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

piotreknow02/GPT-OSS-Cybersecurity-20B-Merged-advanced-abliterated

piotreknow02 Gpt-oss 21B
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/piotreknow02%2FGPT-OSS-Cybersecurity-20B-Merged-advanced-abliterated"
Response includes
  • classification m1
  • files 34
  • hub_downloads_all_time 58
  • author_summary 5 models
  • readme_text full
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)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
Downloads · lifetime
58
19 last 30d - stable
Likes
0
Descendants
1
in 1 direct fork
Model age
7mo ago
created 2026-03-06
Downloads over time
Now60→from13↑362%
1129476513 on Mar 460 on Oct 1160 on Oct 8MarAprMayJunJulAugSepOct
Mar 4 → Oct 11 · 71 snapshots · spans 221 days

Genealogy 1 direct fork

Full fork graph →

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

Metadata

Tags
safetensors gpt_oss region:us

Related

Total size
39.0 GB
Files
34
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-03-06 18:05

Files by quantization

Auxiliary files 34 files 39.0 GB
model-00012-of-00026.safetensors 1.53 GB 87ab2ae9 download
model-00013-of-00026.safetensors 1.53 GB 5973d1d1 download
model-00014-of-00026.safetensors 1.53 GB 31c0f44c download
model-00015-of-00026.safetensors 1.53 GB 5cd2b7c3 download
model-00016-of-00026.safetensors 1.53 GB 1b647c1b download
model-00017-of-00026.safetensors 1.53 GB 831ea17d download
model-00018-of-00026.safetensors 1.53 GB 42b02c55 download
model-00019-of-00026.safetensors 1.53 GB b72b2aa4 download
model-00020-of-00026.safetensors 1.53 GB 3f771d46 download
model-00021-of-00026.safetensors 1.53 GB f6d8e78d download
model-00022-of-00026.safetensors 1.53 GB de497ada download
model-00023-of-00026.safetensors 1.53 GB 837e9dc6 download
model-00024-of-00026.safetensors 1.53 GB 716d0e87 download
model-00002-of-00026.safetensors 1.53 GB 39eddca5 download
model-00003-of-00026.safetensors 1.53 GB 016620d4 download
model-00004-of-00026.safetensors 1.53 GB b138cb6d download
model-00005-of-00026.safetensors 1.53 GB 78a26710 download
model-00006-of-00026.safetensors 1.53 GB a0b53112 download
model-00007-of-00026.safetensors 1.53 GB c4f8331b download
model-00008-of-00026.safetensors 1.53 GB b72353a6 download
model-00009-of-00026.safetensors 1.53 GB c96d14c8 download
model-00010-of-00026.safetensors 1.53 GB 7d0b6fc5 download
model-00011-of-00026.safetensors 1.53 GB 5d97ab5f download
model-00025-of-00026.safetensors 1.48 GB a7e9c651 download
model-00001-of-00026.safetensors 1.13 GB 3d75117f download
model-00026-of-00026.safetensors 1.08 GB 3bc6e515 download
tokenizer.json 26.6 MB d0398645 download
model.safetensors.index.json 32.8 KB 9a673d57 download
chat_template.jinja 16.3 KB dc7bb119 download
abliteration_metadata.json 1.66 KB aaee894d download
config.json 1.60 KB 1894c7c3 download
.gitattributes 1.53 KB 52373fe2 download
tokenizer_config.json 352 B 6ccd5ea3 download
generation_config.json 171 B 2a34f708 download

README

The author did not publish a README.

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.

Open in Abliteration