Lesser Wrong Feed

 Old talks, 2000 - 2020 period.
 Fri, 30 Oct 2026 08:57:59 GMT
 LLM self-decompilation to turn weights into conventional code
 Fri, 30 Oct 2026 07:56:11 GMT
 My experiences with the extremes of post-viral illness and brain retraining/mindbody work
 Fri, 30 Oct 2026 06:07:39 GMT
 Walking Partway to School
 Fri, 30 Oct 2026 02:30:57 GMT
 Interpreting attention head outputs with NLAs
 Fri, 30 Oct 2026 02:09:38 GMT
 Formalising Newcomblike Problems
 Fri, 30 Oct 2026 00:13:15 GMT
 Formally Defining Wavefunction Branches
 Thu, 29 Oct 2026 23:03:53 GMT
 What is OpenAI's Mission Advisory Council?
 Thu, 29 Oct 2026 22:58:42 GMT
 Probability ranges are a cop-out with 90–93% probability
 Thu, 29 Oct 2026 21:14:06 GMT
 [Paper] Propensities Please: Exposing Alignment-Faking LLMs using Synthetic Document Finetuning
 Thu, 29 Oct 2026 22:04:03 GMT