Models Merged
The following models were included in the merge:
Configuration
The following YAML configuration was used to produce this model:
models:
- model: Novaciano/LAMED
- model: Novaciano/VAV
- model: Novaciano/YOD
- model: Novaciano/TAV
- model: Novaciano/NUN-FINAL
merge_method: model_stock
base_model: Novaciano/BAPHOMET
dtype: bfloat16
parameters:
t: [0, 0.5, 1, 0.5, 0]
Open LLM Leaderboard Evaluation Results
Detailed results can be found here! Summarized results can be found here!
| Metric | Value (%) |
|---|---|
| Average | 13.69 |
| IFEval (0-Shot) | 54.94 |
| BBH (3-Shot) | 9.40 |
| MATH Lvl 5 (4-Shot) | 5.44 |
| GPQA (0-shot) | 1.45 |
| MuSR (0-shot) | 1.42 |
| MMLU-PRO (5-shot) | 9.50 |