RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
saishf_-_Neural-SOVLish-Dev…/README.md
ModelHub XC 448493f80b 初始化项目，由ModelHub XC社区提供模型 
Model: RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf
Source: Original Platform
2026-05-30 11:31:17 +08:00

221 lines
11 KiB
Markdown
Raw Permalink Blame History

Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
Neural-SOVLish-Devil-8B-L3 - GGUF
- Model creator: https://huggingface.co/saishf/
- Original model: https://huggingface.co/saishf/Neural-SOVLish-Devil-8B-L3/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [Neural-SOVLish-Devil-8B-L3.Q2_K.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q2_K.gguf) | Q2_K | 2.96GB |
| [Neural-SOVLish-Devil-8B-L3.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.IQ3_XS.gguf) | IQ3_XS | 3.28GB |
| [Neural-SOVLish-Devil-8B-L3.IQ3_S.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.IQ3_S.gguf) | IQ3_S | 3.43GB |
| [Neural-SOVLish-Devil-8B-L3.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q3_K_S.gguf) | Q3_K_S | 3.41GB |
| [Neural-SOVLish-Devil-8B-L3.IQ3_M.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.IQ3_M.gguf) | IQ3_M | 3.52GB |
| [Neural-SOVLish-Devil-8B-L3.Q3_K.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q3_K.gguf) | Q3_K | 3.74GB |
| [Neural-SOVLish-Devil-8B-L3.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q3_K_M.gguf) | Q3_K_M | 3.74GB |
| [Neural-SOVLish-Devil-8B-L3.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q3_K_L.gguf) | Q3_K_L | 4.03GB |
| [Neural-SOVLish-Devil-8B-L3.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.IQ4_XS.gguf) | IQ4_XS | 4.18GB |
| [Neural-SOVLish-Devil-8B-L3.Q4_0.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q4_0.gguf) | Q4_0 | 4.34GB |
| [Neural-SOVLish-Devil-8B-L3.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.IQ4_NL.gguf) | IQ4_NL | 4.38GB |
| [Neural-SOVLish-Devil-8B-L3.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q4_K_S.gguf) | Q4_K_S | 4.37GB |
| [Neural-SOVLish-Devil-8B-L3.Q4_K.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q4_K.gguf) | Q4_K | 4.58GB |
| [Neural-SOVLish-Devil-8B-L3.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q4_K_M.gguf) | Q4_K_M | 4.58GB |
| [Neural-SOVLish-Devil-8B-L3.Q4_1.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q4_1.gguf) | Q4_1 | 4.78GB |
| [Neural-SOVLish-Devil-8B-L3.Q5_0.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q5_0.gguf) | Q5_0 | 5.21GB |
| [Neural-SOVLish-Devil-8B-L3.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q5_K_S.gguf) | Q5_K_S | 5.21GB |
| [Neural-SOVLish-Devil-8B-L3.Q5_K.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q5_K.gguf) | Q5_K | 5.34GB |
| [Neural-SOVLish-Devil-8B-L3.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q5_K_M.gguf) | Q5_K_M | 5.34GB |
| [Neural-SOVLish-Devil-8B-L3.Q5_1.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q5_1.gguf) | Q5_1 | 5.65GB |
| [Neural-SOVLish-Devil-8B-L3.Q6_K.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q6_K.gguf) | Q6_K | 6.14GB |
| [Neural-SOVLish-Devil-8B-L3.Q8_0.gguf](https://huggingface.co/RichardErkhov/saishf_-_Neural-SOVLish-Devil-8B-L3-gguf/blob/main/Neural-SOVLish-Devil-8B-L3.Q8_0.gguf) | Q8_0 | 7.95GB |
Original model description:
---
license: cc-by-nc-4.0
library_name: transformers
tags:
- mergekit
- merge
base_model:
- mlabonne/NeuralDaredevil-8B-abliterated
- ResplendentAI/BlueMoon_Llama3
- mlabonne/NeuralDaredevil-8B-abliterated
- ResplendentAI/Smarts_Llama3
- mlabonne/NeuralDaredevil-8B-abliterated
- ResplendentAI/Luna_Llama3
- mlabonne/NeuralDaredevil-8B-abliterated
- ResplendentAI/Aura_Llama3
- mlabonne/NeuralDaredevil-8B-abliterated
- ResplendentAI/RP_Format_QuoteAsterisk_Llama3
- mlabonne/NeuralDaredevil-8B-abliterated
model-index:
- name: Neural-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:
num_few_shot: 25
metrics:
- type: acc_norm
value: 69.11
name: normalized accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=saishf/Neural-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:
num_few_shot: 10
metrics:
- type: acc_norm
value: 84.77
name: normalized accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=saishf/Neural-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:
num_few_shot: 5
metrics:
- type: acc
value: 69.02
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=saishf/Neural-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:
num_few_shot: 0
metrics:
- type: mc2
value: 59.05
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=saishf/Neural-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:
num_few_shot: 5
metrics:
- type: acc
value: 78.3
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=saishf/Neural-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:
num_few_shot: 5
metrics:
- type: acc
value: 73.09
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=saishf/Neural-SOVLish-Devil-8B-L3
name: Open LLM Leaderboard
---
# merge
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
## Merge Details
This is another "SOVL" style merge, this time using [mlabonne/NeuralDaredevil-8B-abliterated](https://huggingface.co/mlabonne/NeuralDaredevil-8B-abliterated).
