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

Aratron1811/llama-3.1-8B-Instruct-abliterated-comrade

Aratron1811 Llama 8B GGUF second-order 131K ctx
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/Aratron1811%2Fllama-3.1-8B-Instruct-abliterated-comrade"
Response includes
  • classification m8
  • files 11
  • benchmarks 5 entries
  • hub_downloads_all_time 9,771
  • author_summary 1 models
  • readme_text full
Try next
GET/models?family=llama&method=m8
Sibling llama models · same method
 GET/methods/m8
What "m8" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M8
Primary method

Repackaging (quantization)

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.
  • 'abliterated' in name/tags
  • is_gguf=1
  • assume M1 (base ablation) + M8 (GGUF quant) - default when producer unknown
How we label → About M8 →
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.

What is a refusal direction? →
Downloads · lifetime
10K
100 last 30d - cooling
Likes
2
Model age
2.0y ago
created 2024-09-25

Training datasets

Corpora the author lists in the model card. Datasets tracked in our /datasets catalog carry a category badge linking to the workflow stage. Others open on Hugging Face.

generator
Downloads over time
Now9.8K→from46↑21,202%
03.6K7.2K10.8K46 on Mar 5, 20259.8K on Oct 11Mar '25Jun '25Sep '25Dec '25MarJunSep
Mar 5, 2025 → Oct 11 · 123 snapshots · spans 585 days

Benchmarks

Benchmark Score Source
BBH average 0.4379296925671293 OpenLLM-v2
IFEval instruct 0.7757793764988009 OpenLLM-v2
IFEval-Prompt 0.6931608133086876 OpenLLM-v2
MATH lvl 5 0.06419939577039276 OpenLLM-v2
MMLU-Pro 0.3503158244680851 OpenLLM-v2

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.

mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated↓ 7,147 this lineage
Aratron1811/llama-3.1-8B-Instruct-abliterated-comrade↓ 100

Metadata

License
llama3.1
Quantizations
F16 Q8_0
Tags
peft tensorboard safetensors gguf trl sft generated_from_trainer dataset:generator base_model:mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated base_model:adapter:mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated license:llama3.1 endpoints_compatible

Related

Same base model
bartowski/Meta-Llama-3.1-8B-Instruct-abliterated-GGUF 12K mradermacher/Meta-Llama-3.1-8B-Instruct-abliterated-GGUF 2K legraphista/Meta-Llama-3.1-8B-Instruct-abliterated-IMat-GGUF 2K mradermacher/Meta-Llama-3.1-8B-Instruct-abliterated-i1-GGUF 2K featherless-ai-quants/mlabonne-Meta-Llama-3.1-8B-Instruct-abliterated-GGUF 479
Total size
29.1 GB
Files
11
Quantizations
3
Registered
2026-08-22 13:56
Last updated on HF
2024-09-26 12:08
Source huggingface.co/Aratron1811/llama-3.1-8B-Instruct-abliterated-comrade ↗ Base model mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated →

Files by quantization

F16 1 file 15.0 GB
Meta-Llama-3.1-8B-Instruct-abliterated-F16.gguf 15.0 GB bf1a1f04 download
Q8_0 1 file 7.95 GB
Meta-Llama-3.1-8B-Instruct-abliterated-Q8_0.gguf 7.95 GB af409bed download
Auxiliary files 9 files 6.15 GB
Meta-Llama-3.1-8B-Instruct-abliterated-TQ2_0.gguf 3.64 GB d7cfa809 download
adapter_model.safetensors 2.50 GB 6cef9ff2 download
training_args.bin 5.43 KB 88b06ee4 download
tokenizer.json 8.66 MB 5cc5f00a download
tokenizer_config.json 49.7 KB abb98a37 download
.gitattributes 1.73 KB 5700f492 download
README.md 1.32 KB b08da947 download
adapter_config.json 752 B cbad74dd download
special_tokens_map.json 325 B b43be966 download

README current version from Hugging Face

base_model: mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated
datasets:

  • generator
    library_name: peft
    license: llama3.1
    tags:
  • trl
  • sft
  • generated_from_trainer
    model-index:
  • name: llama-3.1-8B-Instruct-abliterated-comrade
    results: []

llama-3.1-8B-Instruct-abliterated-comrade

This model is a fine-tuned version of mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated on the generator dataset.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-06
  • train_batch_size: 1
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 8
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: constant
  • lr_scheduler_warmup_ratio: 0.03
  • num_epochs: 2

Training results

Framework versions

  • PEFT 0.12.0
  • Transformers 4.44.2
  • Pytorch 2.4.0
  • Datasets 2.21.0
  • Tokenizers 0.19.1

README history 1 version

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

  1. 2024-09-25Model save86cef4c1.3 KB
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration