Quantization made by Richard Erkhov.

Github

Discord

Request more models

SOVLish-Devil-8B-L3 - GGUF

• Model creator: https://huggingface.co/saishf/
• Original model: https://huggingface.co/saishf/SOVLish-Devil-8B-L3/

Name	Quant method	Size
SOVLish-Devil-8B-L3.Q2K.gguf	Q2K	2.96GB
SOVLish-Devil-8B-L3.IQ3XS.gguf	IQ3XS	3.28GB
SOVLish-Devil-8B-L3.IQ3S.gguf	IQ3S	3.43GB
SOVLish-Devil-8B-L3.Q3KS.gguf	Q3K_S	3.41GB
SOVLish-Devil-8B-L3.IQ3M.gguf	IQ3M	3.52GB
SOVLish-Devil-8B-L3.Q3K.gguf	Q3K	3.74GB
SOVLish-Devil-8B-L3.Q3KM.gguf	Q3K_M	3.74GB
SOVLish-Devil-8B-L3.Q3KL.gguf	Q3K_L	4.03GB
SOVLish-Devil-8B-L3.IQ4XS.gguf	IQ4XS	4.18GB
SOVLish-Devil-8B-L3.Q40.gguf	Q40	4.34GB
SOVLish-Devil-8B-L3.IQ4NL.gguf	IQ4NL	4.38GB
SOVLish-Devil-8B-L3.Q4KS.gguf	Q4K_S	4.37GB
SOVLish-Devil-8B-L3.Q4K.gguf	Q4K	4.58GB
SOVLish-Devil-8B-L3.Q4KM.gguf	Q4K_M	4.58GB
SOVLish-Devil-8B-L3.Q41.gguf	Q41	4.78GB
SOVLish-Devil-8B-L3.Q50.gguf	Q50	5.21GB
SOVLish-Devil-8B-L3.Q5KS.gguf	Q5K_S	5.21GB
SOVLish-Devil-8B-L3.Q5K.gguf	Q5K	5.34GB
SOVLish-Devil-8B-L3.Q5KM.gguf	Q5K_M	5.34GB
SOVLish-Devil-8B-L3.Q51.gguf	Q51	5.65GB
SOVLish-Devil-8B-L3.Q6K.gguf	Q6K	6.14GB
SOVLish-Devil-8B-L3.Q80.gguf	Q80	7.95GB

Original model description:

---

license: cc-by-nc-4.0
library_name: transformers
tags:

• mergekit
• merge

base_model:

• mlabonne/Daredevil-8B-abliterated
• ResplendentAI/RPFormatQuoteAsteriskLlama3
• mlabonne/Daredevil-8B-abliterated
• ResplendentAI/BlueMoonLlama3
• mlabonne/Daredevil-8B-abliterated
• ResplendentAI/LunaLlama3
• mlabonne/Daredevil-8B-abliterated
• mlabonne/Daredevil-8B-abliterated
• ResplendentAI/AuraLlama3
• mlabonne/Daredevil-8B-abliterated
• ResplendentAI/Smarts_Llama3

model-index:

• name: SOVLish-Devil-8B-L3

results:
- task:
type: text-generation
name: Text Generation
dataset:
name: AI2 Reasoning Challenge (25-Shot)
type: ai2_arc
config: ARC-Challenge
split: test
args:
numfewshot: 25
metrics:
- type: acc_norm
value: 69.2
name: normalized accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/openllmleaderboard?query=saishf/SOVLish-Devil-8B-L3
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: HellaSwag (10-Shot)
type: hellaswag
split: validation
args:
numfewshot: 10
metrics:
- type: acc_norm
value: 84.44
name: normalized accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/openllmleaderboard?query=saishf/SOVLish-Devil-8B-L3
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MMLU (5-Shot)
type: cais/mmlu
config: all
split: test
args:
numfewshot: 5
metrics:
- type: acc
value: 68.97
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/openllmleaderboard?query=saishf/SOVLish-Devil-8B-L3
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: TruthfulQA (0-shot)
type: truthful_qa
config: multiple_choice
split: validation
args:
numfewshot: 0
metrics:
- type: mc2
value: 57.95
source:
url: https://huggingface.co/spaces/HuggingFaceH4/openllmleaderboard?query=saishf/SOVLish-Devil-8B-L3
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: Winogrande (5-shot)
type: winogrande
config: winogrande_xl
split: validation
args:
numfewshot: 5
metrics:
- type: acc
value: 78.14
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/openllmleaderboard?query=saishf/SOVLish-Devil-8B-L3
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: GSM8k (5-shot)
type: gsm8k
config: main
split: test
args:
numfewshot: 5
metrics:
- type: acc
value: 72.48
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/openllmleaderboard?query=saishf/SOVLish-Devil-8B-L3
name: Open LLM Leaderboard

---

merge

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

Merge Details

!image/png
Devil >:3

This is another "SOVL" style merge, this time using mlabonne/Daredevil-8B-abliterated.

Daredevil is the first abliterated model i've tried that feels as smart as base llama-3-instruct while also being willing to give instructions to do all kinda of illegal things

This model should do well in rp, I'm yet to test it (waiting for gguf files @_@)

Merge Method

This model was merged using the Model Stock merge method using mlabonne/Daredevil-8B-abliterated as a base.

Models Merged

The following models were included in the merge:

• mlabonne/Daredevil-8B-abliterated + ResplendentAI/RPFormatQuoteAsteriskLlama3
• mlabonne/Daredevil-8B-abliterated + ResplendentAI/BlueMoonLlama3
• mlabonne/Daredevil-8B-abliterated + ResplendentAI/LunaLlama3
• mlabonne/Daredevil-8B-abliterated + ResplendentAI/AuraLlama3
• mlabonne/Daredevil-8B-abliterated + ResplendentAI/SmartsLlama3

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: mlabonne/Daredevil-8B-abliterated+ResplendentAI/Aura_Llama3
  - model: mlabonne/Daredevil-8B-abliterated+ResplendentAI/Smarts_Llama3
  - model: mlabonne/Daredevil-8B-abliterated+ResplendentAI/Luna_Llama3
  - model: mlabonne/Daredevil-8B-abliterated+ResplendentAI/BlueMoon_Llama3
  - model: mlabonne/Daredevil-8B-abliterated+ResplendentAI/RPFormatQuoteAsterisk_Llama3
mergemethod: modelstock
base_model: mlabonne/Daredevil-8B-abliterated
dtype: bfloat16

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	71.86
AI2 Reasoning Challenge (25-Shot)	69.20
HellaSwag (10-Shot)	84.44
MMLU (5-Shot)	68.97
TruthfulQA (0-shot)	57.95
Winogrande (5-shot)	78.14
GSM8k (5-shot)	72.48