CITADEL
StandbyNothing is live right now.
Eval accuracylast measured Mon 2 Nov, 19:05 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
Mon 2 Nov, 18:59 UTC
ran for
06:24
config
batch 8, seed 2 of 3

run11/22step150/150epoch5/5config5/11seed2/3batch8modellaya 421Mrev7b928d8items479+130

Loss train 0.0122 ▼ −1.2677 eval 0.7534 lowest -0.4812 at step 35 run before, run 10 ▲ +0.0122 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, 2 min/div2 min/div1 min/div
-101200:0002:0004:0006:00phase starting, 0.0 s to 36.8 sphase training, 36.8 s to 06:22phase eval, 06:22 to 06:24epoch 2, 02:37epoch 3, 03:43epoch 4, 04:49epoch 5, 05:55trainingE2E3E4E5−1.2677since first1.2799Lowest last train loss across this project's 22 runs, run 19lowest last -0.0937, run 190.0122-101200:0002:0004:0006:00phase starting, 0.0 s to 36.8 sphase training, 36.8 s to 06:22phase eval, 06:22 to 06:24epoch 1, 91.2 sepoch 2, 02:37epoch 3, 03:43epoch 4, 04:49epoch 5, 05:55trainingepoch 1epoch 2epoch 3epoch 4E51.2799Lowest last train loss across this project's 22 runs, run 19lowest last -0.0937, run 19eval 0.75340.0122▼ −1.2677-101200:0001:0002:0003:0004:0005:0006:00phase starting, 0.0 s to 36.8 sphase training, 36.8 s to 06:22phase eval, 06:22 to 06:24epoch 1, 91.2 sepoch 2, 02:37epoch 3, 03:43epoch 4, 04:49epoch 5, 05:55startingtrainingepoch 1epoch 2epoch 3epoch 4epoch 51.2799Lowest last train loss across this project's 22 runs, run 19lowest last -0.0937, run 19eval 0.75340.0122▼ −1.2677

Eval measures

Eval losslower is better
0.7534
02 →
ECEcalibration error, lower is better
0.1704
01
Brierlower is better
0.3582
02 →

Elapsed

66.2 s

epoch interval

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

2.20 s per step, 27.3 steps a minute

Against the other runs

18 of 22

by last eval accuracy

best so far 0.9231, run 19

▼ −0.0385 against run 10

Train minus eval

−0.7412

step 150

train 0.0122, eval 0.7534

Set-up

479 train, 130 eval

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

Seed spread, observed

0.0462
0.750.85

3 finished seeds

Samples per second

7.66

▲ +0.17 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 36.8 sphase training, 36.8 s to 06:22phase eval, 06:22 to 06:24E2E3E4E5traininglowest last -0.0937Train loss0.0122▼ −1.26772.00-1.001.27990.0122last ▼ −1.2677Samples per second7.66▲ +0.17877.48Learning rate1.00e-6▼ −2.40e-53.0e-50.0e+02.50e-5Power, whole chip, CPU and GPU28 W▼ −56 W100 W25 W84 W28 Wlast ▼ −56 WHottest, one board zone83 °C▼ −5 °C88 °C80 °C88 °Clast 83 °C00:0001:0002:0003:0004:0005:0006:00
GPUpower: whole chip, CPU and GPUtemperature: one board zone

GPU 0

Wwhole chip, CPU and GPU

28 W whole chip, CPU and GPU

use0%
memory, whole GPU13.3 GB

this run's peak 8.0 GB

83 °C one board zone

history 0 to 94 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

11 of 22 finished, 0.8000, config 5 of 11 (batch 8), seed 2 of 3, Mon 2 Nov, 18:59 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 2
Hardware
Radeon 8060S, ROCm

Log

Since startMark
0.0 sphase starting
7.5 swent public
36.8 sphase training
91.2 sepoch 1
157.4 sepoch 2
223.6 sepoch 3
289.4 sepoch 4
355.8 sepoch 5
382.9 sphase eval
382.9 seval 1

Seeds

SeedStatusEval accuracy
seed 0| finished0.7923
seed 1| finished0.8385
seed 2| finished0.8000

observed spread 0.0462, 3 finished seeds

Readings

31 rows as a table
StepSecondsTrain lossEval lossEval accuracyECEBrierSamples per secondLearning rateGPU GB, this run
036.7-----7.4832.50e-56.5
547.41.2799----7.4562.48e-56.5
1058.41.0189----7.2512.45e-56.5
1569.4-0.3356----7.2582.41e-56.5
2080.40.0386----7.2732.35e-56.5
2591.00.2359----7.5582.29e-56.5
30103.30.5987----6.5252.21e-56.5
35114.0-0.4812----7.5002.12e-56.5
40124.6-0.0174----7.5322.03e-56.5
45135.50.6412----7.3151.92e-56.5
50146.30.2523----7.4021.81e-56.5
55157.21.0032----7.3331.69e-56.5
60169.50.3204----6.5411.57e-56.5
65180.20.3978----7.4161.45e-56.5
70191.10.2565----7.3441.33e-56.5
75201.90.3873----7.4221.20e-56.5
80212.60.2661----7.5151.08e-56.5
85223.4-0.0201----7.3819.53e-66.5
90235.7-0.1018----6.5318.35e-66.5
95246.3-0.1545----7.5447.22e-66.5
100257.1-0.0880----7.4136.15e-66.5
105268.0-0.0285----7.2745.16e-66.5
110278.90.3219----7.3504.25e-66.5
115289.30.0000----7.7353.44e-66.5
120301.30.4691----6.6312.74e-66.5
125312.10.0571----7.4432.14e-66.5
130323.00.0826----7.3381.67e-66.5
135334.1-0.0099----7.2261.32e-66.5
140345.2-0.2549----7.1831.09e-66.5
145355.6-0.0032----7.6551.00e-66.5
150384.20.01220.75340.80000.17040.3582---

GPU readings

31 rows as a table
StepGPUUse %Power W, whole chip, CPU and GPUMemory GB, whole GPUTemp C, one board zone
004583.813.287.8
1501983.813.381.8
3003185.313.386.8
4504683.713.382.8
600484.513.386.8
7508484.413.382.8
9006985.113.385.8
10503083.313.381.8
12004694.013.385.8
13502982.213.382.8
1500027.513.382.8

Worked out

ReadoutValueWorked out from
Run complete100%150 steps over 150
Step time2.20 s319.0 s over 145 steps, between readings. Startup and intervals that hold an eval are left out
Steps per minute27.360 s over the step time of 2.20 s
Epoch interval66.2 sMean of 4 gaps between epoch marks
Train loss since first reading−1.26770.0122 at step 150 minus 1.2799 at step 5
Lowest train loss-0.4812Lowest of 30 readings, at step 35
Eval accuracy, as shown80.0%0.8000 times 100, to one decimal. Changes on that plate are percentage points
Train minus eval loss−0.74120.0122 minus 0.7534, both at step 150
Wrong answers26130 eval items minus 104 correct. 104 over 130 is the only count that gives 0.8000
Rank18 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−0.03850.8000 minus 0.8385 on run 10
Lowest last train loss-0.0937Lowest last train loss across this project's 22 runs, run 19
Seed spread, observed0.0462High 0.8385 minus low 0.7923 across 3 finished seeds
Train loss against run 10+0.01220.0122 minus 0.0000 on run 10, both at step 150
Samples per second since first reading+0.177.66 minus 7.48
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