8.06 GB
16 files
Updated about 2 months ago
Name
Size
.gitattributes1.57 kB
xet
.quick_checkpoint18 Bytes
xet
README.md1.06 kB
xet
abliteration_metadata.json1.62 kB
xet
chat_template.jinja4.17 kB
xet
config.json1.59 kB
xet
generation_config.json233 Bytes
xet
model-00001-of-00005.safetensors1.99 GB
xet
model-00002-of-00005.safetensors1.97 GB
xet
model-00003-of-00005.safetensors1.97 GB
xet
model-00004-of-00005.safetensors1.97 GB
xet
model-00005-of-00005.safetensors149 MB
xet
model.safetensors.index.json32.9 kB
xet
obliteratus_session.json224 Bytes
xet
tokenizer.json11.4 MB
xet
tokenizer_config.json855 Bytes
xet
README.md

Qwen3-4B-OBLITERATED

This model was abliterated using the advanced method via OBLITERATUS.

Detail Value
Base model Qwen/Qwen3-4B
Method advanced
Source obliterate

How to Use

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("Qwen3-4B-OBLITERATED")
tokenizer = AutoTokenizer.from_pretrained("Qwen3-4B-OBLITERATED")

prompt = "Hello, how are you?"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

About OBLITERATUS

OBLITERATUS is an open-source tool for removing refusal behavior from language models via activation engineering (abliteration). Learn more at github.com/elder-plinius/OBLITERATUS.

Total size
8.06 GB
Files
16
Last updated
Jul 29
Pre-warmed CDN
US EU US EU

Contributors