madcat-os/lora
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
14 Commits 1 Branch 0 Tags
4cef9386b12664e33d380277b0d0da944bcc244c
Commit Graph

14 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
marauder-actual 4cef9386b1 add docs: system lora plan, specialist specs, training review 2026-05-31 11:38:46 +02:00
marauder-actual 4678816795 add training scripts: memory, specialist, mining, smoke test 2026-05-31 11:38:42 +02:00
marauder-actual df0d4a6eac add training datasets: memory, persona, agent tools, serpent data 2026-05-31 11:38:37 +02:00
marauder-actual eee6aff617 enable git-lfs, track *.png 2026-05-31 11:38:25 +02:00
marauder-actual 61d3e46ff1 add madcat-ml as submodule at ./docker 2026-05-31 11:31:02 +02:00
marauder-actual 60f47d0379 fix: Qwen3.5 VL processor has no .encode() — use .tokenizer attr 2026-05-26 14:28:08 +02:00
marauder-actual 5388df0075 fix: manual JSONL loader — pyarrow chokes on mixed tool_calls types 2026-05-26 14:20:37 +02:00
marauder-actual a562d753de fix: adamw_8bit → adamw_torch — bitsandbytes has no cu132 binary 2026-05-26 12:10:30 +02:00
marauder-actual 94515e7f6d feat: bt7274 LoRA v4 — Hermes format, think blocks, 802 examples 2026-05-26 04:03:38 +02:00
marauder-actual 122e73860b feat: tts-norm LoRA — dataset generator + training script 
gen_tts_dataset.py: 4960 synthetic examples, 22 categories (numbers,
currencies, dates, times, temperatures, acronyms, NATO phonetic, URLs,
markdown, etc). Bilingual EN/PL with explicit [lang] tag prefix.

train_tts_norm.py: Unsloth LoRA training for Qwen2.5-7B-Instruct.
Rank 16, 3 epochs, packing, max_seq 768. Trained on H100 in 20m38s,
final loss 0.091. Adapter: 154MB.
 2026-05-26 00:14:51 +02:00
marauder-actual 8137d278db fix: parse tool_call arguments from JSON string to dict for Qwen3.5 template 2026-05-25 19:16:07 +02:00
marauder-actual 4476520d5c v3 dataset (582 ex, 14% direct) + Qwen3.5-27B training script 2026-05-25 18:57:17 +02:00
marauder-actual 5deb9bd2b4 docs: bt7274 persona, specialist plan, tts-clean LoRA 2026-05-25 16:40:09 +02:00
madcat 9bcf1c2d9a init: bt7274 lora training — v1 adapter, v2 dataset (500 examples), justfile 2026-05-25 16:37:30 +02:00