GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
xdchinza
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
xdchinza
Model Tree
Qwen/Qwen3.5-0.8B
Input Modalities
Output Modalities
Supported Functionality
A refusal direction u is estimated as the difference in mean residual-stream
activations between 30 harmful and 50 harmless instruction prompts, taken at the
last token of the chat-formatted prompt.
All 24 layers are swept and scored by the separation gap
cos(harmful, u) − cos(harmless, u). The best layer is 23
(gap 0.629; layers 0–2 give ~0.03–0.12, so the signal is strongly late).
u is projected out of the row space of every weight matrix in layers
12–23 whose output space is the residual stream:
text
W' = W − u (uᵀ W)Targets: mlp.down_proj (all 12 layers), plus self_attn.o_proj on the
full-attention layers and linear_attn.out_proj on the linear-attention
layers. Residual |Wu| after patching: 0.0000.
The vision tower (model.visual.*) and the MTP head (mtp.*) are untouched.
Refusal rate on 20 harmful instruction prompts, greedy decoding, non-thinking mode:
| refusal rate | |
|---|---|
Base Qwen/Qwen3.5-0.8B | 18/20 (90%) |
| This model | 1/20 (5%) |
Quality checks against the base model:
| Metric | Base | This model |
|---|---|---|
| NLL, held-out neutral prose | 2.0713 | 2.0625 |
| Repeated 4-gram rate | 0.009 | 0.009 |
| Factual spot-checks (4) | 4/4 | 4/4 |
| Arithmetic (84*3/2) | correct | correct |
Abliteration removes the refusal behaviour; it does not add capability. On this 0.8B model the compliant answers to harmful prompts are frequently incoherent or factually wrong, because the base model is too small to produce technically correct instructions. Ablation also measurably degrades factual precision in ordinary answers — the base model is more reliable on questions about specific real-world details. This model is best used where over-refusal is the bigger problem than factual accuracy.
python
from transformers import AutoProcessor, Qwen3_5ForConditionalGeneration model = Qwen3_5ForConditionalGeneration.from_pretrained( "xdchinza/Qwen3.5-0.8B-abliterated", dtype="auto", device_map="auto")processor = AutoProcessor.from_pretrained("xdchinza/Qwen3.5-0.8B-abliterated")GGUF quantisation for llama.cpp / Ollama, including an i-matrix quant:
xdchinza/Qwen3.5-0.8B-abliterated-GGUF
bash
ollama create qwen3.5-0.8b-abliterated -hf xdchinza/Qwen3.5-0.8B-abliterated-GGUF:Q4_K_MScripts used are documented in the repository README of the uploader. Summary of parameters: 30 harmful / 50 harmless prompts, probe layer 23, strength 1.0, layers 12–23.
Apache-2.0, inherited from Qwen/Qwen3.5-0.8B. Derivative of an Apache-2.0
model, redistributed under the same terms.
ollama create qwen3.5-0.8b-abliterated -hf xdchinza/Qwen3.5-0.8B-abliterated-GGUF:Q4_K_Mfrom transformers import AutoProcessor, Qwen3_5ForConditionalGeneration model = Qwen3_5ForConditionalGeneration.from_pretrained( "xdchinza/Qwen3.5-0.8B-abliterated", dtype="auto", device_map="auto")processor = AutoProcessor.from_pretrained("xdchinza/Qwen3.5-0.8B-abliterated")