summaryrefslogtreecommitdiff
path: root/ep_run/casc_bp_train.py
diff options
context:
space:
mode:
Diffstat (limited to 'ep_run/casc_bp_train.py')
-rw-r--r--ep_run/casc_bp_train.py3
1 files changed, 2 insertions, 1 deletions
diff --git a/ep_run/casc_bp_train.py b/ep_run/casc_bp_train.py
index 987277f..d45327b 100644
--- a/ep_run/casc_bp_train.py
+++ b/ep_run/casc_bp_train.py
@@ -146,7 +146,8 @@ if args.resume:
print(f'[resume] loaded {args.resume} at step {start_step}', flush=True)
if args.opt == 'muon':
from muon import build_hybrid
- opt, sched = build_hybrid(blocks, params, args.lr, args.muon_lr, args.warmup)
+ opt, sched = build_hybrid(blocks, params, args.lr, args.muon_lr, args.warmup,
+ total_steps=(args.steps if args.cosine else 0), lr_min_ratio=args.lr_min_ratio)
else:
if args.wd >= 0: # OLMo2-style grouped decay
nodecay = {id(p) for p in tok.parameters()} | {id(p) for p in pos.parameters()} | \