Sequence to Sequence Learning with Keras
Updated 2026-07-27 11:03:00 +08:00
Implementing jQueryUI Drag and Drop functionality in AngularJS (with Animation) is easier than ever
Updated 2026-07-27 11:02:57 +08:00
openshift-scalable-celery-cartridge
Updated 2026-07-27 11:02:56 +08:00
The brass buckles on lodash's utility belt
Updated 2026-07-27 11:02:50 +08:00
investigate Quiet-STaR paper, and it's thought scratchpad
Updated 2026-07-27 11:02:48 +08:00
Quick comparison of neural processes vs kriging for gridding soil samples
Updated 2026-07-27 11:02:27 +08:00
Tools for debugging, analyzing, and validating 3D Tiles tilesets 🚦
Updated 2026-07-27 11:02:14 +08:00
inspired by circuit breakers paper. honesty>harmless
Updated 2026-07-27 11:02:11 +08:00
Curse Modpack downloader
Updated 2026-07-27 11:02:08 +08:00
Facebook Mobile Timeline Cleaner. This is
Updated 2026-07-27 00:26:47 +08:00
Updated 2026-07-27 00:24:12 +08:00
Each lora type adapter can tell us something about how to look at transformer internals, and they some with causal evidence
Updated 2026-07-19 11:39:41 +08:00
tiny moral foundations vignettes. logprob eval for steering
Updated 2026-07-19 09:17:23 +08:00
Updated 2026-07-18 10:03:49 +08:00
Updated 2026-07-14 13:55:25 +08:00
ANZ 2040: the decisions Australia and New Zealand still get to make (draft proposal, retelling AI-2040 / Plan A from a middle-power view)
Updated 2026-07-12 09:20:28 +08:00
Hypothesis: you can distill a steering vector into LoRA weights and "heal" the incoherency the vector injects by regularising the training (KL to base, or weight decay). Then loop and see what multiple rounds give you.
Updated 2026-06-30 15:37:34 +08:00
Curated list of datasets for evaluating moral/ethical behaviour of LLMs: dilemmas, norms, moral foundations, ethics, honesty, personas
Updated 2026-06-30 08:22:29 +08:00
A hackable, single-file-per-variant LoRA library built on PyTorch forward hooks.
Updated 2026-06-19 08:47:41 +08:00