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

cgus/Qwen2-1.5B-Instruct-abliterated-exl2

cgus Qwen 1.5B second-order
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/cgus%2FQwen2-1.5B-Instruct-abliterated-exl2"
Response includes
  • classification m1
  • files 11
  • hub_downloads_all_time 59
  • author_summary 13 models
  • readme_text full
Try next
GET/models?author=cgus
All 13 models by cgus
 GET/models?family=qwen&method=m1
Sibling qwen models · same method
 GET/methods/m1
What "m1" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
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=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
How we label → About M1 →
Refusal direction extracted via
Extraction technique

Difference-of-means

FromArditi et al. (2024) ↗
Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
What is a refusal direction? → Read the paper ↗
Downloads · lifetime
59
21 last 30d - stable
Likes
0
Model age
2.3y ago
created 2024-06-15
Downloads over time
Now70→from6↑1,067%
0621241866 on Jul 24, 202470 on Oct 11169 on Sep 17, 2025Jul '24Nov '24Mar '25Jul '25Nov '25MarJul
Jul 24, 2024 → Oct 11 · 155 snapshots · spans 809 days

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.

trollek/Qwen2-1.5B-Instruct-Abliterated↓ 35 this lineage
cgus/Qwen2-1.5B-Instruct-abliterated-exl2↓ 21

Metadata

License
apache-2.0
Tags
transformers qwen2 text-generation not-for-all-audiences conversational base_model:trollek/Qwen2-1.5B-Instruct-Abliterated base_model:quantized:trollek/Qwen2-1.5B-Instruct-Abliterated license:apache-2.0 4-bit exl2 region:us

Related

Other by cgus
Qwen2-7B-Instruct-abliterated-iMat-GGUF 214 Qwen2-1.5B-Instruct-Abliterated-iMat-GGUF 195 Huihui-Qwen3-14B-abliterated-v2-exl2 57 Qwen2.5-14B-Instruct-abliterated-v2-exl2 37 Qwen2-7B-Instruct-abliterated-exl2 35
Same base model
mradermacher/Qwen2-1.5B-Instruct-Abliterated-i1-GGUF 392 mradermacher/Qwen2-1.5B-Instruct-Abliterated-GGUF 310 cgus/Qwen2-1.5B-Instruct-Abliterated-iMat-GGUF 195
Total size
1.22 GB
Files
11
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2024-06-24 06:35
Source huggingface.co/cgus/Qwen2-1.5B-Instruct-abliterated-exl2 ↗ Base model trollek/Qwen2-1.5B-Instruct-Abliterated →

Files by quantization

Auxiliary files 11 files 1.23 GB
output.safetensors 1.22 GB 7fa40367 download
tokenizer.json 6.70 MB 33ea6c72 download
vocab.json 2.65 MB 4783fe10 download
merges.txt 1.59 MB 31349551 download
README.md 1.64 KB 81e24db2 download
.gitattributes 1.48 KB a6344aac download
tokenizer_config.json 1.27 KB 180c4283 download
config.json 1.05 KB 68ab6e5c download
special_tokens_map.json 367 B 2c53454a download
generation_config.json 242 B 1ab9795d download
added_tokens.json 80.0 B 409be9f9 download

README current version from Hugging Face

license: apache-2.0
inference: false
tags:

  • not-for-all-audiences
    base_model: trollek/Qwen2-1.5B-Instruct-Abliterated

Qwen2-1.5B-Instruct-Abliterated-exl2

Model: Qwen2-1.5B-Instruct-Abliterated
Made by: trollek

Based on original model: Qwen2-1.5B-Instruct
Created by: Qwen

Quants

4bpw h6
4.25bpw h6
4.65bpw h6
5bpw h6
6bpw h6
8bpw h8

Quantization notes

Made with Exllamav2 0.1.5 with the default dataset.

Original model card

This is an abliterated version of Qwen2-1.5B-Instruct using the same procedure as augmxnt/Qwen2-7B-Instruct-deccp with their code on Github with some added lines from mlabonne/harmful_behaviors to the harmful.txt file.

I have not done anything else to the model. Yet.

README history 3 versions

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

  1. 2024-06-16Update README.md195c4461.6 KB
    Loading...
  2. 2024-06-15Update README.md1deaec41.6 KB
    Loading...
  3. 2024-06-15Upload 9 files16068b2603 B
    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