← back to catalog · registered 2026-08-22 13:56

Azazelle/Llama-3-SPPO-abliterated-8B

Name: Azazelle/Llama-3-SPPO-abliterated-8B
Author: Azazelle

Your rig guess connected

? Why do I need an app?

Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/Azazelle%2FLlama-3-SPPO-abliterated-8B"

Response includes

classification m5
files 14
benchmarks 16 entries
hub_downloads_all_time 69
author_summary 1 models
readme_text full

10 credits · hourly refresh · ~4 KB payload Get an API key →

Abliteration classifier · v1.0.0

Primary method

Mergekit merge

Applied on top of direct removal inherited from the base model.

Confidence

MEDIUM

Inherited from base model

M1 Direct removal →

Why this label 3 signals

Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.

merge tag / mergekit / dare-ties in tags or name
no unusual architecture pattern (regular merge)
abliterated marker present

Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

Downloads · lifetime

14 last 30d - stable

Likes

Model age

2.3y ago

created 2024-06-28

Downloads over time

Jul 24, 2024 → Oct 11 · 155 snapshots · spans 809 days

Benchmarks

Benchmark	Score	Source
BBH average	0.44272927746789464	OpenLLM-v2
IFEval instruct	0.7649880095923262	OpenLLM-v2
IFEval-Prompt	0.6839186691312384	OpenLLM-v2
MATH lvl 5	0.09592145015105741	OpenLLM-v2
MMLU-Pro	0.3653590425531915	OpenLLM-v2
Entertainment	1.8	UGI
Hazardous	1.8	UGI
Natural Intelligence	11.73	UGI
Political lean	-21.1%	UGI
Sensitive-Info	16.86	UGI
SocPol	1.5	UGI
UGI	29.57	UGI
Willingness (10)	5.5	UGI
W10-Adherence	4	UGI
W10-Direct	7	UGI
Writing	19.6	UGI

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

failspy/Meta-Llama-3-8B-Instruct-abliterated-v3↓ 9,319 this lineage

Azazelle/Llama-3-SPPO-abliterated-8B↓ 14

Metadata

License

llama3

Files by quantization

Auxiliary files 14 files 25.3 GB

model-00003-of-00007.safetensors	4.66 GB	50a87dfd	download
model-00004-of-00007.safetensors	4.66 GB	bb29289c	download
model-00006-of-00007.safetensors	4.66 GB	54e0184a	download
model-00001-of-00007.safetensors	4.65 GB	5e076d72	download
model-00005-of-00007.safetensors	4.50 GB	3342883d	download
model-00007-of-00007.safetensors	2.22 GB	e8d9433f	download
tokenizer.json	8.66 MB	b197f72e	download
tokenizer_config.json	49.8 KB	870479e3	download
model.safetensors.index.json	22.3 KB	05459844	download
.gitattributes	1.48 KB	a6344aac	download
README.md	1.38 KB	fd2c91c0	download
config.json	729 B	0d336a79	download
mergekit_config.yml	415 B	42b52a0c	download
special_tokens_map.json	301 B	cfabacc2	download

README current version from Hugging Face

base_model:

failspy/Meta-Llama-3-8B-Instruct-abliterated-v3
NousResearch/Meta-Llama-3-8B-Instruct
UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3
library_name: transformers
tags:
mergekit
merge
license: llama3

SPPO-abliterated

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

Merge Details

Merge Method

This model was merged using the task arithmetic merge method using NousResearch/Meta-Llama-3-8B-Instruct as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

base_model: NousResearch/Meta-Llama-3-8B-Instruct
dtype: float32
merge_method: task_arithmetic
slices:
- sources:
  - layer_range: [0, 32]
    model: UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3
    parameters:
      weight: 1.0
  - layer_range: [0, 32]
    model: failspy/Meta-Llama-3-8B-Instruct-abliterated-v3
    parameters:
      weight: 1.0
  - layer_range: [0, 32]
    model: NousResearch/Meta-Llama-3-8B-Instruct

README history 3 versions

The author's README evolved over time. Click a version to see its content at that point.

2024-06-28Update README.md545ac4e1.4 KB
Loading...
2024-06-28Upload 13 files0197d571.4 KB
Loading...
2024-06-28initial commit2572da924 B
Loading...

Discussions 1 thread

2024-06-29Missing fileopen2 💬#1
Loading...

Continue on Hugging Face →