Continuum-AI-Corp/OrcaBonsai-27B-Uncensored
Runtime behavioral ablation for compressed LLMs. First target: Ternary Bonsai 2 27B — no weight modification or re-quantization. by OrcaRouter team
About the project
A tool for runtime refusal ablation in the compressed Ternary Bonsai 2 27B LLM: it projects the residual stream to remove the refusal direction without modifying or re-quantizing weights. Runs on Apple Silicon via MLX.
Useful for
- Run the model with alpha=1 to get uncensored responses on the same weights
- Compare model behavior at alpha=0 and alpha=1 for A/B testing
- Verify instrumentation correctness with scripts/selfcheck.py
README summarized by DeepSeek V4.1 Flash. Details may be inaccurate.
Why it’s trending
- 291 stars so far today, about 365 expected by the end of the day.
- The repository is 0 days old and already has 286 stars.
- Top new repositories this week: #35.
- About 77 forks a day — people are taking the code.
Stars per day
Bars are daily stars, the line is the usual pace. Red marks spike days.
Numbers
- Total stars
- 286
- Stars in a day
- 365
- Forks
- 27
- Issues and pull requests
- 0
- Watchers
- 0
- Language
- Python
- Created
- September 18, 2026
- Last push
- September 18, 2026
Star trust
Growth looks organic: forks and discussion are in line with active projects, and stars arrive unevenly, the way people give them.
These are heuristics, not a verdict: we judge by the repository’s behavior, not by a list of stargazers.
Spotted in
- September 18, 2026Top new repositories this week: #35
Share
README badge
Paste it into your README — the badge shows the current magnitude and links to this page.
[](https://gitnova.dev/en/r/Continuum-AI-Corp/OrcaBonsai-27B-Uncensored)Similar projects
-
8.6
TheoLeeCJ/SemIf
An open reproduction of the Jev-style semantic decision interface: reads typed option probabilities directly from a 4B model's logits without generating text. Runs on a single RTX 3090.
-
7.1
vinnylarouge/jevlike
A small trainable model that picks one option from a changing list of text options in a single forward pass, an open alternative to TypeSafe's Jev. Aimed at developers who need fast classification or menu selection…
-
6.6
TianyuCodings/NanoJev
A nano replica of Jev built on Qwen3-0.6B that outputs probability distributions over candidates in a single forward pass with no token decoding, plus a training pipeline and game demos (maze, Snake).
-
6.3
zhengkid/Dream-RSI
Research project on recursive self-improvement: past discovery histories become a replay simulator used to evaluate and refine exploration policies. Code is still being prepared; paper and demo are available.
-
4.9
ekzhang/openjev-sglang
HTTP server implementing the TypeSafe/Jev API for structured text classification on Qwen3.6-35B-A3B via SGLang; returns answer probabilities without generating a chain of thought.
-
4.8
kvmem/kvmem-llama.cpp
A llama.cpp port with tiered KV memory: a bounded GPU KV working set, host-RAM storage and query-based retrieval of relevant blocks. Runs Qwen3.8-27B with a 256K logical workspace on 16 GiB VRAM for long-running agent…