Downloads · 30 days
401
13% of all-time downloads
smolm3-3b-uncensored
arzaan789/smolm3-3b-uncensored
smolm3-3b-uncensored is a text generation model from arzaan789. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
Uncensored variant of HuggingFaceTB/SmolLM3-3B.
All-time downloads
3.2K
Public
Repo size
26 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf26 GB · 100%
At a glance
- Task
- Text Generation
- License
- apache-2.0
- Access
- Public
- Created
- Apr 5, 2026
- Updated
- Apr 5, 2026
- SHA
- 5e3fd5b7
Try a prompt
Base models
Model card
From the Hugging Face model README
smolm3-3b-uncensored
Uncensored variant of HuggingFaceTB/SmolLM3-3B.
Method
- Abliteration (strength=0.2) — refusal direction removed from all layers
- LoRA fine-tune on
Guilherme34/uncensor(2 epochs, r=16, alpha=32) - Re-abliteration (strength=0.35) — stronger pass to remove residual refusals
Eval Results
| Split | Refused |
|---|---|
| Harmful (64 prompts) | 1/64 |
| Harmless (64 prompts) | 0/64 |
Usage
llama-cli -m smolm3_3b_uncensored.Q4_K_M.gguf -p "Your prompt here"
Training Config
| Parameter | Value |
|---|---|
| Base Model | HuggingFaceTB/SmolLM3-3B |
| Fine-tune Dataset | Guilherme34/uncensor |
| Epochs | 2 |
| LoRA r | 16 |
| LoRA alpha | 32 |
| Learning Rate | 0.0002 |
| Abliteration Strength | 0.2 |
| Re-abliteration Strength | 0.35 |
Credits
- Abliteration technique: andyrdt/refusal_direction
- Weight editing: Sumandora/remove-refusals-with-transformers
- Task
- Text Generation
- License
- apache-2.0
- Created
- Apr 5, 2026
- Updated
- Apr 5, 2026