Skip to main content
picocreator
Projects
RWKV-InfCtx-Validation
Log in
Sign up
Overview
Workspace
Runs
Automat.
Sweeps
Reports
Artifacts
Picocreator's workspace
Personal workspace
Automated workspace
Changes are only visible to you.
Runs
70
Name
5 visualized
infctx-v5-deepspeed-test (deepspeed_stage_3_offload, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_3_offload, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_3, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_3, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_2_offload, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_2_offload, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_2, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_2, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_1, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_1, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_1, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_1, train-ctx=4096, data-ctx=4096)
infctx-validation-baseline (train-ctx=1024, data-ctx=1024)
infctx-validation-baseline (train-ctx=1024, data-ctx=1024)
(loss_bias=0.5) infctx-validation-baseline (train-ctx=1024, data-ctx=1024)
(loss_bias=0.5) infctx-validation-baseline (train-ctx=1024, data-ctx=1024)
infctx-validation ckpt-test (train-ctx=1024, data-ctx=1024)
infctx-validation ckpt-test (train-ctx=1024, data-ctx=1024)
infctx-bptt-validation L6-D512 (last-2-bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (last-2-bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (last-1-bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (last-1-bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (last-2-bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (last-2-bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (last-1-bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (last-1-bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (last-2-bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (last-2-bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (last-1-bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (last-1-bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (bptt, train-ctx=512, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (bptt, train-ctx=512, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (full, train-ctx=1024, data-ctx=1024, deepspeed_stage_1)
infctx-bptt-validation L6-D512 (full, train-ctx=1024, data-ctx=1024, deepspeed_stage_1)
bptt-test (bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_3_offload)
bptt-test (bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_3_offload)
bptt-test (bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_3_offload)
bptt-test (bptt, train-ctx=128, data-ctx=1024, deepspeed_stage_3_offload)
1-20
of 70
Add panels
train
3
train/loss
train/loss
0
50
100
150
Step
7
8
9
10
train/loss
train/loss
0
50
100
150
Step
7
8
9
10
train/loss
train/loss
0
50
100
150
Step
5
6
7
8
9
10
validation
1
validation/loss
validation/loss
infctx-v5-deepspeed-test (deepspeed_stage_3_offload, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_3, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_2_offload, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_2, train-ctx=4096, data-ctx=4096)
infctx-v5-deepspeed-test (deepspeed_stage_1, train-ctx=4096, data-ctx=4096)
0
2
4
6
8
Charts
5
1-5 of 5
trainer
2
Add section
8