From bdf0c0d5201795fcbc0ec20487a03561b407cdcc Mon Sep 17 00:00:00 2001 From: dumko2001 Date: Sat, 7 Mar 2026 14:46:43 +0530 Subject: [PATCH] Allow agent to diagnose crashes by reading the python stack trace --- program.md | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/program.md b/program.md index cec54a5..c591578 100644 --- a/program.md +++ b/program.md @@ -98,9 +98,10 @@ LOOP FOREVER (until I wake up and come back in the morning): 3. git commit 4. run the experiment: `uv run train.py > run.log 2>&1` (redirect everything — do NOT use tee or let output flood your context) 5. read out the results: `grep "^val_bpb:\|^peak_vram_mb:" run.log` -6. record the results in the tsv -7. if val_bpb improved (lower), you "advance" the branch, keeping the git commit -8. if val_bpb is equal or worse, you git reset back to where you started +6. If the grep output is empty, the run crashed. You MUST run `tail -n 50 run.log` to read the Python stack trace so you can actually diagnose and fix the bug. +7. record the results in the tsv +8. if val_bpb improved (lower), you "advance" the branch, keeping the git commit +9. if val_bpb is equal or worse, you git reset back to where you started The idea is that you are a completely autonomous researcher trying things out. If they work, keep. If they don't, discard. And you're advancing the branch so that you can iterate. If you feel like you're getting stuck in some way, you can rewind but you should probably do this very very sparingly (if ever).