Skip to main content
mluo
Projects
deepswe
Log in
Sign up
Overview
Workspace
Runs
Automat.
Sweeps
Reports
Artifacts
Mluo's workspace
Personal workspace
Automated workspace
Changes are only visible to you.
Runs
4
Name
2 visualized
swe-sft-rl-fail
swe-sft-rl-fail
swe-14b-no-overlong-filter-fail
swe-14b-no-overlong-filter-fail
deepswe-preview-part2
deepswe-preview-part2
deepswe-preview-part1
deepswe-preview-part1
1-4
of 4
Settings
Add panels
actor
7
1-6 of 7
actor/ppo_kl
actor/ppo_kl
50
100
150
Step
-0.00015
-0.0001
-0.00005
0
0.00005
0.0001
0.00015
deepswe-preview-part2
deepswe-preview-part1
actor/pg_loss
actor/pg_loss
50
100
150
Step
-0.04
-0.02
0
0.02
0.04
deepswe-preview-part2
deepswe-preview-part1
actor/pg_clipfrac_lower
actor/pg_clipfrac_lower
50
100
150
Step
0
0.000002
0.000004
0.000006
0.000008
0.00001
0.000012
0.000014
deepswe-preview-part2
deepswe-preview-part1
actor/pg_clipfrac
actor/pg_clipfrac
50
100
150
Step
0.0005
0.001
0.0015
0.002
deepswe-preview-part2
deepswe-preview-part1
actor/lr
actor/lr
50
100
150
Step
0
5e-7
0.000001
0.0000015
0.000002
deepswe-preview-part2
deepswe-preview-part1
actor/grad_norm
actor/grad_norm
50
100
150
Step
0.1
0.2
0.3
0.4
0.5
deepswe-preview-part2
deepswe-preview-part1
batch
2
critic
12
1-6 of 12
Add section