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

IlyaGusev/saiga_llama3_70b_sft_m1_d5_abliterated

IlyaGusev Llama 71B
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/IlyaGusev%2Fsaiga_llama3_70b_sft_m1_d5_abliterated"
Response includes
  • classification m1
  • files 37
  • hub_downloads_all_time 229
  • author_summary 16 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
229
13 last 30d - cooling
Likes
1
Model age
2.3y ago
created 2024-06-13
Downloads over time
Now233→from19↑1,126%
08517125619 on Jul 24, 2024233 on Oct 11233 on Oct 10Jul '24Nov '24Mar '25Jul '25Nov '25MarJul
Jul 24, 2024 → Oct 11 · 155 snapshots · spans 809 days

Metadata

Tags
transformers safetensors llama text-generation conversational text-generation-inference endpoints_compatible region:us

Related

Total size
131 GB
Files
37
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2024-06-13 16:43

Files by quantization

Auxiliary files 37 files 131 GB
model-00008-of-00030.safetensors 4.66 GB 9cc5058c download
model-00013-of-00030.safetensors 4.66 GB f02f23f7 download
model-00018-of-00030.safetensors 4.66 GB df2a41a0 download
model-00023-of-00030.safetensors 4.66 GB 3f73f35e download
model-00028-of-00030.safetensors 4.66 GB dc7b0d4c download
model-00003-of-00030.safetensors 4.66 GB 763eb8d7 download
model-00029-of-00030.safetensors 4.63 GB f8d5f8d7 download
model-00009-of-00030.safetensors 4.63 GB 0ae2db61 download
model-00014-of-00030.safetensors 4.63 GB f2af6596 download
model-00019-of-00030.safetensors 4.63 GB e921c884 download
model-00024-of-00030.safetensors 4.63 GB ce4234ba download
model-00004-of-00030.safetensors 4.63 GB eb4e7064 download
model-00006-of-00030.safetensors 4.34 GB 1d04a04a download
model-00007-of-00030.safetensors 4.34 GB d05551ac download
model-00011-of-00030.safetensors 4.34 GB 31eaca11 download
model-00012-of-00030.safetensors 4.34 GB f2f0eebb download
model-00016-of-00030.safetensors 4.34 GB 02520039 download
model-00017-of-00030.safetensors 4.34 GB ea209fd4 download
model-00021-of-00030.safetensors 4.34 GB 1daf657b download
model-00022-of-00030.safetensors 4.34 GB bb6522d5 download
model-00026-of-00030.safetensors 4.34 GB 76cfa09f download
model-00027-of-00030.safetensors 4.34 GB a776528a download
model-00002-of-00030.safetensors 4.34 GB f7665070 download
model-00005-of-00030.safetensors 4.34 GB 59018936 download
model-00010-of-00030.safetensors 4.34 GB d0714796 download
model-00015-of-00030.safetensors 4.34 GB aeb2228c download
model-00020-of-00030.safetensors 4.34 GB 68a5c184 download
model-00025-of-00030.safetensors 4.34 GB 1c386678 download
model-00001-of-00030.safetensors 4.27 GB 6489d63d download
model-00030-of-00030.safetensors 1.96 GB 8185e676 download
tokenizer.json 8.66 MB a3c73537 download
model.safetensors.index.json 58.2 KB 37b1afe6 download
tokenizer_config.json 49.8 KB bdcb8c50 download
.gitattributes 3.55 KB 47554a13 download
config.json 730 B 56cf5932 download
special_tokens_map.json 446 B 4dd5e992 download
generation_config.json 277 B 8726aaf0 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