Skip to main content
zaydzuhri
Projects
fla
Log in
Sign up
Overview
Workspace
Runs
Automat.
Sweeps
Reports
Artifacts
Zaydzuhri's workspace
Personal workspace
Automated workspace
Changes are only visible to you.
Runs
322
Name
9 visualized
mtp_transformer-mtp.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
mtp_transformer-mtp.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
mtp_transformer-mtp.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
mtp_transformer-mtp.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
mtp_transformer-mtp.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
mtp_transformer-mtp.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
mtp_transformer-mtp.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
mtp_transformer-mtp.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
mtp_transformer-mtp.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
mtp_transformer-mtp.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
top_transformer-top.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
top_transformer-top.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
transformer-vanilla.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
transformer-vanilla.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
top_transformer-top.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
top_transformer-top.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
transformer-vanilla.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
transformer-vanilla.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
top_transformer-top.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
top_transformer-top.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
transformer-vanilla.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
transformer-vanilla.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
transformer-vanilla.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
transformer-vanilla.7B.batch8.seqlen4096.context4096.warmup2000.update2.steps200000.lr1.2e-4.cosine
transformer-vanilla.7B.batch4.seqlen4096.context4096.warmup2000.update4.steps200000.lr1.2e-4.cosine
transformer-vanilla.7B.batch4.seqlen4096.context4096.warmup2000.update4.steps200000.lr1.2e-4.cosine
1-13
of 13
Add panels
loss_metrics
6
1-3 of 6
loss_metrics/global_avg_loss, loss_metrics/global_avg_ntp_loss, loss_metrics/global_avg_myopic_loss, loss_metrics/global_avg_top_loss
loss_metrics/global_avg_loss, loss_metrics/global_avg_ntp_loss, loss_metrics/global_avg_myopic_loss, loss_metrics/global_avg_top_loss
50k
100k
150k
200k
Step
2
3
4
5
6
7
8
9
10
loss_metrics/global_avg_top_loss
loss_metrics/global_avg_top_loss
50k
100k
150k
200k
Step
4
6
8
10
loss_metrics/global_avg_mtp_loss
loss_metrics/global_avg_mtp_loss
50k
100k
150k
200k
Step
15
20
25
30
optimizer
3
optimizer/grad_norm
optimizer/grad_norm
50k
100k
150k
200k
Step
1
100
optimizer/lr
optimizer/lr
50k
100k
150k
200k
Step
0.00002
0.00004
0.00006
0.00008
0.0001
optimizer/skipped_step
optimizer/skipped_step
50k
100k
150k
200k
Step
-2
-1
0
1
2
time_metrics
3
Charts
3
memory
6
1-6 of 6
optim
2
Add section