wassname
0 Followers · 0 Following
  • Joined on 2026-02-05
Repositories
255
 Projects Packages Public Activity Starred Repositories
Filter
Sort
ml_debug
Python 0 0 0
Updated 2026-08-14 15:42:47 +08:00
ml-debug
Python 0 0 0
Updated 2026-08-14 15:42:47 +08:00
isokl_steering_calibration
Python 0 0 0
Updated 2026-08-13 05:45:48 +08:00
vGROUT_pub
Python 0 0 0
Updated 2026-08-12 15:39:07 +08:00
persona-steering-template-library
Python 0 0 0
Measured persona prompt templates and contrastive persona pairs for steering experiments
Updated 2026-08-09 10:47:36 +08:00
Unsupervised-Elicitation
Python 0 0 0
Updated 2026-08-04 13:29:18 +08:00
machiavelli_deep_value
Python 0 0 0
Minimal code for reproducing the MACHIAVELLI Deep Value dataset
Updated 2026-07-30 15:38:00 +08:00
moral-maps
Python 0 0 0
tiny moral foundations vignettes. logprob eval for steering
Updated 2026-07-19 09:17:23 +08:00
soul-ab-test-scope-test
Python 0 0 0
Updated 2026-07-18 10:03:49 +08:00
openrouter-python-sdk-retry-errors
Python 0 0 0
Updated 2026-07-15 06:26:08 +08:00
steer-heal-love
Python 0 0 0
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
lora-lite
Python 0 0 0
A hackable, single-file-per-variant LoRA library built on PyTorch forward hooks.
Updated 2026-06-19 08:47:41 +08:00
evil_MoE
Python 0 0 0
Putting the E in MoE with an evil expert (can initial seeding, cause follow up unwated behaviour to absorb into a MoE)
Updated 2026-06-14 13:06:38 +08:00
grpo_proj2
Python 0 0 0
Updated 2026-06-01 14:30:20 +08:00
minicache
Python 0 0 0
Updated 2026-05-15 14:44:57 +08:00