Seth 48b627d498 Add LoRA training scripts and fix bake-off token budget ...

- training/scripts/train_lora.py: Unsloth QLoRA trainer for qwen3:8b
- training/scripts/train_lora.sh: Launch script for steel141 RTX 3090 Ti
- eval/bakeoff.py: Fixed token budget (400->1500) that caused qwen3
  models to exhaust tokens on thinking, added --no-think flag
- agent/serve.py: Default model changed to gemma3n:e4b

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

2026-03-18 10:40:18 -04:00

agent

Add LoRA training scripts and fix bake-off token budget

2026-03-18 10:40:18 -04:00

data

Initial project scaffold: dataset schema, 31 seed training examples, Mineflayer bot framework, and 7-phase roadmap

2026-03-18 01:51:28 -04:00

eval

Add LoRA training scripts and fix bake-off token budget

2026-03-18 10:40:18 -04:00

ingame

Initial project scaffold: dataset schema, 31 seed training examples, Mineflayer bot framework, and 7-phase roadmap

2026-03-18 01:51:28 -04:00

knowledge

Add knowledge corpus: 14 command references, server context, and TF-IDF search index (Phase 1.3)

2026-03-18 02:01:12 -04:00

training

Add LoRA training scripts and fix bake-off token budget

2026-03-18 10:40:18 -04:00

.gitignore

Initial project scaffold: dataset schema, 31 seed training examples, Mineflayer bot framework, and 7-phase roadmap

2026-03-18 01:51:28 -04:00

IDEA.md

Initial project scaffold: dataset schema, 31 seed training examples, Mineflayer bot framework, and 7-phase roadmap

2026-03-18 01:51:28 -04:00

PLAN.md

Add baseline assistant with tools, guardrails, and system prompts (Phase 1.4)

2026-03-18 02:12:20 -04:00

SESSION.default.md

Initial project scaffold: dataset schema, 31 seed training examples, Mineflayer bot framework, and 7-phase roadmap

2026-03-18 01:51:28 -04:00

S

Description

An open-source AI God for Minecraft servers — trained to understand natural language, execute commands, and play a divine character.

14 MiB

Languages

Python 83.5%

JavaScript 10.5%

HTML 4.4%

Shell 1.6%