wassname

vGROUT_pub

Python 0 0

Updated 2026-06-27 17:49:37 +08:00

ml_debug

Markdown 0 0

Updated 2026-06-26 09:52:43 +08:00

ml-debug

Markdown 0 0

Updated 2026-06-26 09:52:43 +08:00

persona-steering-template-library

Python 0 0

Measured persona prompt templates and contrastive persona pairs for steering experiments

Updated 2026-06-25 14:08:19 +08:00

steer-heal-love

Python 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 2026-06-24 20:50:29 +08:00

lora-lite

Python 0 0

A hackable, single-file-per-variant LoRA library built on PyTorch forward hooks.

Updated 2026-06-19 08:47:41 +08:00

pi-plan

TypeScript 0 0

pi extension: plan-mode goals with evidence in one plan.md, signed off by a read-only subagent check. Small successor to pi-lgtm.

Updated 2026-06-17 18:21:45 +08:00

pi-goals

TypeScript 0 0

pi extension: plan-mode goals with evidence in one plan.md, signed off by a read-only subagent check. Small successor to pi-lgtm.

Updated 2026-06-17 18:21:45 +08:00

pi-lgtm

TypeScript 0 0

UAT-style task tree with verify commands and done criteria for pi coding agent

Updated 2026-06-15 17:29:24 +08:00

evil_MoE

Python 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

copilot-gpt4-service

0 0

Convert Github Copilot to ChatGPT

Updated 2026-06-13 13:55:35 +08:00

adapters_as_hypotheses

Markdown 0 0

Each lora type adapter can tell us something about how to look at transformer internals, and they some with causal evidence

Updated 2026-06-11 10:29:47 +08:00

wassname

0 0

Updated 2026-06-11 09:53:28 +08:00

grpo_proj2

Python 0 0

Updated 2026-06-01 14:30:20 +08:00

minicache

Python 0 0

Updated 2026-05-15 14:44:57 +08:00