CITADEL
StandbyNothing is live right now.
Eval accuracylast measured Tue 3 Nov, 01:26 UTC
0%25%50%75%100%0%25%50%75%100%0%25%50%75%100%

Claviger

finishedRadeon 8060S

A small model that reads one Discord message and decides whether an agent should reply, and which one. Fine-tuned from Laya on invented chat.

started
Tue 3 Nov, 01:13 UTC
ran for
12:50
config
batch 8, seed 1 of 3

run13/22step150/150epoch5/5config6/11seed1/3batch8modellaya 421Mrev7b928d8items479+130

Loss train 0.0000 ▼ −0.4307 eval 0.6453 lowest -0.7846 at step 80 run before, run 12 ▲ +0.0000 step 150 Laya's objective: soft cross-entropy plus a policy-gradient term, so it can go below zero.
Timebase
Range
Smoothing
Reference
30 readings, 5 min/div5 min/div2 min/div
-202400:0005:0010:00phase starting, 0.0 s to 81.8 sphase training, 81.8 s to 12:44phase eval, 12:44 to 12:50epoch 2, 05:11epoch 3, 07:21epoch 4, 09:28epoch 5, 11:34trainingE2E3E4E50.4307Lowest last train loss across this project's 22 runs, run 19lowest last -0.0937, run 190.0000-202400:0005:0010:00phase starting, 0.0 s to 81.8 sphase training, 81.8 s to 12:44phase eval, 12:44 to 12:50epoch 1, 02:59epoch 2, 05:11epoch 3, 07:21epoch 4, 09:28epoch 5, 11:34trainingepoch 1epoch 2epoch 3epoch 4E50.4307Lowest last train loss across this project's 22 runs, run 19lowest last -0.0937, run 19eval 0.64530.0000▼ −0.4307-202400:0002:0004:0006:0008:0010:0012:00phase starting, 0.0 s to 81.8 sphase training, 81.8 s to 12:44phase eval, 12:44 to 12:50epoch 1, 02:59epoch 2, 05:11epoch 3, 07:21epoch 4, 09:28epoch 5, 11:34startingtrainingepoch 1epoch 2epoch 3epoch 4epoch 50.4307Lowest last train loss across this project's 22 runs, run 19lowest last -0.0937, run 19eval 0.64530.0000▼ −0.4307

Eval measures

Eval losslower is better
0.6453
02 →
ECEcalibration error, lower is better
0.1250
01
Brierlower is better
0.3020
02 →

Elapsed

128.8 s

epoch interval

150 of 150 steps, 5 of 5 epochs, 100%

4.23 s per step, 14.2 steps a minute

Against the other runs

14 of 22

by last eval accuracy

best so far 0.9231, run 19

▲ +0.0000 against run 12

Train minus eval

−0.6453

step 150

train 0.0000, eval 0.6453

Set-up

479 train, 130 eval

synthetic, batch size 8, grad accum 2, epochs 5, loss rlcd, seed 1

Seed spread, observed

0.0231
0.800.85

3 finished seeds

Samples per second

4.22

▲ +0.11 since first

Learning rate

1.00e-6

▼ −2.40e-5 since first

GPU memory, this run

6.5 GB

On one axis

phase starting, 0.0 s to 81.8 sphase training, 81.8 s to 12:44phase eval, 12:44 to 12:50E2E3E4E5traininglowest last -0.0937Train loss0.0000▼ −0.43074.00-2.000.43070.0000last ▼ −0.4307Samples per second4.22▲ +0.11524.11Learning rate1.00e-6▼ −2.40e-53.0e-50.0e+02.50e-5Power, whole chip, CPU and GPU84 W0 W85 W82 W84 W84 Wlast 0 WHottest, one board zone94 °C▼ −2 °C96 °C92 °C96 °Clast 94 °C00:0002:0004:0006:0008:0010:0012:00
GPUpower: whole chip, CPU and GPUtemperature: one board zone

GPU 0

Wwhole chip, CPU and GPU

84 W whole chip, CPU and GPU

use0%
memory, whole GPU13.3 GB

this run's peak 8.0 GB

94 °C one board zone

history 0 to 84 W

Input

  • Runs128 runs
  • Agentsnone showing
  • Treesnone showing
  • Projectsnone showing

Check

Every entry is checked on its way out; some are held for review.

Not shown on any run: data samples, predictions, raw logs, file paths, machine names

Delay

2 min, set

Watching the agents work, about 2 min behind.

One way

Nothing here reaches the agents.

This page

This page updates by itself.

visits counted with Google Analytics · choose

Runsheight is eval accuracyall recordings ▸ | finished× failed■ stopped○ lost● live
0.950.750.950.75finished, eval accuracy 0.8077, config 1 of 11 (batch 8), seed 2, Mon 2 Nov, 13:19 UTCfinished, eval accuracy 0.8231, config 2 of 11 (batch 8), seed 2 of 3, Mon 2 Nov, 14:03 UTCfinished, eval accuracy 0.8231, config 3 of 11 (batch 8), seed 2, Mon 2 Nov, 14:09 UTCfinished, eval accuracy 0.7846, config 2 of 11 (batch 8), seed 0 of 3, Mon 2 Nov, 14:16 UTCfinished, eval accuracy 0.8385, config 2 of 11 (batch 8), seed 1 of 3, Mon 2 Nov, 14:22 UTCfinished, eval accuracy 0.7923, config 4 of 11 (batch 8), seed 0 of 3, Mon 2 Nov, 18:15 UTCfinished, eval accuracy 0.8000, config 4 of 11 (batch 8), seed 1 of 3, Mon 2 Nov, 18:22 UTCfinished, eval accuracy 0.8154, config 4 of 11 (batch 8), seed 2 of 3, Mon 2 Nov, 18:28 UTCfinished, eval accuracy 0.7923, config 5 of 11 (batch 8), seed 0 of 3, Mon 2 Nov, 18:46 UTCfinished, eval accuracy 0.8385, config 5 of 11 (batch 8), seed 1 of 3, Mon 2 Nov, 18:52 UTCfinished, eval accuracy 0.8000, config 5 of 11 (batch 8), seed 2 of 3, Mon 2 Nov, 18:59 UTCfinished, eval accuracy 0.8231, config 6 of 11 (batch 8), seed 0 of 3, Tue 3 Nov, 00:59 UTCfinished, eval accuracy 0.8231, config 6 of 11 (batch 8), seed 1 of 3, Tue 3 Nov, 01:13 UTCfinished, eval accuracy 0.8000, config 6 of 11 (batch 8), seed 2 of 3, Tue 3 Nov, 01:26 UTCfinished at step 336 of 339, eval accuracy 0.8769, config 7 of 11 (batch 8), seed 0 of 3, Thu 19 Nov, 21:49 UTCfinished at step 336 of 339, eval accuracy 0.8846, config 7 of 11 (batch 8), seed 1 of 3, Thu 19 Nov, 22:06 UTCfinished at step 336 of 339, eval accuracy 0.8846, config 7 of 11 (batch 8), seed 2 of 3, Thu 19 Nov, 22:21 UTCfinished at step 336 of 339, eval accuracy 0.8615, config 8 of 11 (learning rate 0.0001, batch 8), seed 0, Sat 21 Nov, 23:04 UTCfinished at step 336 of 339, eval accuracy 0.9231, config 9 of 11 (learning rate 0.00002, batch 8), seed 0, Sat 21 Nov, 23:17 UTCfinished at step 336 of 339, eval accuracy 0.8538, config 10 of 11 (learning rate 0.00005, batch 8), seed 0, Sat 21 Nov, 23:29 UTCfinished at step 336 of 339, eval accuracy 0.9231, config 11 of 11 (learning rate 0.00002, batch 8), seed 1 of 2, Sun 22 Nov, 03:49 UTCfinished at step 336 of 339, eval accuracy 0.8846, config 11 of 11 (learning rate 0.00002, batch 8), seed 2 of 2, Sun 22 Nov, 04:01 UTC0.950.75finished, eval accuracy 0.8077, config 1 of 11 (batch 8), seed 2, Mon 2 Nov, 13:19 UTCfinished, eval accuracy 0.8231, config 2 of 11 (batch 8), seed 2 of 3, Mon 2 Nov, 14:03 UTCfinished, eval accuracy 0.8231, config 3 of 11 (batch 8), seed 2, Mon 2 Nov, 14:09 UTCfinished, eval accuracy 0.7846, config 2 of 11 (batch 8), seed 0 of 3, Mon 2 Nov, 14:16 UTCfinished, eval accuracy 0.8385, config 2 of 11 (batch 8), seed 1 of 3, Mon 2 Nov, 14:22 UTCfinished, eval accuracy 0.7923, config 4 of 11 (batch 8), seed 0 of 3, Mon 2 Nov, 18:15 UTCfinished, eval accuracy 0.8000, config 4 of 11 (batch 8), seed 1 of 3, Mon 2 Nov, 18:22 UTCfinished, eval accuracy 0.8154, config 4 of 11 (batch 8), seed 2 of 3, Mon 2 Nov, 18:28 UTCfinished, eval accuracy 0.7923, config 5 of 11 (batch 8), seed 0 of 3, Mon 2 Nov, 18:46 UTCfinished, eval accuracy 0.8385, config 5 of 11 (batch 8), seed 1 of 3, Mon 2 Nov, 18:52 UTCfinished, eval accuracy 0.8000, config 5 of 11 (batch 8), seed 2 of 3, Mon 2 Nov, 18:59 UTCfinished, eval accuracy 0.8231, config 6 of 11 (batch 8), seed 0 of 3, Tue 3 Nov, 00:59 UTCfinished, eval accuracy 0.8231, config 6 of 11 (batch 8), seed 1 of 3, Tue 3 Nov, 01:13 UTCfinished, eval accuracy 0.8000, config 6 of 11 (batch 8), seed 2 of 3, Tue 3 Nov, 01:26 UTCfinished at step 336 of 339, eval accuracy 0.8769, config 7 of 11 (batch 8), seed 0 of 3, Thu 19 Nov, 21:49 UTCfinished at step 336 of 339, eval accuracy 0.8846, config 7 of 11 (batch 8), seed 1 of 3, Thu 19 Nov, 22:06 UTCfinished at step 336 of 339, eval accuracy 0.8846, config 7 of 11 (batch 8), seed 2 of 3, Thu 19 Nov, 22:21 UTCfinished at step 336 of 339, eval accuracy 0.8615, config 8 of 11 (learning rate 0.0001, batch 8), seed 0, Sat 21 Nov, 23:04 UTCfinished at step 336 of 339, eval accuracy 0.9231, config 9 of 11 (learning rate 0.00002, batch 8), seed 0, Sat 21 Nov, 23:17 UTCfinished at step 336 of 339, eval accuracy 0.8538, config 10 of 11 (learning rate 0.00005, batch 8), seed 0, Sat 21 Nov, 23:29 UTCfinished at step 336 of 339, eval accuracy 0.9231, config 11 of 11 (learning rate 0.00002, batch 8), seed 1 of 2, Sun 22 Nov, 03:49 UTCfinished at step 336 of 339, eval accuracy 0.8846, config 11 of 11 (learning rate 0.00002, batch 8), seed 2 of 2, Sun 22 Nov, 04:01 UTC

13 of 22 finished, 0.8231, config 6 of 11 (batch 8), seed 1 of 3, Tue 3 Nov, 01:13 UTC

Runs as a table
StartedStatusEval accuracyStepsConfig
Sun 22 Nov, 04:01 UTC| finished0.8846336 of 339learning rate 0.00002, batch 8, seed 2 of 2
Sun 22 Nov, 03:49 UTC| finished0.9231336 of 339learning rate 0.00002, batch 8, seed 1 of 2
Sat 21 Nov, 23:29 UTC| finished0.8538336 of 339learning rate 0.00005, batch 8, seed 0
Sat 21 Nov, 23:17 UTC| finished0.9231336 of 339learning rate 0.00002, batch 8, seed 0
Sat 21 Nov, 23:04 UTC| finished0.8615336 of 339learning rate 0.0001, batch 8, seed 0
Thu 19 Nov, 22:21 UTC| finished0.8846336 of 339batch 8, seed 2 of 3
Thu 19 Nov, 22:06 UTC| finished0.8846336 of 339batch 8, seed 1 of 3
Thu 19 Nov, 21:49 UTC| finished0.8769336 of 339batch 8, seed 0 of 3
Tue 3 Nov, 01:26 UTC| finished0.8000150 of 150batch 8, seed 2 of 3
Tue 3 Nov, 01:13 UTC| finished0.8231150 of 150batch 8, seed 1 of 3
Tue 3 Nov, 00:59 UTC| finished0.8231150 of 150batch 8, seed 0 of 3
Mon 2 Nov, 18:59 UTC| finished0.8000150 of 150batch 8, seed 2 of 3
Mon 2 Nov, 18:52 UTC| finished0.8385150 of 150batch 8, seed 1 of 3
Mon 2 Nov, 18:46 UTC| finished0.7923150 of 150batch 8, seed 0 of 3
Mon 2 Nov, 18:28 UTC| finished0.8154150 of 150batch 8, seed 2 of 3
Mon 2 Nov, 18:22 UTC| finished0.8000150 of 150batch 8, seed 1 of 3
Mon 2 Nov, 18:15 UTC| finished0.7923150 of 150batch 8, seed 0 of 3
Mon 2 Nov, 14:22 UTC| finished0.8385150 of 150batch 8, seed 1 of 3
Mon 2 Nov, 14:16 UTC| finished0.7846150 of 150batch 8, seed 0 of 3
Mon 2 Nov, 14:09 UTC| finished0.8231150 of 150batch 8, seed 2
Mon 2 Nov, 14:03 UTC| finished0.8231150 of 150batch 8, seed 2 of 3
Mon 2 Nov, 13:19 UTC| finished0.8077150 of 150batch 8, seed 2

Plate

Model
convaiinnovations/laya, 421M parameters revision 7b928d8
Data
Synthetic Discord triage set, synthetic479 train130 eval
Config
batch size 8, grad accum 2, epochs 5, loss rlcd, seed 1
Hardware
Radeon 8060S, ROCm

Log

Since startMark
0.0 sphase starting
1.8 swent public
81.8 sphase training
179.1 sepoch 1
311.4 sepoch 2
441.2 sepoch 3
568.2 sepoch 4
694.4 sepoch 5
764.4 sphase eval
764.4 seval 1

Seeds

SeedStatusEval accuracy
seed 0| finished0.8231
seed 1| finished0.8231
seed 2| finished0.8000

observed spread 0.0231, 3 finished seeds

Readings

31 rows as a table
StepSecondsTrain lossEval lossEval accuracyECEBrierSamples per secondLearning rateGPU GB, this run
081.6-----4.1072.50e-56.5
5100.60.4307----4.2202.48e-56.5
10120.01.0819----4.1132.45e-56.5
15138.90.3514----4.2502.41e-56.5
20158.70.2497----4.0382.35e-56.5
25179.00.9126----3.9462.29e-56.5
30211.00.8281----2.4972.21e-56.5
35231.40.2653----3.9212.12e-56.5
40252.10.4311----3.8672.03e-56.5
45272.23.5385----3.9731.92e-56.5
50292.00.3317----4.0511.81e-56.5
55311.20.1993----4.1611.69e-56.5
60340.00.5015----2.7761.57e-56.5
65359.40.1702----4.1191.45e-56.5
70379.8-0.0475----3.9271.33e-56.5
75400.1-0.6396----3.9391.20e-56.5
80420.4-0.7846----3.9341.08e-56.5
85441.00.8554----3.8859.53e-66.5
90469.52.7800----2.8108.35e-66.5
95489.41.5983----4.0337.22e-66.5
100508.6-0.0718----4.1646.15e-66.5
105528.40.1299----4.0405.16e-66.5
110548.2-0.1139----4.0384.25e-66.5
115568.0-0.0053----4.0293.44e-66.5
120595.50.0000----2.9152.74e-66.5
125615.00.0000----4.1012.14e-66.5
130634.90.0000----4.0291.67e-66.5
135654.90.0560----3.9821.32e-66.5
140675.3-0.4934----3.9271.09e-66.5
145694.30.0000----4.2201.00e-66.5
150771.00.00000.64530.82310.12500.3020---

GPU readings

31 rows as a table
StepGPUUse %Power W, whole chip, CPU and GPUMemory GB, whole GPUTemp C, one board zone
002383.613.095.8
50083.413.292.8
150083.413.393.8
2502182.913.393.8
3002283.813.395.8
3502683.613.393.8
4504683.413.393.8
550082.613.393.8
600883.413.395.8
650384.413.393.8
750183.713.394.8
8502683.713.394.8
9005983.513.395.8
9504883.613.394.8
10504483.413.394.8
11502683.313.393.8
12004783.513.395.8
1300083.713.394.8
1350083.313.394.8
1450483.413.393.8
1500083.613.393.8

Worked out

ReadoutValueWorked out from
Run complete100%150 steps over 150
Step time4.23 s612.6 s over 145 steps, between readings. Startup and intervals that hold an eval are left out
Steps per minute14.260 s over the step time of 4.23 s
Epoch interval128.8 sMean of 4 gaps between epoch marks
Train loss since first reading−0.43070.0000 at step 150 minus 0.4307 at step 5
Lowest train loss-0.7846Lowest of 30 readings, at step 80
Eval accuracy, as shown82.3%0.8231 times 100, to one decimal. Changes on that plate are percentage points
Train minus eval loss−0.64530.0000 minus 0.6453, both at step 150
Wrong answers23130 eval items minus 107 correct. 107 over 130 is the only count that gives 0.8231
Rank14 of 22By last eval accuracy, across this project's 22 runs with an eval
Best so far0.9231Highest last eval accuracy, run 19
Against the run before+0.00000.8231 minus 0.8231 on run 12
Lowest last train loss-0.0937Lowest last train loss across this project's 22 runs, run 19
Seed spread, observed0.0231High 0.8231 minus low 0.8000 across 3 finished seeds
Train loss against run 12+0.00000.0000 minus 0.0000 on run 12, both at step 150
Samples per second since first reading+0.114.22 minus 4.11
Learning rate since first reading−2.40e-51.00e-6 minus 2.50e-5
This run's peak memory per GPU8.0 GBAs the run reports it: its own memory, not the whole GPU's