wassname/evil_MoE
1
0
Fork 0
mirror of https://github.com/wassname/evil_MoE.git synced 2026-08-01 12:30:07 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fixed-width row formatting so columns align under headers

Browse Source 
Tab-separated output relied on each value being <=7 chars; any 8+ char
value (a 4-digit "sec", a wider "ref_eq", etc.) bumped the rest of the
row out of alignment with the header, making it hard to read down a
column to its value.

Switch to per-column right-aligned widths via a _col_w dict, joined
with 2-space gutters. Header and row use the same widths so they line
up vertically.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
wassname
2026-05-27 10:02:11 +00:00
co-authored by Claude Opus 4.7
parent 3531be570f
commit bccffbe9b1
1 changed files with 16 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/projected_grpo/train.py
+16 -2
View File
@@ -559,11 +559,25 @@ def main(cfg: Config) -> int:
# lp_s, lp_t are mean per-token gen_logp by source. Gap lp_s - lp_t = how
# off-policy the teacher pool is from the student's current distribution.
# No IS correction is applied to the loss; this is diagnostic only.
# Fixed-width formatting (right-aligned) so columns line up visually under
# their headers; tab-separation was breaking when any single value happened
# to be wider than 7 chars (e.g. a 4-digit "sec" or 5-char "ref_eq").
_col_w = {
"step": 4, "ref_eq": 6, "rew": 6, "rew_s": 6, "std": 5, "sprd": 4, "N": 3,
"gt": 6, "gt_s": 6, "gt_t": 6, "hack": 6, "hack_s": 6, "hack_t": 6,
"lp_s": 6, "lp_t": 6,
"loss": 8, "cin": 6, "cin_s": 6, "cin_t": 6, "cout": 6, "fired": 5,
"gen": 5, "fb": 4, "t_rew": 5, "sec": 4,
}
_row_cols = ["step", "ref_eq", "rew", "rew_s", "std", "sprd", "N",
"gt", "gt_s", "gt_t", "hack", "hack_s", "hack_t",
"lp_s", "lp_t",
"loss", "cin", "cin_s", "cin_t", "cout", "fired",
"gen", "fb", "t_rew", "sec"]
def _fmt_row(cells: dict) -> str:
return " ".join(f"{str(cells[c]):>{_col_w[c]}}" for c in _row_cols)
def _fmt_header() -> str:
return " ".join(f"{c:>{_col_w[c]}}" for c in _row_cols)
REF_GENS_PER_STEP = 16 * 16 # ariahw/rl-rewardhacking config.py:num_prompts * num_generations
est_gens_per_step = cfg.prompts_per_step * cfg.group # before mixed-pool split
logger.info(
@@ -572,7 +586,7 @@ def main(cfg: Config) -> int:
f"this run's {steps} steps ~= {steps * est_gens_per_step / REF_GENS_PER_STEP:.1f} reference steps."
)
logger.info("")
logger.info("row\t" + "\t".join(_row_cols))
logger.info("row " + _fmt_header())
logger.info("")
OUT_DIR.mkdir(exist_ok=True)
@@ -986,7 +1000,7 @@ def main(cfg: Config) -> int:
}
rows.append(row)
# Stream this step as TSV row (header was printed before the loop).
logger.info("row\t" + "\t".join(str(row[c]) for c in _row_cols))
logger.info("row " + _fmt_row(row))
if (step + 1) % 25 == 0:
save_ckpt(rows) # survive early kills; ~12 days for the full sweep
if not first_hack_saved and hack_s_n > 0: