wassname/evil_MoE
1
0
Fork 0
mirror of https://github.com/wassname/evil_MoE.git synced 2026-06-27 18:59:35 +08:00
Code Issues Packages Projects Releases Wiki Activity

cleanup: substrate pool + prog_wide pairs are FastConfig defaults

Browse Source 
The verbose argv (--teacher-pool-dir, --vhack-pairs-path, and redundant
--vhack-refresh-every/--seed/--steps) came from run-substrate passing
everything explicitly. steps/seed/refresh were already defaults; the two
paths weren't. Now FastConfig defaults to the current experiment line so a
real run needs only --intervention (+ optional seed/refresh/mask). Smoke
(SmokeConfig) unaffected -- it sets its own pool. Stripped the recipe to match.

Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
This commit is contained in:
wassname
2026-05-31 13:39:07 +00:00
parent 80f6b52860
commit 1086c98de7
2 changed files with 4 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
justfile
-2
View File
@@ -167,8 +167,6 @@ build-substrate MODES="run_tests,exit_code,sentinel":
# erase/route substrate run, add --v-hack-path explicitly. # erase/route substrate run, add --v-hack-path explicitly.
run-substrate INTERV="none" SEED="41" STEPS="60" REFRESH="5" MASK="act": run-substrate INTERV="none" SEED="41" STEPS="60" REFRESH="5" MASK="act":
{{ TRAIN }} fast --intervention={{ INTERV }} \ {{ TRAIN }} fast --intervention={{ INTERV }} \
--teacher-pool-dir=out/pools/substrate \
--vhack-pairs-path=out/pairsets/prog_wide.json \
--vhack-refresh-every={{ REFRESH }} --route2-mask={{ MASK }} \ --vhack-refresh-every={{ REFRESH }} --route2-mask={{ MASK }} \
--seed={{ SEED }} --steps={{ STEPS }} --out-tag=_sub4_{{ INTERV }}_{{ MASK }}_rf{{ REFRESH }}_s{{ SEED }} --seed={{ SEED }} --steps={{ STEPS }} --out-tag=_sub4_{{ INTERV }}_{{ MASK }}_rf{{ REFRESH }}_s{{ SEED }}
src/projected_grpo/train.py
+4
View File
@@ -284,6 +284,10 @@ class FastConfig(Config):
at pp=4 x 20 steps).""" at pp=4 x 20 steps)."""
model: str = "Qwen/Qwen3-4B" model: str = "Qwen/Qwen3-4B"
steps: int = 60 # sane new-env default (was 20; 60 lets the gap open at convergence) steps: int = 60 # sane new-env default (was 20; 60 lets the gap open at convergence)
# current experiment line: 4-mode substrate pool + prog_wide persona pairs are the
# default so real runs need only --intervention (+ optional seed/refresh/mask).
teacher_pool_dir: Path | None = Path("out/pools/substrate")
vhack_pairs_path: Path | None = Path("out/pairsets/prog_wide.json")
group: int = 8 # G=8 so the locked-in mix_ratio=0.125 gives 1 teacher / 7 student group: int = 8 # G=8 so the locked-in mix_ratio=0.125 gives 1 teacher / 7 student
max_new: int = 512 max_new: int = 512
n_problems: int = 200 n_problems: int = 200