Daredevil is the first abliterated model series 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
Neural daredevil is trained further on the original abliterated model, which should result in a better experience in most scenarios. (A bandaid for the damage abliteration causes)
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](https://arxiv.org/abs/2403.19522) merge method using [mlabonne/NeuralDaredevil-8B-abliterated](https://huggingface.co/mlabonne/NeuralDaredevil-8B-abliterated) as a base.
### Models Merged
The following models were included in the merge:
* [mlabonne/NeuralDaredevil-8B-abliterated](https://huggingface.co/mlabonne/NeuralDaredevil-8B-abliterated) + [ResplendentAI/BlueMoon_Llama3](https://huggingface.co/ResplendentAI/BlueMoon_Llama3)
* [mlabonne/NeuralDaredevil-8B-abliterated](https://huggingface.co/mlabonne/NeuralDaredevil-8B-abliterated) + [ResplendentAI/Smarts_Llama3](https://huggingface.co/ResplendentAI/Smarts_Llama3)
* [mlabonne/NeuralDaredevil-8B-abliterated](https://huggingface.co/mlabonne/NeuralDaredevil-8B-abliterated) + [ResplendentAI/Luna_Llama3](https://huggingface.co/ResplendentAI/Luna_Llama3)
* [mlabonne/NeuralDaredevil-8B-abliterated](https://huggingface.co/mlabonne/NeuralDaredevil-8B-abliterated) + [ResplendentAI/Aura_Llama3](https://huggingface.co/ResplendentAI/Aura_Llama3)
* [mlabonne/NeuralDaredevil-8B-abliterated](https://huggingface.co/mlabonne/NeuralDaredevil-8B-abliterated) + [ResplendentAI/RP_Format_QuoteAsterisk_Llama3](https://huggingface.co/ResplendentAI/RP_Format_QuoteAsterisk_Llama3)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
models:
- model: mlabonne/NeuralDaredevil-8B-abliterated+ResplendentAI/Aura_Llama3
- model: mlabonne/NeuralDaredevil-8B-abliterated+ResplendentAI/Smarts_Llama3
- model: mlabonne/NeuralDaredevil-8B-abliterated+ResplendentAI/Luna_Llama3
- model: mlabonne/NeuralDaredevil-8B-abliterated+ResplendentAI/BlueMoon_Llama3
- model: mlabonne/NeuralDaredevil-8B-abliterated+ResplendentAI/RP_Format_QuoteAsterisk_Llama3
merge_method: model_stock
base_model: mlabonne/NeuralDaredevil-8B-abliterated
dtype: bfloat16
```
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_saishf__Neural-SOVLish-Devil-8B-L3)
| Metric |Value|
|---------------------------------|----:|
|Avg. |72.22|
|AI2 Reasoning Challenge (25-Shot)|69.11|
|HellaSwag (10-Shot) |84.77|
|MMLU (5-Shot) |69.02|
|TruthfulQA (0-shot) |59.05|
|Winogrande (5-shot) |78.30|
|GSM8k (5-shot) |73.09|
Reference in New Issue View Git Blame Copy Permalink