Skip to main content
baiqingl
Projects
GRPO comparison
Log in
Sign up
Overview
Workspace
Runs
Automat.
Sweeps
Reports
Artifacts
Baiqingl's workspace
Personal workspace
Automated workspace
Changes are only visible to you.
Runs
4
Name
4 visualized
Unsloth Full Training Run
Unsloth Full Training Run
Lora Extended Training Run
Lora Extended Training Run
Full Training Run
Full Training Run
Lora Training Run
Lora Training Run
1-4
of 4
Settings
Add panels
eval
12
1-6 of 12
val-core
1
training
2
timing_s
8
1-6 of 8
timing_per_token_ms
3
response_length
4
1-4 of 4
prompt_length
4
1-4 of 4
perf
7
1-6 of 7
global_seqlen
6
1-6 of 6
critic
12
1-6 of 12
critic/score/min
critic/score/min
0
20
40
60
80
Step
-8
-6
-4
-2
0
Full Training Run
critic/score/mean
critic/score/mean
0
20
40
60
80
Step
0
1
2
3
Full Training Run
critic/score/max
critic/score/max
0
20
40
60
80
Step
3.6
3.8
4
4.2
4.4
4.6
4.8
5
Full Training Run
actor
7
1-6 of 7
Add section