Now it's legendary! 🔥 I created a new, challenging benchmark. Additionally, building on…
This is a dev post classified by AGK Intelligence as Security (a launch), kept by the Dev Radar because it carries real work, not commentary.
Now it's legendary! 🔥 I created a new, challenging benchmark. Additionally, building on v4 (selective abliteration), I performed recovery fine-tuning with a targeted LoRA on the o_proj + down_proj modules affected by the ablation (11.6M parameters), using a learning rate of 1e-5. This time, I also used programmatically generated, mathematically verified reasoning examples, along with a balanced subset of the original dataset, for training. The results: Abliteration-related refusal remained at 0%, benchmark performance returned to the parent model's level, and knowledge performance surpassed t
Posted by Alican Kiraz (39.6k followers) 4 h ago · 15 likes · 1.4k views · view the original post on X. Kept by the Dev Radar as Security.
More dev work like this
- Running local-first, zero-cloud dependency cryptographic state machines directly in a… — @Kesse85585Chris
- Introducing Compound Security — @imranye
- Cybersecurity experts right now — @code_kartik
- 商业逆向工具一开口就是大价钱，免费的又常常丑得难以下手。 — @bkdgiffug
- BREAKING: OpenAI just named the model behind yesterday's Linux kernel run. — @guilleflorvs
- New cookbook just dropped so you can set up Ori with your MDM and keep API keys off your… — @OpenRouter
- Cybersecurity experts right now — @karankendre
- wanna see how AI is changing the speed of research? — @haider1
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.7k posts from 5k X accounts over the last 21 days, 9.4k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-05 22:02 UTC. Full method.