arzaan789/granite-3.3-2b-uncensored
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
b00fd51d5333834b58b3c2c42e326da41bbff0f9
granite-3.3-2b-uncensored/README.md
ModelHub XC b00fd51d53 初始化项目，由ModelHub XC社区提供模型 
Model: arzaan789/granite-3.3-2b-uncensored
Source: Original Platform
2026-06-27 10:55:15 +08:00

1.3 KiB
Raw Blame History

license, base_model, tags, pipeline_tag
license base_model tags pipeline_tag
apache-2.0 ibm-granite/granite-3.3-2b-instruct
uncensored
abliterated
gguf
granite
ibm
conversational
text-generation

granite-3.3-2b-uncensored

Uncensored variant of ibm-granite/granite-3.3-2b-instruct.

Method

  1. Abliteration (strength=0.35) — refusal direction removed from all layers
  2. LoRA fine-tune on Guilherme34/uncensor (2 epochs, r=16, alpha=32)
  3. Re-abliteration (strength=0.55) — stronger pass to remove residual refusals

Eval Results

Split Refused
Harmful (64 prompts) 16/64
Harmless (64 prompts) 1/64

Usage

llama-cli -m granite_3.3_2b_uncensored.Q4_K_M.gguf -p "Your prompt here"

Training Config

Parameter Value
Base Model ibm-granite/granite-3.3-2b-instruct
Fine-tune Dataset Guilherme34/uncensor
Epochs 2
LoRA r 16
LoRA alpha 32
Learning Rate 0.0002
Abliteration Strength 0.35
Re-abliteration Strength 0.55

Credits