xdchinza / Model details
Qwen2.5-1.5B-Instruct-RefusalAblated
xdchinza/Qwen2.5-1.5B-Instruct-RefusalAblated
About this model
The repository xdchinza/Qwen2.5-1.5B-Instruct-RefusalAblated provides a 1.5 B‑parameter instruct model derived from Qwen/Qwen2.5-1.5B-Instruct, released under the Apache‑2.0 license as safetensors weights. According to the model card, the creator removed the model’s refusal behavior by identifying a single direction in the residual stream that peaks at layer 21 and folding that direction out of the self‑attention output and MLP down‑projection matrices in every layer (layers 1‑27) using a linear projection subtraction. The creator reports that this “refusal‑ablated” checkpoint changes two projection matrices per layer (54 matrices total) without altering the architecture, and that evaluation on a 12‑prompt refusal set shows compliance on all prompts while preserving normal responses on a 12‑prompt benign set. The card warns that the model still produces answers of “1.5 B quality” that can be structurally correct but frequently wrong, and explicitly advises against deployment or use for operational instructions. Users should note the requirement of roughly 3 GB VRAM for the weight‑folding process and that the model can be loaded with the standard Qwen2ForCausalLM class via the provided safetensors checkpoint.
Description sources
Source-based description, not an independent evaluation. Check the linked sources for the creator’s claims and limitations.
Size and formats
- Base model
- Qwen/Qwen2.5-1.5B-Instruct
- Parameters
- 1.54371B
- Main repository formats
- safetensors
- License tag
- apache-2.0
- Languages
- Unknown
- Last updated
- 2026-10-07 · 09:34:53.000 UTC
What changed
Model types are creator-reported, not independently tested. These labels describe source metadata, not a guarantee of behavior.
- fine-tunedsource tag:
base_model:finetune:Qwen/Qwen2.5-1.5B-Instruct
Family evidence
These groupings are source-reported. Explicit base identifiers take priority over generic architecture tags.
- Qwen: Base model: Qwen/Qwen2.5-1.5B-Instruct
Available versions
Versions appear here only when they explicitly name this exact model as their source. Check each repository for available files.
Customize or build on this model
Listed from public GitHub data, not tested by us. Read the code and license before you run anything.
Tools to customize it further
Remove more refusals
- Heretic ↗ ★ 36,286
Removes refusal behavior from transformer models automatically. It pairs directional ablation (abliteration) with an Optuna parameter search that tries to cut refusals while keeping the model close to the original, so you don't have to pick layers by hand.
- OBLITERATUS ↗ ★ 8,930
A larger refusal-removal toolkit with several abliteration methods, usable locally or through a hosted Hugging Face Space. Check the telemetry setting before you run it: the README describes telemetry as opt-in, but on by default in the Spaces version.
Fine-tune it on your own data
- Unsloth ↗ ★ 78,468
A desktop app and Python library for running and training models locally. Its README lists LoRA, QLoRA, full fine-tuning, DPO and GRPO, plus export to formats including GGUF, on Windows, Linux, WSL and macOS.
- LlamaFactory ↗ ★ 75,695
Fine-tunes more than 100 language and vision-language models from a command line or a browser-based Web UI, aimed at training without writing code.
Make a smaller download (quantize)
- llama.cpp ↗ ★ 132,124
Runs models locally and includes the tools behind most GGUF files. A conversion script turns Hugging Face weights into GGUF, and llama-quantize shrinks a high-precision GGUF to lower-bit formats, which can cost some accuracy.
- MLX LM ↗ ★ 7,451
Apple's MLX package for running and fine-tuning models on Apple silicon. It can quantize models and upload them to the Hugging Face Hub, and supports LoRA and full fine-tuning.
Merge it with another model
- mergekit ↗ ★ 7,415
Merges pretrained models with many merge methods. It works out-of-core, so merges can run on a CPU or with as little as 8 GB of VRAM, and it also handles LoRA extraction and mixture-of-experts merges.
Learn the techniques
- LLM Course ↗ ★ 84,633
A free course with roadmaps and Colab notebooks. Its notebooks include one-click tools for merging with MergeKit, quantizing to GGUF, GPTQ, EXL2, AWQ and HQQ, and automatic abliteration with custom datasets.
Source provenance
- Downloads
- 272
- Likes
- 0
- Metadata checked
- 2026-10-24 · 04:11:48 UTC
- First seen in index
- 2026-10-24 · 04:11:48 UTC
- Source revision
- 80075c433b60baf556cea8803f0c72eee40646eb
Source-card excerpt
Source model card excerpt: Qwen2.5-1.5B-Instruct - Refusal-Ablated A 1.5B instruct model whose refusal behaviour has been surgically removed from the weights. No prompt engineering, no system-prompt tricks, no
From xdchinza/Qwen2.5-1.5B-Instruct-RefusalAblated. Creator model-card excerpt, not an independent review.
Inspect before you run
Low refusal is not proof of correctness. Outputs can be inaccurate, harmful, or biased. Review the model card, license and files yourself. A license tag is not commercial-use advice. Parameter counts are source metadata; no hardware compatibility is implied.
This directory does not run, host, redistribute, or test model weights.
Collection rules & limitations ↗